The Structure of Evolutionary Theory

by Stephen Jay Gould

  I suggest that we use the list of minimal commitments to gauge the status of Darwin's errors. Very few faults of simple fact can, as individual items, be of much consequence unless they confute a core commitment. Darwin argued, for example, that swimbladders evolved into lungs (see p. 107) though ex­actly the opposite occurred — but no premise of the general theory suffers any injury by this mistake, however embarrassing. What about more important theoretical claims like Darwin's hypothesis of “pangenesis” as a mechanism of heredity (Darwin, 1868)? Again, Darwin's view of life would have been easier to vindicate if the theory had been affirmed, but none of his three essen­tial postulates about the nature of variation fell with the disproof of pangen­esis, and the core commitments remained intact, if unproven. What about the impact of major claims that turn out to be basically true, Mendelism for ex­ample? We must make our judgment by assessing their engagement with the core commitments. In the first decade of the 20th century, most evolution­ists invoked Mendelism as a saltational theory of macromutation against the Darwinian core commitment to small-scale variation (see Chapter 5). Later, largely through R. A. Fisher's analysis and the resolution of the Mendelian vs. biometrician debate, macromutations were rejected, “ordinary” small-scale variation granted a Mendelian basis, and Mendelism comfortably reinter­preted as support for the same core commitment. Again, challenges and new proposals must be judged and ranked by their engagement with the essence of a reigning theory. Darwinism embodies a definable set of minimal commit­ments; all great theories do and must. [Page 167]

  We should use this perspective of engagement with the core commitments to assess the relative theoretical importance of issues now commanding at­tention among evolutionists. For example, Kimura's theory of neutralism (1983) ranks as fundamental and reformative for proposing a new domain of causation at high relative frequency. I regard as unfair, and disrespectful of Darwin's clear commitments, the common rhetorical strategy of arguing, as Stebbins and Ayala did for example (1981), that selection and neutralism should be judged as competing paradigms comfortably embraced within the Modern Synthesis. The Synthesis, as an intellectual structure, has always been understood as Darwinism strengthened by modern knowledge about ge­netics and heredity. The Synthesis must therefore assert a dominant relative frequency for selection. Of course such a theory allows for neutrality — one could scarcely deny either the mathematics or the conditions of potential op­eration — but only at a low relative frequency, so that the preeminence of se­lection will remain unchallenged.

  Kimura's claim for high, even dominant, relative frequency of neutral change at the nucleotide level introduces a world different from Darwin's. At most, one can say that this world, largely invisible at the organismic level, does not subvert Darwin's proposal that selection dominates the phenotypic realm of overt form, function, and physiology of organisms. But in so saying, we must admit that a large part of reality, though unaddressed by Darwin himself, cannot be explained on Darwinian principles if Kimura's theory holds. Darwinism does not fall thereby, but a new and distinct domain, pri­marily regulated by a different style of causality, has been added to evolution­ary explanation. How can one deny that evolutionary theory becomes sub­stantially reformulated and enriched thereby? Why would one want to issue such a denial, unless psychic health depends upon the continued assertion of comfortable orthodoxy, whatever the required twist of logic?

  My own expertise lies in paleontology, and this book shall emphasize critiques from the attendant domain of macroevolution, descriptively defined as patterns and causes of evolution at and above the species level. (I acknowl­edge, of course, the fascination and transforming power of work at the mo­lecular level. I also recognize that macroevolution must shake hands with mo­lecular genetics in order to forge the new consensus. If this book slights the molecular side, my own ignorance stands as the only cause, and this work necessarily suffers thereby.)

  Basically, I shall defend the view that each leg of Darwin's essential tripod, as explicated in this chapter, now faces a serious critique from the domain of macroevolutionary change. These critiques rank as auxiliaries to Darwin­ism in Kellogg's sense; for they either expand or add to the core commit­ments. But the expansions are large and the additions substantial — so the resulting revision can no longer be called ordinary Darwinism in any conven­tional meaning. I am convinced that the three critiques intertwine in a poten­tially unified way. But consensus is premature and we can only see the result­ing shape of the revised and unified theory through a glass darkly — though in the future, no doubt, face to face. [Page 168]

  Proceeding in reverse order through critiques of Darwinism's three core claims, catastrophic mass extinction, and more general views about fortu­ity in abiotically driven extinction at all levels, challenge Darwin's essential notion of a dominant relative frequency for biotic struggle in a crowded world — the third leg of the tripod, as represented by the geological stage re­quired for an evolutionary play based entirely on extrapolation of microevolutionary principles (Chapters 6 and 12). The general idea of constraint — more in the positive sense of internally biased channels for change, rather than the negative meaning of limited variation for potentially useful alter­ations (see Gould, 1989a) — rejects the key Darwinian notion of isotropy in raw material, and consequent control of evolutionary direction by natural se­lection. Constraint therefore challenges the second leg of the tripod — the “creativity of natural selection” — not by confuting the proposition that natu­ral selection acts as a creative force, but by insisting on diminished relative frequency and a sharing of control. Moreover, by reasserting the structuralist side of the old dichotomy between structure and function in biology — an is­sue far predating evolution, and inherent in the struggle between continental vs. Paleyan approaches to natural theology — the idea of constraint reengages one of the deepest issues in all the life sciences (Chapters 4–5 and 10–11).

  Most importantly, and as the best integrator of all three critiques, the hierarchical theory of natural selection, by asserting both the existence and rela­tive importance of selection at all levels from genes to species, challenges the first leg of the tripod — the insistence, so crucial to Darwin's radical overthrow of Paley via Adam Smith, that selection works almost exclusively on organ­isms (Chapters 3 and 8–9). I believe that this hierarchical theory provides the most fundamental, and potentially unifying, of all critiques — for I suspect that many constraints will be explained as effects of lower level selection indi­rectly expressed in phenotypes; while the contribution of mass extinction to repatterning life's history will include a crucial component of selection at lev­els above the organismic. Moreover, the attendant need to reconceptualize trends and stabilities not as optimalities of selection upon organisms alone, but as outcomes of interactions among numerous levels of selection, implies an evolutionary world sufficiently at variance from Darwin's own conception that the resulting theory, although still “selectionist” at its core, must be rec­ognized as substantially different from current orthodoxy — and not just as a dash of spice on an underflavored dish. I therefore devote the largest section of this book's second half (Chapters 8 and 9) to defining and defending this hierarchical theory of selection.

  If the next generation of evolutionists follows and extends this protocol at the outset of our new millennium, as presaged by the tentative work and ex­ploration of so many scientists at the close of the last millennium, then we shall honor, all the more, the vitality of the tight definitions and firm commit­ments proposed by Darwin himself at the foundation of our discipline. Few theories hold the range of power, and the intricacy of logic, necessary to gen­erate an intellectual structure of such continuing fascination and relevance. We do not pay our proper respect to Darwin by bowing before the icons of [Page 169] his central propositions, but by engaging these focal precepts as living pres­ences, ripe for reformulation, almost 150 years after their initial presentation. In Darwin's own world of continuou
s flux, anything that lasts so long be­comes a many-splendored thing. In a revised world of structuralism, we might say that Darwin first located and embellished one of the few brilliant and coherent positions in an intellectual universe with few nucleating places. Either formulation engenders the same result of abiding respect for Darwin's view of life — leading to proper thanks owed by all of us for the good fortune of such an interesting founder. What greater pleasure can we know than to engage Darwin in dialogue — as we can and must do, because his theory rests upon a powerful and defining essence. Darwin, in short, is the extraordinary man who, all by himself, embodied the only three beings proclaimed worthy of respect by Baudelaire — for he pulled down an old order, and came to know a large part of the new world that he created. Il n'existe que trois etres re­spectables: le pretre, le guerrier, le poete. Savoir, tuer, et creer. There exist only three beings worthy of our respect: the priest, the warrior, and the poet. Know, kill, and create.

  [Page 170]

  CHAPTER THREE

  Seeds of Hierarchy

  Lamarck and the Birth of Modern Evolutionism in

  Two-Factor Theories

  THE MYTHS OF LAMARCK

  In 1793, the French revolutionary government, having expunged the past by executing a monarch, proclaimed a new beginning of time. They renamed the months, and started the calendar all over again with the foundation of the Republic in September 1792. The old months had honored emperors and gods, but the new months would celebrate the passing of seasons by weather and activity — Brumaire (the foggy month in fall), Thermidor (the hot times of mid summer), and Nivose (for the depth of a frosty winter), for example.

  Jean-Baptiste-Pierre-Antoine de Monet, Chevalier de Lamarck (1744-1829) — now redesignated, with democratic brevity, as Citoyen Lamarck — became professor of “lower” animals (the old Linnaean classes Insecta and Vermes, later renamed “invertebrates” by Lamarck himself) at the newly founded Museum d'Histoire Naturelle in 1793. (His previous work, nearly all in botany, had not prepared him for this new role, though he had long been an avid shell collector and student of conchology.) Until 1797, he had supported the conventional idea of species as fixed entities. But he then be­came an evolutionist, first expressing this new view of life in his inaugural lec­ture for the Museum course of 1800, and then in three major works — the Recherches sur I'organisation des corps vivans of 1802; his most famous work, the Philosophie zoologique of 1809; and the Histoire naturelle des animaux sans vertebres of 1815-1822.

  In an ironic symbol, Lamarck first presented his evolutionary theories in an inaugural lecture pronounced on the 21st day of Floreal, year VIII (May 11, 1800) — in the month of flowering. For Lamarck's theory suffered the oppo­site fate of withering, and the scorn of inattention. We all know the image of Lamarck — an impression carefully nurtured, for different reasons, by friends and foes alike — as a lonely man (a prophet before his time to some, a kook to others), penniless, friendless, and, finally, blind; living out the last days of a long and sad life, supported only by his devoted daughters.

  This image of a forgotten failure was fostered by the two greatest figures of 19th century natural history — first by Cuvier, and later by Darwin. Darwin said little about Lamarck (see pp. 192–197), but his derision still permeates [Page 171] our view. Cuvier did far more damage. I don't know what lip service Cuvier gave to the ancient maxim de mortuis nil nisi bonum (say only good of the dead), but he violated this precept with avidity in writing eloges (eulogies) of deceased colleagues. Cuvier, the consummate politician, understood the power thus granted to shape history in his own favor. For what forum could be less subject to rebuttal, and therefore more suited for easy passage into re­ceived truth. As master of eloges, Cuvier held enormous power over his col­leagues, as long as he could outlive them! (see pp. 309–312 on Geoffroy's re­venge for the same reason). His official eloge of Lamarck is a masterful, if repugnant, document of propaganda directed against a close colleague and former friend who had (in Cuvier's view) gone beyond the pale in both meth­odology of research and content of belief. Cuvier used his eloge as an oppor­tunity to castigate Lamarck, and thus provide a lesson in proper procedure for aspiring scientists.

  Cuvier began with cloying praise, and then portrayed his need to criticize as a sad duty: “In sketching the life of one of our most celebrated naturalists, we have conceived it to be our duty, while bestowing the commendation they deserve on the great and useful works which science owes to him, likewise to give prominence to such of his productions in which too great indulgence of a lively imagination had led to results of a more questionable kind, and to indi­cate, as far as we can, the cause or, if it may be so expressed, the genealogy of his deviations” (1832, 1984 edition, p. 435).

  Cuvier then dismembered Lamarck on two grounds. First, with justice in the claim (however unkind the rhetoric), he castigated Lamarck for reaching too far without foundation, and for building all-encompassing systems in the speculative mode. (This criticism reflected Cuvier's main unhappiness with Lamarck's science. Cuvier viewed himself as a modernist, committed to rigor­ous empirical documentation, and no extension beyond direct evidence in the search for explanations — as opposed to Lamarck's unfruitful, comprehensive speculation in the antiquated esprit de systeme, or spirit of system): “He had meditated on the general laws of physics and chemistry, on the phenomena of the atmosphere, on those of living bodies, and on the origin of the globe and its revolutions. Psychology, and the higher branches of metaphysics, were not beyond the range of his contemplations; and on all these subjects he had formed a number of definite ideas . . . calculated to place every branch of knowledge on a new foundation” (1832, 1984 edition, p. 442).

  Cuvier acknowledged Lamarck's excellent efforts in morphology and taxonomy, but then damned him for denigrating this admirable work as a trifle compared with all-embracing and useless theories. What a sorry spectacle: Lamarck in his armchair, challenging the great Lavoisier, the icon and martyr of true science. (Lavoisier was beheaded during the Reign of Terror.) “So inti­mately did he identify himself with his systems, and such was his desire that they should be propagated, that all other objects seemed to him subordinate, and even his greatest and most useful works appeared in his own eyes merely as the slight accessories of lofty speculations. Thus, while Lavoisier was creat­ing in his laboratory a new chemistry, founded on a beautiful and methodical [Page 172] series of experiments, M. de Lamarck, without attempting experiment, and destitute of the means of doing so, imagined that he had discovered another” (1832, 1984 edition, p. 442).

  After ridiculing Lamarck's general method of system building, Cuvier mounted his second attack and dismembered the particular content of Lamarck's system, especially his evolutionary views. Cuvier did his former colleague a lasting disservice by caricaturing Lamarckian evolution, as the outcome of organic will, based on desires, translated into phyletic progress. Cuvier's rhetoric was brilliant, his characterization grossly distorted:

  Wants and desires, produced by circumstances, will lead to other efforts, which will produce other organs. ... It is the desire and the attempt to swim that produces membranes in the feet of aquatic birds; wading in the water, and at the same time the desire to avoid wet, has lengthened the legs of such as frequent the sides of rivers . . . These principles once admitted, it will easily be perceived that nothing is wanting but time and circumstances to enable a monad or a polypus gradually and indiffer­ently to transform themselves into a frog, a stork, or an elephant (1832, 1984 edition, p. 446).

  Finally, in an ultimate dismissal from a “hard” scientist (and with the tone of the Yahoo), Cuvier concludes: “A system established on such foundations may amuse the imagination of a poet; a metaphysician may derive from it an entirely new series of systems; but it cannot for a moment bear the examina­tion of anyone who has dissected a hand ... or even a feather” (1832, 1984 edition, p. 447).

  Cuvier's caricature remains potent in our worst modern misunderstanding of La
marck as a mystical vitalist, advancing the idea of an ineffable organic will against the ordinary physical causality of science. (Tit for tat, and how­ever unfairly, Lyell hurt Cuvier even more in return by caricaturing him as a theologically tainted, antiscientific catastrophist in geology.) But Lamarck, schooled (along with Cuvier) in the ideals of the French Enlightenment, was an ardent materialist. His idiosyncratic and unfruitful views about the nature of matter (arising primarily from his anti-Lavoisierian chemistry) led to pre­dictions of odd behavior for living bodies, but his basic notions of reduction and causality remained in the scientific mainstream. In his last great work, and in the context of his evolutionary theory, Lamarck defended a conven­tional view of mechanistic causality, and derided all teleological interpreta­tions. Goals, he argued, are false appearances reflecting an underlying causal necessity:

  It is chiefly among the living, and most notably among Animals, that some have claimed to glimpse a purpose in nature's operations. Even in this case the purpose is mere appearance, not reality. Indeed, in every type of animal organism, there subsists an order of things ... whose only effect is to lead to what seems to us to be a goal, but is essentially a neces­sity. The order achieves this necessity through the progressive development [Page 173] of parts, which are [also] shaped by environmental conditions (1815, in Corsi, 1988, p. 190).

  Since watchdogs tend to be more vigilant than publicists, Lamarck's opponents among the natural theologians often noted (and deplored) his materialism. The pious Reverend William Kirby, one of Britain's greatest entomolo­gists, made a statement that I regard as both trenchant and descriptively accurate (in Burkhardt, 1977, p. 189): “Lamarck's great error, and that of many other of his compatriots, is materialism; he seems to have no faith in anything but body, attributing every thing to a physical, and scarcely any­thing to a metaphysical cause. Even when, in words, he admits the being of a God, he employs the whole strength of his intellect to prove that he had noth­ing to do with the works of creation. Thus he excludes the Deity from the government of the world that he has created, putting nature in his place.”