The Structure of Evolutionary Theory

by Stephen Jay Gould

  Owen and Darwin first met for professional reasons after Darwin's return on the Beagle: Darwin had gathered the material (important bones of South American fossil mammals), and Owen possessed the anatomical skills. Lyell wrote to Owen on October 26, 1836, inviting him to dinner: “Among others you will meet Mr. Charles Darwin, whom I believe you have seen, just re­turned from South America, where he has labored for zoologists as well as for hammer-bearers” (in Owen, 1894, vol. 1, p. 102). The two men met and liked each other well enough. Darwin entrusted his Beagle material to his anatomi­cal colleague, and Owen became the taxonomic author of Toxodon and Dar­win's other spectacular finds.

  Their later antagonism arose for several reasons, some obvious, others less clear. Owen could surely be devious, arrogant and unpleasant. Darwin had struck a blow to the heart of Owen's system by substituting a flesh and blood ancestor, a concrete beastly thing, for the lovely, abstract, Platonic archetype. But something deeper and more intellectually honorable than simple jealousy lay at the core of their growing antipathy.

  Owen often enters the false dichotomies of standard histories as a virulent antievolutionist, the man who whispered into Wilberforce's ear before the famous debate with T. H. Huxley. If true, simple jealousy might provide an adequate motive: he who overturns my world, and (implicitly at least) makes me a fool in a profession I once dominated, can scarcely remain my compan­ion. Darwin contributed to this impression of Owen as a special creationist [Page 327] by so identifying him in the Origin and other writings — and Darwin cannot be entirely blamed for this mischaracterization. For Owen, never the clearest of writers, and ever the diplomat in an aristocratic world where he hob­nobbed with skill as a social climber and a seeker of support for his Museum, could be infuriatingly opaque in his stated commitments. In fact, Darwin, al­though characteristically genial and conciliatory to a fault in his writings, permitted himself a rare burst of trenchant irony in expressing his frustration at Owen's slippery attitude toward evolution. In the historical sketch added to later editions of the Origin, Darwin wrote:

  When the first edition of this work was published, I was so completely deceived, as were many others, by such expressions as “the continuous operation of creative power,” that I included Professor Owen with other paleontologists as being firmly convinced of the immutability of species; but it appears that this was on my part a preposterous error. In the last edition of this work, I inferred, and the inference still seems to me per­fectly just, . . . that Professor Owen admitted that natural selection may have done something in the formation of a new species; but this it ap­pears is inaccurate and without evidence. I also gave some extracts from a correspondence between Professor Owen and the editor of the “Lon­don Review” from which it appeared manifest to the editor as well as to myself, that Professor Owen claimed to have promulgated the theory of natural selection before I had done so; and I expressed my surprise and satisfaction at this announcement; but as far as it is possible to un­derstand certain recently published passages I have either partially or wholly again fallen into error. It is consolatory to me that others find Professor Owen's controversial writings as difficult to understand and to reconcile with each other, as I do (Darwin, 1872b, pp. xvii-xviii).

  One can certainly appreciate Darwin's frustrations. Owen did tailor his statements to circumstances and audiences, appearing cautious or critical as the case warranted, and always taking as much credit as possible. For exam­ple, Owen wrote a particularly nasty notice of the Origin in the April 1860 issue of the Edinburgh Review (published anonymously, following the tradi­tion of several leading journals at the time. Guessing the identity of review­ers — not at all difficult in this case — became a favorite Victorian intellectual pastime).

  In this commentary, Owen did proclaim the origin of species as the greatest of biological problems: “The origin of species is the question of questions in zoology; the supreme problem which the most untiring of our original la­borers, the clearest zoological thinkers, and the most successful generalizers have never lost sight of, whilst they have approached it with due reverence” (Owen, 1860, in Hull, 1973, p. 77).

  Writing anonymously, Owen praised himself as Darwin's unacknowledged predecessor in accepting the fact of evolution, but more cautious, and there­fore more worthy and philosophical, on the question of mechanisms: [Page 328]

  The great names to which the steady inductive advance of zoology has been due during these periods, have kept aloof from any hypothesis on the origin of species. One only, in connection with his paleontological discoveries, with his development of the law of irrelative repetition and of homologies, including the relation of the latter to an archetype, has pronounced in favor of the view of the origin of species by continuously operative creational law; but he, at the same time, has set forth some of the strongest objections or exceptions to the hypothesis of the nature of that law as a progressively and gradually transmutational one (Owen, 1860, in Hull, 1973, p. 184).

  Lest the term “creational law” still seem ambiguous, and lest anyone put the wrong name to the description, Owen later became more explicit. (In the jargon of Owen's day, a secondary cause operated under natural law, thus representing the subject matter of science. God, as “first cause,” may have es­tablished natural laws, and therefore secondary causes, at the beginning of time, but nature then unfolds under these invariable laws, thus defining the domain of science): “Owen has long since stated his belief that some preor­dained law or secondary cause is operative in bringing about the change” (Owen, 1860, in Hull, 1973, p. 210).

  I acknowledge Owen's opacity and shiftiness, but I also think that we should take him at face value here, for his claim follows his earlier writings, and also accords with the standard view of most formalist thinkers, espe­cially Geoffroy. I believe that Owen had, for more than a decade before the Origin appeared, accepted a limited form of evolution — within archetypes, and along channels preordained by archetypal constraints. He never accepted global transmutation, for his brand of limited evolution could not generate the archetypes themselves (which stand as primitive terms, or “givens” in his system), but could only produce variety within their permitted channels. I don't know how to read the famous last lines of the Nature of Limbs (see p. 322), except as a genuine statement of the usual formalist commitment to evolution in this admittedly restricted but entirely legitimate sense: “To what natural laws or secondary causes the orderly succession and progression of such organic phenomena may have been committed we are as yet ignorant. ... But if, without derogation of the Divine power, we may conceive the exis­tence of such ministers, and personify them by the term 'Nature' ...” (1849, p. 86).

  If we thus accept Owen as at least a halfhearted evolutionist long before the Origin, then the basis of his unhappiness with Darwin becomes easier to grasp. Darwin mocked Owen's caution, consumed his precious arche­type, made evolution global, and then proposed a central mechanism of change (natural selection) in the functionalist mode, diametrically opposed to Owen's formalist inclinations. Owen despised the extent and character of Darwin's evolutionism, but not the idea of evolution itself. All ideologists know that the enemy within provokes more intellectual danger and emo­tional distress than the enemy without. [Page 329]

  Consider Darwin's two principal apostasies from Owen's point of view. First, Darwin reconfigured the abstract archetype as a material ancestor, thus converting Platonism to materialism. Darwin wrote his challenge right in the margin of his personal copy of Owen's Nature of Limbs (quoted in Ospovat, 1981, p. 146): “I look at Owen's archetypes as more than ideal, as a real rep­resentation as far as the most consummate skill and loftiest generalization can represent the parent form of the Vertebrata.”

  Second, and even more disturbingly, Darwin inverted Owen's system, and the entire formalist program, by explaining the archetype in functional terms as a congeries of past adaptations materially inherited by descendants. Dar­win practically mocked Owen's formalism in the crucial p
aragraph on p. 206 that I cited to introduce this chapter (pp. 251–257). For Darwin used the jar­gon of the formalist-functionalist dichotomy — unity of type vs. conditions of existence — and then sank the formalist pole (along with Owen's most pre­cious concept of the archetype) into the alien functionalist sea of natural se­lection. In an important sense, Owen vs. Darwin on evolution replays Geoffroy vs. Cuvier on morphology.

  Others appreciated this reincarnation of formalism vs. functionalism, and cheered from the sidelines or within the fray. St. George Mivart, regarded by Darwin as his most cogent critic (see pp. 1218–1224), construed Owen's for­malism as a proper evolutionary exit from Darwin's bankrupt’s natural selec­tion:

  Owen . . . spread abroad in England the perception that a deep sig­nificance underlies the structure of animals — a significance for which no stress or strain and no influence of heredity, and certainly no mere practi­cal utility, can account. The temporary overclouding of this perception through the retrograde influence of Darwin's hypothesis of “natural se­lection” is now slowly but surely beginning to pass away ... Homologies for which neither heredity nor utility will account reveal themselves in the limbs of chelonians, birds, beasts, and most notably in those of man (from an 1893 statement by Mivart quoted in Owen, 1894, vol. 2, pp. 94-95).

  From the other side, Asa Gray understood Darwin's central contribution as the proper reintroduction of purpose, or functionalism, into biology. In 1874, Gray wrote to Nature (quoted in Ospovat, 1981, p. 148) that Darwin had done great service for biology by “bringing back to it teleology; so that, in­stead of morphology vs. teleology, we shall have morphology wedded to tele­ology” — in other words, functionalist hegemony by proper criteria of pri­macy and relative frequency. Darwin certainly appreciated the argument, for he wrote back to Gray: “What you say about teleology pleases me espe­cially.”

  In his usual perceptive manner, Russell (1916, p. 78) wrote: “The problem as Geoffroy and Cuvier understood it was not an evolutionary one. But the problem exists unchanged for the evolutionist, and evolution theory is essen­tially an attempt to solve it in one direction or the other.” So the problem ap­peared to Owen and Darwin; and so it remains for us today. [Page 330]

  Darwin's Strong but Limited Interest in Structural Constraint

  DARWIN'S DEBT TO BOTH POLES OF THE DICHOTOMY

  Darwin, who understood so well that all organisms must be shaped by their history, could scarcely build a great structure in the realm of ideas without subjecting his mental edifice to the same causal influences. Darwin's theory grew in the context of the formalist-functionalist struggle, and he knew and appreciated the issues and terminology (see Section I of this chapter).

  A cardinal premise of this book holds that Darwin must be ranked in the functionalist line — for the causal mechanics of his theory grant such clear pri­macy to adaptation, however subtly the argument develops.* But any evolu­tionary theory, in adding a historical dimension to reshape the simpler world of the formalist-functionalist dichotomy, would necessarily draw upon both axes of the old system to build the new, orthogonal dimension of temporal change. In two senses, formalist thought included great potential to influence Darwin's evolutionary views.

  First, and most obvious in its possibilities (though largely unrealized by Darwin, I shall argue), classical formalism, with its key concept of transformational channels within the bounds of archetypal design, followed a logic intrinsically favorable to a limited form of evolution, while the optimalist functionalism of Paley, or of Cuvier, rooted the impossibility of transmuta­tion in the core of their central argument (Cuvier on correlation of parts, for example). Many of the leading pre-Darwinian formalists supported evolution in this restricted sense — Geoffroy in France, Meckel in Germany, Owen in England. Yet, although Darwin could not have been isolated from this influ­ence, I see no strong evidence that his decision to embrace evolutionism de­rived from this source (Gruber and Barrett, 1974; Schweber, 1977; Sulloway, 1982). Moreover, the most prominent and fully elaborated evolutionary the­ory available to Darwin (however strongly he rejected the formulation — see pp. 192–197) resided in the opposite camp of Lamarck's functionalism. Finally, the content of Darwin's theory, from his earliest codification in 1838, stood clearly outside formalist thinking — both in replacing archetypes with flesh and blood ancestors built by adaptation, and in advocating the function­alism of natural selection itself.

  Secondly, Ospovat (1981) presented the important thesis that Darwin's chief intellectual change within his theory of natural selection between codification [Page 331] in 1838 and publication in 1859, lay in abandoning an original be­lief in perfect adaptation, and in accepting the crucial concepts of relative ad­aptation (“locally better than”) and imperfection imposed by constraints of phyletic history — the argument that completed Darwin's rejection of Paleyan optimality, first in accepting evolution rather than God as the architect of morphology, and only later in recognizing that history implies imperfection in current design. I accept Ospovat's claim that this change must be interpreted as fundamental to the structure of Darwin's theory — see my own extended treatment of Darwin's crucial use of imperfection as primary evidence for evolution itself (pp. 111-116). Finally, Ospovat demonstrates that Owen's disparagement of teleology, and the formalist notion of constraint imparted by archetypes and rules for their transformation, played an important role in Darwin's shift of attitude.

  If Owen's formalism influenced Darwin in this manner, then why should Darwin be placed so firmly in the functionalist line? The first and most defin­itive answer must cite the obvious statement that Darwin explicitly so identi­fied himself — in using the defining terms of the formalist-functionalist debate (unity of type vs. conditions of existence), and in declaring his allegiance to the functionalist proposition as a “higher law” subsuming unity of type as past adaptation (see pp. 251–260).

  The broader reason lies in a decision to concentrate on the causal mechan­ics of explanatory theories for evolutionary change. Evolution opens its um­brella over a vast subject, with many concerns and meanings. Taxonomy of attitudes might designate several alternative criteria for subdivisions. I believe that the logic of the inner workings of the primary causal theory — the struc­ture of evolutionary theory, as I named this book — should hold primacy in definitions.* Darwin developed strong ideas about history, attitudes towards analogy, convictions about geology, and philosophical grounding in much that the Victorian age held dear. But his mechanism of evolutionary change — the theory of natural selection — rests upon a central logic, a mode of working, [Page 332] that should define the basic attributes of Darwinism. The theory of natu­ral selection is functionalist, by Darwin's own recognition and definition, and by routes (and roots) of causality inherent in its proposed mechanics. In­trinsic factors supply copious, small-scale, isotropic variation — raw material only, and no direct cause or impetus of change. Evolutionary change occurs by natural selection, as organisms adapt to modify local environments. The mechanism of evolutionary change therefore remains functionalist in Dar­win's theory; selection powers change, and organisms adapt as a primary re­sult. Darwin also defined his major problem squarely within the functionalist theme of adaptation, as he wrote, in the oft-quoted statement in the Introduc­tion to the Origin (1859, p. 3), that much evidence, from many sources, could validate the factuality of evolution itself, but that “nevertheless, such a con­clusion, even if well founded, would be unsatisfactory until it could be shown how the innumerable species inhabiting this world have been modified, so as to acquire that perfection of structure and coadaptation which most justly ex­cites our admiration.” The argument from imperfection (pp. 111-116) dem­onstrates the factuality of evolution, but we need to explain adaptation if we wish to understand the mechanism.

  DARWIN ON CORRELATION OF PARTS

  Many evolutionists can cite the specific example given by Darwin in Chapter 6 of the Origin (“Difficulties on Theory”) in allowin
g that not all useful structures arise by natural selection for their current role, however essential that function may be to the life of the organism: “The sutures in the skulls of young mammals have been advanced as a beautiful adaptation for aiding par­turition, and no doubt they facilitate, or may be indispensable for this act; but as sutures occur in the skulls of young birds and reptiles, which have only to escape from a broken egg, we may infer that this structure has arisen from the laws of growth, and has been taken advantage of in the parturition of the higher animals” (1859, p. 197).

  But I suspect that few of our colleagues know that Darwin took this exam­ple directly from the much longer and more detailed treatment in Richard Owen's famous essay (1849) On the Nature of Limbs (see quote from Owen on p. 325). This example not only provides direct evidence for an influence of the leading British structuralist thinker upon Darwin (a link often denied be­cause we misread Owen as a creationist, and a leader of the rearguard), but also illustrates a fascinating change of emphasis between Owen's invocation and Darwin's borrowing. For Owen, the example lies at the heart of nature's primary causal structure, and serves as a point of entry to his most trenchant critique of functionalism in general — particularly his claim that explanations of organic form in terms of utility match the barrenness of the Vestal Virgins in Bacon's famous aphorism. For Darwin, on the other hand, the example il­lustrates an exception to natural selection included within a chapter entitled “Difficulties on Theory.”

  But Darwin's serious concern with structural constraint cannot be denied [Page 333] (and his borrowed example from Owen surely indicates respect and atten­tion). Darwin linked the efficacy of natural selection to a set of assumptions about the nature of variation (see pp. 141–146). But he could not be satisfied with such an abstraction, and he recognized that any complete theory re­quired an understanding of mechanisms of variation. Darwin presented his major discussion of constraint in Chapter 5 on “Laws of Variation” — for any exception to his trio of necessary properties for variation (copious, small in extent, and undirected) would compromise the exclusive power of natural se­lection by granting a role to “internal” principles of variation in the direction of evolutionary change. Any exception, in short, would represent a “law of variation” acting not only as a source of raw material, but also as a subsid­iary to natural selection among causes of change. (In both the Origin (1859) and in his extended treatise on The Variation of Animals and Plants Under Domestication (1868), Darwin employed the phrase “laws of variation” to specify properties that could produce evolutionary change independent of natural selection. He proposed a threefold taxonomy of “use and disuse,” “direct action of the external environment” — the bases of evolutionary theo­ries often attributed to Lamarck and Geoffroy respectively — and “correlation of growth,” or structural constraint as discussed in this chapter. Properties of variation that merely supplied raw material for natural selection — the indis­pensable trio of copious, small, and undirected — apparently did not count among the “laws,” for Darwin viewed these properties as auxiliaries or hand­maidens of selection.)