The Structure of Evolutionary Theory

by Stephen Jay Gould

  * Since their debate has so often been misunderstood as an argument about evolution, something should be said about this anachronistic error. Cuvier's philosophy did foreclose any possibility of evolution. Geoffroy did accept a limited form of transmutationism, and he did write on the subject (particularly in his monographs on fossil crocodiles). We might also allow that his brand of formalism did encourage an acceptance of evolution as a possi­bility (as Owen's approbation also testifies) — for the generation of great and continuous di­versity (within channels) from an underlying archetype does establish a friendly climate for transmutation (at least within Bauplan), while Cuvier's optimized functionalism discour­ages any thought of evolutionary intermediacy. Evolution played a very minor role in the 1830 debate. Still, we will never understand the great antithesis of functionalism and for­malism — a subject that has pervaded the history of biology — if we misread this dichotomy in the later light of evolutionary theory. The debate between Cuvier and Geoffroy centered upon the primacy of form or function in morphology, hardly at all on evolution. Russell (1916, p. 66) has neatly summarized Geoffroy's views on evolution: “That he did believe in evolution to a limited extent is certain; that his theory of evolution was, as it were, a by­product of his lifework, is also certain. Geoffroy was primarily a morphologist and a seeker after the unity hidden under the diversity of organic form. His theory of evolution had as good as no influence on his morphology, for he did not to any extent interpret unity of plan as being due to community of descent.”

  * Owen conceived his mission as marrying the schools of “morphology and teleology,” or formalism and functionalism; but, as we shall see, he forged this union with a clearly dominant formalist partner, and therefore wins primary allegiance to this school by the proper criteria of relative frequency and primacy of cause. This casting may seem ironic or contradictory in the light of Owen's common designation as the “British Cuvier,” and of his youthful visit to Cuvier in 1831. But this common appellation primarily honored Owen's professional skill and his domination of the discipline, not his ideology (the intended comparison being to Cuvier's power, not to his ideas). Records of the Paris visit paradoxically affirm Cuvier's lack of influence; for, although Owen frequently visited Cuvier, his notes and diaries feature a strange lack of commentary on Cuvier's science or thought. Owen's grandson wrote in the standard life and letters of his grandfather (1894, vol. 1, p. 50): “His rough diary, which he kept during his stay in Paris, seldom mentions the fossil vertebrate collection, and shows that his interviews with Baron Cuvier were for the most part of a purely social character. It notes, for example, that he attended pretty regularly Cuvier's soi­rees held on Saturday evenings, and that he enjoyed the music. With the diary agree his letters. Both devote page after page to the sights and amusement of Paris. Owen, in fact, seems to have regarded his stay in Paris as an exceedingly pleasant and entertaining holiday.”

  Huxley's remarkably fair eloge of Owen makes the same point: “It was not uncommon to hear our countryman called 'the British Cuvier.' ... But when we consider Owen's contribution to “philosophical anatomy,” I think the epithet ceases to be appropriate. For there can be no question that he was deeply influenced by, and inclined towards, those speculations of Oken and Geoffroy St. Hilaire, of which Cuvier was the declared antagonist and often the bitter critic” (in Owen, 1894, vol. 2, p. 312).

  * The subtleties, centering upon two arguments, express great intellectual force, and con­stitute the power of natural selection in Darwin's version. First, since selection makes noth­ing, and can only choose among variations supplied by other causes, Darwin developed a theory of variation (copious, small in extent and isotropic — see pp.141-146) that reduced this evolutionary factor to a status as raw material only, and granted all power of change to the functional force of natural selection. Second, Darwin brilliantly used the classical rela­tive frequency arguments of “nooks and crannies” and “sequelae” to place formalist influ­ence upon evolutionary change at a periphery of unimportance relative to selection (see pp. 255-256).

  * Pardon a personal footnote, but this criterion — mechanics of the causal theory as a ba­sis for a taxonomy of ideas — fills me with deja vu, I pursued the same argument in my first book, Ontogeny and Phylogeny (1977b), when I made a primary separation between von Baer's laws of embryonic repetition and later divergence, and Haeckel's doctrine of recapit­ulation. Since these two principles often yield the same data for phylogenetic inferences (greater similarity of embryonic stages to ancestral morphologies), many biologists had tended to lump the two accounts together. Yet their causal mechanics are not only different, but strictly opposed — von Baer's as a theory of embryonic retention by unaltered inheri­tance, Haeckel's as a theory of active evolutionary change by acceleration of previously adult morphologies into early stages of descendant's ontogenies. I was able to show: (1) his­torically, that major actors in the great late 19th century debate on this topic supported my division by viewing von Baer and Haeckel as opposed; and (2) intrinsically, that the logic of my argument fell into a greatly clarified, and much more useful, structure under such a pri­mary division by causal mechanics. I believe that the general acceptance of my division, and the emergence of heterochrony (the active Haeckelian theme) as a useful and well defined concept in evolutionary theory (Alberch et ah, 1979; McKinney, 1999; Gould, 2000e) have vindicated my approach.

  * Darwin faced the same issue when he realized that natural selection, as an agent of anagenesis, could not fully encompass evolution without a separate explanation for multiplication of species — a gap that Darwin attempted to fill with his “principle of divergence” (see pp. 224-236).

  * In an obvious foray against Weismann, Eimer (1897) used this word in the subtitle of his second and final volume on orthogenesis — Ein Beweis bestimmt gericbteter Entwickelung und Ohnmacht der naturlichen Zuchtwahl bei der Artbildung (“a proof of definitely directed development and the weakness of natural selection in the origin of spe­cies”). I focus here on the more balanced view of external and internal forces presented in Eimer's first volume of 1888, his only major work translated into English, and therefore the main source of his influence among anglophone evolutionists. Natural selection gets short enough shrift in this balanced view (for Eimer, as we shall see, grants most external power to Lamarckian forces and little to Darwinian selection). But in the 1897 work, and largely (I suspect) as a result of long and bitter polemics with Weismann, Eimer became ever more dismissive about natural selection (while still giving lip service to the importance of envi­ronment, though largely through use and disuse). Eimer continually contrasts the Ohn­macht der Selection with the Herrschaft (domination) der Orthogenese. He considers natu­ral selection (1897, p. i) “von der geringsten Bedeutung” (of the smallest significance) as a factor of change. He cites the standard argument — that selection creates nothing but can only choose among variants presented by another process — as the Fundamental Einwurf (objection) to Darwin's system: “Selection can create nothing new, but can only work with characters that already exist and are useful in and of themselves” (1897, p. iii). Eimer waxes polemical about “the exaggerated, blinkered presentation of the principle of selec­tion, right up to the proclamation of Allmacht” (1897, p. iv), and he particularly lambastes assumptions made by strict selectionists about adaptation: “They satisfy themselves either in simply stating that this or that is 'adaptive,' thereby bringing investigation to an end; or they begin a round of groundless speculation, which surely has nothing to do with exact science” (1897, p. iv). Further comparison of Eimer's earlier “balanced” version of 1888 with his anti-Weismann polemic of 1897 would make an interesting historical project. The Weismann connection, as well as the polemical tone, may be sampled in Eimer's motto: “orthogenesis is the mortal enemy [Todfeind] not only of the omnipotence [Allmacht] of natural selection, but also of the hypothesis of the germ plasm that is based upon it.” (1897, pp. xiv-xv).

  * As a small
footnote in the logic of evolutionary theory and the history of Darwinian ar­guments, this notion of “phyletic life cycles” provides the best historiographic refutation of the old canard that natural selection is a tautology and therefore empty of content (see Bethel, 1976, and refutation in Gould, 1977c). This hoary claim, still a favored gambit of creationists, brands selection as a useless concept because its watchword — “survival of the fittest” — becomes meaningless when fitness is defined in terms of survival. The argument can be refuted in several ways (including the value of tautology in many scientific con­texts — see Sober, 1993), but Darwin's own rebuttal seems most compelling to me. Darwin did not define fitness retrospectively by observed survival. He insisted, in principle at least, that fitter organisms could be identified before any environmental test by features of pre­sumed biomechanical or ecological advantage. (The speediest deer can be specified before­hand, and their differential survival in a world of wolves can then be tested empirically.) Some critics dismiss Darwin's claim by arguing that no one would be foolish enough to pre­dict differential survival of the less adapted — and that such a Gedanken experiment there­fore becomes meaningless. But the theory of racial life cycles proves empirically that several leading evolutionary thinkers once made predictions of exactly this type as central proposi­tions of influential theories. As an essential postulate, Hyatt's theory held that less fit forms (by Darwin's a priori definition) would prevail over better-adapted individuals during peri­ods of phyletic old age and racial senescence. “Survival of the fittest” cannot be dismissed as an empty statement if alternative empirical claims not only can be formulated in princi­ple, but also actually build the core of historically important theories.

  * Cope recognized, of course, that many small-scale adaptations (changes in color and proportion, for example) could not be rendered as stages in a phyletic series of acceleration and terminal addition. These functional contingencies became the basis for recognizing dif­ferences at the species level — whereas new steps in the programmed sequence (or retreats down the staircase by retardation) set the chief criterion for establishing new genera. This attempt to designate taxonomic rank by the theoretical status of new features, while funda­mentally misguided by current views, gives us insight into older concepts about progress and predictability in evolution. In this feature, and ironically, Cope's early system is more truly Lamarckian than his later and more explicit “Neo-Lamarckism” — as this distinction between criteria of central and superficial change mirrors Lamarck's central concept of dif­ferent causes for upward and tangential evolution (see Chapter 3, pp. 186-189). Cope's view also entailed a logical paradox that helped to sink the theory — for Cope could not deny that two genera might reside in the same species if a small heterochronic change, and no other, occurred in one population of a lineage.

  * Hyatt's apparent illogic can be comprehended when we recognize that most scientists of his day regarded the biogenetic law as sufficiently well validated to represent a fact of na­ture. Many paleontologists maintained this conviction to quite recent times. My late senior colleague Bernie Kummel told me of an argument he once had with the most powerful American paleontologist in the generation just before his own — R. C. Moore. Kummel, as a young Turk, had disputed the universal validity of recapitulation over cocktails one evening. Moore slammed down his glass and bellowed: “Bernie, do you deny the Law of Gravity!”

  * We should give the last word, if only in a footnote, to the ever-perceptive T. H. Huxley, Darwin's stoutest supporter, but an incisive critic for several aspects of natural selection in its strict form. Whitman cited this passage from Huxley (Darwiniana) as an epigraphic quotation to one of his articles on orthogenesis (1919, p. 64): “But the causes and condi­tions of variation have yet to be thoroughly explored, and the importance of natural selec­tion will not be impaired even if further inquiries should prove that variability is definite and is determined in certain directions rather than in others, by conditions inherent in that which varies. It is quite conceivable that every species tends to produce varieties of a limited number and kind and that the effect of natural selection is to favor the development of some of these, while it opposes the development of others along their predetermined lines of modification.”

  * I also feel the strong tug of this theme, and for a personal reason. I was trained as a strict adaptationist, and I accepted and vigorously promoted this worldview in my early pa­pers. These works now embarrass me, with such statements as: “I acknowledge a nearly complete bias for seeking causes framed in terms of adaptation” (1966, p. 588 — at least I labelled the preference as a “bias”); and “... the fundamental problem of evolutionary pa­leontology — the explanation of form in terms of adaptation” (1967, p. 385). Yet I also felt the pull and fascination of the opposite set — though I had no inkling of the coordinated force behind the varied concepts, no explicit idea of the coherence (or even the terminol­ogy), and certainly no sense of the challenge thus posed to my juvenile certainties. I do not know why I felt the tug so strongly. But the ideas must cohere intrinsically if a young scholar can be so pulled by all of them, yet so unaware of their aggregation or their import. I discuss these personal aspects further in Chapter 1.

  * Central though this question may be to Bateson's inquiry (and sympathetic as I am to his book), I confess that I have never understood why Bateson regards this point as so tell­ing and decisive. I think that Darwin presented a simple and perfectly satisfactory solution to this dilemma (which he clearly recognized and discussed at length in early chapters of the Origin) — namely, that forms once filling the gaps between modern discontinuities have now become extinct. (Most intermediates, after all, are not contemporary creatures, but graded series on two lineages of ancestors running back to a common branching point, of­ten deep in the geological distance.)

  * Note Bateson's use of the term “position of organic stability.” Bateson here cites Galton's phrase, borrowed from his passages on the polyhedron metaphor. Bateson uses the phrase throughout his text, often with quotation marks to indicate its source. Thus, Bateson clearly embraced, as did many of his contemporaries, Galton's doubled-edged met­aphor for formalism against pure Darwinian externalism — though Bateson emphasized the facet flipping (saltational) rather than the inherently directional (orthogenetic) theme of the metaphor.

  * For this reason, I emphasize the logic of argument, rather than the psychologic of pre­sentation, throughout this book. If a minimal Darwinian “essence” resides within the logic of my three key statements about levels of selection, creativity in selection, and extrapolationism, then the Darwinian commitments of other scientists can be judged by degree of consonance with this necessary foundation. By this proper criterion, de Vries' saltational mechanism can only be labeled as anti-Darwinian (as the most astute of all commentators, Vernon Kellogg, recognized and illustrated) — whatever de Vries' psychological need for fe­alty to a personal hero.

  * De Vries wrote this book in English, so “species selection” represents his own chosen term and not the product of a translation. I note this fact with personal chagrin. “Species selection” has been a central component in the debates about punctuated equilibrium, and paleontologists have been discussing this idea intensely since the mid 1970's. We have all attributed the term to Stanley (1975), who brilliantly articulated the concept and its impli­cations. No one recognized that de Vries — in a book written in English by the world's lead­ing evolutionist at the time — not only developed the idea, but designated the concept by the same name (and realized the full set of implications). De Vries did not emphasize “species selection” or discuss the concept at length, but I can offer no excuse beyond my own inade­quate research for my ignorance of this point. See Gould (1993b) for my attempt at amends to this great scientist.

  * Note how Goldschmidt here uses the words that would become his title in 1940 — but only for the developmental theme in this passage.

  * Every time Darwin makes
such an overextended statement, his own honesty and sub­tlety draw him back immediately. The very next line presents the obvious caveat: “A num­ber of species, however, keeping in a body might remain for a long period unchanged, whilst within the same period, several of these species, by migrating into new countries and coming into competition with foreign associates, might become modified; so that we must not overrate the accuracy of organic change as a measure of time” (p. 488).

  * Not nefariously, in this case, for the great works of Cuvier and other catastrophists have always remained on library shelves, and have been much valued by historians and collectors. But never doubt the power of false characterization to ban effective consideration of the readily available. A scientist beyond the pale becomes an object of ridicule without being read — and the force of silence should never be underestimated. To cite just one per­sonal anecdote about Cuvier and his Discours preliminaire: The stereotyped Cuvier stands accused in most textbooks for arguing that catastrophes wipe all life off the face of the earth, and that God then creates new biotas from scratch. But Cuvier never advanced such a claim. No doubt, when pressed, he would have accepted some new creation to replenish a depleted world. But he attributed much local faunal change across stratigraphic boundaries to migration from previously isolated areas following geographic alterations that accom­pany episodes of rapid geological change (citing, as a potential example, the migration of Asian mammals to Australia should a land bridge ever connect these continents). Cuvier didn't hide this argument; he presents his viewpoint prominently in Section 30 of the ca­nonical Jameson translation (1818, pp. 128-129). Yet, at least a half dozen times in my professional life, colleagues ranging from graduate students to senior professors have ap­proached me with excitement, thinking that they had just made an important and original discovery: “Hey, look at this. Cuvier didn't believe in complete replacement by new cre­ation...” “Yes,” I reply, “page 128; the passage has always been there.”