The Structure of Evolutionary Theory

by Stephen Jay Gould

  And yet, though Darwin now presents more examples of correlated vari­ability than he had provided in 1859, the main thrust of the 1868 volumes, as we shall see in the next section, only accentuates the dominance of selection over these exceptions to its domination.

  THE “QUITE SUBORDINATE POSITION” OF CONSTRAINT

  TO SELECTION

  Paradoxically perhaps, an absence of strong concern for a subject may be better expressed by inadequacy of treatment than by total evasion. Had Dar­win simply omitted the subject of correlated variability altogether, we would know little about his attitude toward constraint (however much we might in­fer from his indifference). But the weakness of his limited discussion reveals far more. As E. S. Russell rightly remarks (1916, p. 240): “Darwin's concep­tion of correlation was singularly incomplete. As examples of correlation he advanced such trivial cases as the relation between albinism, deafness and blue eyes in cats, or between the tortoise-shell color and the female sex. He used the word only in connection with what he called 'correlated variation.' ... He took it for granted that the 'correlated variations' would be adapted to the original variation which was acted upon by natural selection.” (I find Russell a bit harsh in this particular claim. Correlated variability may be “adapted” to the primary target of selection in the sense that organic coher­ence in growth must be maintained, for the correlation marks an inherited pathway. But Darwin clearly designates the realized feature as potentially in­dependent of utility, though admittedly not harmful, for selection would then work to eliminate it.)

  Darwin's discussion of variation in the Origin clearly illustrates his convic­tion about the primacy of selection over any internal drive in supply of varia­tion or constraint. Variation (with its three key properties) must be available as raw material, but all shaping and change arise by natural selection. In those rare cases when change can be traced directly to variability itself — that is, when a phenomenon caused by variability becomes visible as a property of a population, rather than the oddity of an individual — we must seek an ex­planation in the relaxation of selection's usual vigilance and control. Rudi­mentary parts, for example, become free from selective control, and may therefore manifest the influence of subsidiary factors, including constraint, that selection would ordinarily mask: “Rudimentary parts, it has been stated by some authors, and I believe with truth, are apt to be highly variable . . . their variability seems to be owing to their uselessness, and therefore to natu­ral selection having no power to check deviations in their structure. Thus ru­dimentary parts are left to the free play of the various laws of growth, to the effects of long continued disuse, and to the tendency to reversion” (1859, pp. 149-150).

  The concluding statement in Chapter 5 on “Laws of Variation” clearly expresses the domination of selection: “Whatever the cause may be of each [Page 340] slight difference in the offspring from their parents, and a cause for each must exist, it is the steady accumulation, through natural selection, of such differ­ences, when beneficial to the individual, that gives rise to all the more impor­tant modifications of structure, by which the innumerable beings on the face of this earth are enabled to struggle with each other, and the best adapted to survive” (1859, p. 170).

  But the fuller treatment of the 1868 work — Darwin's longest book, and primarily written as a treatise on variation, not selection — asserts this funda­mental claim even more forcefully. As in the Origin, Darwin does allow (1868, vol. 2, p. 320 and p. 355, for example) that correlations of growth produce features (including taxonomically important markers) independent of utility. But the domination of selection, by arguments of relative frequency and importance, now becomes even more explicit, as in this statement from the introductory chapter: “I shall in this volume treat, as fully as my materials permit, the whole subject of variation under domestication. We may thus hope to obtain some light, little though it be, on the causes of variability.... During this investigation we shall see that the principle of Selection is all im­portant. Although man does not cause variability and cannot even prevent it, he can select, preserve, and accumulate the variations given to him by the hand of nature in any way which he chooses; and thus he can certainly pro­duce a great result” (1868, vol. 1, p. 3).

  Darwin, as I argued previously, may not be a consistently brilliant writer in the tradition of Huxley or Lyell. But he did exceed his more stylish colleagues in a literary gift for inventing metaphors that capture the essence of complex ideas. Most of Darwin's enduring lines fall into this category — the face of na­ture bright with gladness, struggle for existence, survival of the fittest, wedg­ing as an image for competition. The Origin introduces strikingly appropriate and beautifully crafted metaphors in crucial places — the tree of life in the summary of natural selection at the end of Chapter 4 (pp. 129-130), and the entangled bank of the book's final flourish (p. 489).

  Darwin also developed a remarkable metaphor to summarize his convic­tion about the relative importance of selection and variation. He introduced this long passage at the end of his chapter on selection in his 1868 work — the “interloper” chapter, if you will, where the dominating force surveys an en­tire volume devoted to subservients. We might label this image as the meta­phor of building stones for the house of morphology. No other passage in all Darwin's writing so strongly illustrates the domination of selection over raw material.

  All components for the primacy of selection, and for the inconsequentiality of constraint (and other internal factors), flow together in this striking image: Selection may depend upon variation, but the character of variation hardly matters (so long as appropriate amounts and styles be present), given the power of selection. Variation cannot be truly random, and we should interest ourselves in its particular forms and biases (the shapes of stones used by the mason). But, in the deepest sense, these preferred forms exert no influence upon the final building when selection (the architect) takes charge. For laws [Page 341] of variation (shapes of stones) “bear no relation to the living structure which is slowly built up” (the form of the building). An architect, armed with a blueprint and enough stones, can build the desired structure, whatever the shapes of pieces available for construction. Thus, “variability sinks to a quite subordinate position in importance in comparison with selection.”

  Throughout this chapter and elsewhere I have spoken of selection as the paramount power, yet its action absolutely depends on what we in our ignorance call spontaneous or accidental variability. Let an architect be compelled to build an edifice with uncut stones, fallen from a precipice. The shape of each fragment may be called accidental; yet the shape of each has been determined by the force of gravity, the nature of the rock, and the slope of the precipice — events and circumstances, all of which depend on natural laws; but there is no relation between these laws and the purpose for which each fragment is used by the builder. In the same manner the variations of each creature are determined by fixed and im­mutable laws; but these bear no relation to the living structure which is slowly built up through the power of selection, whether this be natural or artificial selection.

  If our architect succeeded in rearing a noble edifice, using the rough wedge-shaped fragments for the arches, the longer stones for the lintels and so forth, we should admire his skill even in a higher degree than if he had used stones shaped for the purpose. So it is with selection, whether applied by man or by nature; for though variability is indispensably necessary, yet, when we look at some highly complex and excellently adapted organism, variability sinks to a quite subordinate position in im­portance in comparison with selection, in the same manner as the shape of each fragment used by our supposed architect is unimportant in com­parison with his skill (1868, vol. 2, pp. 248-249).

  I suggest, as a major theme of this book, that Darwinian evolutionists, ever since, have placed too much confidence in this edifice. Darwin's metaphorical structure, fully shaped by the architect of natural selection, cannot be dis­missed as a house of cards, bu
t the walls have developed some cracks and may even be ripe for a breach.

  [Page 342]

  CHAPTER FIVE

  The Fruitful Facets of Galton's

  Polyhedron: Channels and Saltations

  in Post-Darwinian Formalism

  Galton's Polyhedron

  Charles Darwin often remarked, as in the Descent of Man (1871, vol. 1, pp. 152-153), that he had pursued two different goals as his life's work: “I had two distinct objects in view; firstly, to show that species had not been sep­arately created, and secondly, that natural selection had been the chief agent of change.” Darwin spoke wisely (and practically) in these lines. He lies in Westminster Abbey for his unbounded success in the first endeavor; whereas, unbeknownst to many evolutionists who have experienced only the age of natural selection's triumph since the 1930's, Darwin's theory, his second en­deavor, never enjoyed much success in his lifetime, and never attracted more than a modest number of adherents. (The titles of Peter Bowler's excellent his­torical treatises on late 19th century evolutionary thought capture this para­dox well — The Eclipse of Darwinism, 1983; and The Non-Darwinian Revo­lution, 1988.)

  As discussed in Chapter 2 (see pp. 137–141), Darwin's evolutionary critics encountered their greatest stumbling block in their inability to envision natu­ral selection as a creative force. Natural selection could surely serve as an executioner or headsman — the eliminator of the unfit. But such a negative role must occupy a distinctly secondary rank in the panoply of evolutionary forces. The central question of evolutionary theory remains: what creates the fit? The difficulty of this question, and the supposed inadequacy of natural se­lection as a solution, inspired a vast literature, including two famous Darwin­ian title parodies by two leaders of his opposition — The Origin of the Fittest (1887) by the American Neo-Lamarckian E. D. Cope, and The Genesis of Species (1871) by the British structuralist and saltationist St. George Mivart. (Darwin regarded Mivart's criticism as especially serious; the only chapter that he ever added to later editions of the Origin — Chapter 7 on “Miscel­laneous objections” — largely presents a rebuttal of Mivart's critique — see Chapter 11, pp. 1218–1224, for a full analysis.)

  I cannot present a complete taxonomy of alternative proposals in late 19th [Page 343] century thought (lest this section become a multivolumed book in itself). I will, instead, commit the primary historical “sin” of self-serving retrospec­tion, and focus on those critiques of creativity that stressed formalist or structuralist themes now relevant in modern reformulations of evolutionary theory. Thus, I ignore several of the most important currents in late 19th cen­tury debate, particularly the strong Lamarckism of many thinkers, and the various threads of theistic and other forms of finalistic directionality.

  Since Darwin's essential trio of assumptions about variability — copious, small, and undirected (see pp. 141–146) — does permit natural selection to act as the creative force of change, non-Darwinian alternatives, by logical neces­sity, deny one or more of these assertions. The diverse formalist theories of this chapter gain conceptual unity in granting directional power to internal factors, and not only to the interaction of environment with isotropic raw material. Darwinian claims for the small size and nondirectional character of variations become the obvious candidates for confutation — for formalist alternatives to these Darwinian bastions grant directional power to inter­nal causes (whereas a denial of the third claim of copiousness only places lim­its upon natural selection without supplying any substitute as a cause of change). Thus, in this late 19th century heyday of alternatives to Darwinism, formalist and structuralist thought centered upon claims for the evolutionary importance of saltational and directional variation.

  The most striking model and epitome for this formalist opposition derives from a source that will strike many evolutionists as anomalous or paradoxi­cal — Darwin's brilliant and eccentric cousin Francis Galton. (The two men shared Erasmus Darwin as a grandfather, surely the most eminent member of the family before Charles himself.) Galton did study continuous variation extensively, and he therefore gained a reputation as guiding spirit for the lead­ing biometricians, Pearson and Weldon. Moreover, his long-term trumpeting of eugenic improvement also promoted the assumption that he favored insen­sibly gradual and continuous change in evolution.

  But Galton, a pluralist in his views on evolutionary causality, viewed discontinuous variation as even more efficacious. Echoing Huxley's frequent plea to Darwin for a larger permissible size in useful variants (advice that Darwin explicitly rejected because he understood so well that the creativity of natural selection would be seriously compromised thereby), Galton wrote (1889, p. 32) that evolutionary theory “might dispense with a restriction for which it is difficult to see either the need or the justification, namely, that the course of evolution always proceeds by steps that are severally minute, and that become effective only through accumulation. That the steps may be small and that they must be small are very different views; it is only to the lat­ter that I object.”

  We all recognize Galton's main contribution to the study of continuous variation in his recognition and elucidation of the crucial concept of regres­sion toward the mean. But Galton did not interpret regression in a modern genetic light. For him, regression guaranteed that continuous variation could not yield progressive evolutionary change, because all favorable extremes [Page 344] will, in subsequent generations, regress towards the mean, and no permanent or directional modification can therefore accrue. Substantial change to new “types” and taxa must occur by the occasional production of true-breeding “sports” — discontinuous variants that do not meld to intermediacy in hybrid offspring, and are therefore not subject to regression. In thus emphasizing “sports” for evolutionarily efficacious variation, and stability for taxa at other times (as regression to the mean holds continuous variants in check), Galton also became a hero of the early Mendelians, Bateson and de Vries, particularly for his role in formulating a general rationale for their non-Dar­winian concept of saltational origin for new species by macromutation.

  Galton subsumed both non-Darwinian formalist themes of discontinuity in effective variation and internally-generated, preferred channels of change (constraints) in a brilliant metaphor that I have called “Galton's polyhedron” (see Gould and Lewontin, 1979). This image had been forgotten by 20th cen­tury biologists, but many of Galton's contemporaries discussed the model and its implications. Mivart (1871) invoked the polyhedron as a centerpiece of the critique that most attracted Darwin's attention and response (see Mivart, 1871, pp. 97, 113, and 228); W. K. Brooks (1883, p. 296) cited this image in the most important American treatise on variation, a book that strongly in­fluenced Brooks's visiting student, William Bateson. Bateson (1894, p. 42) then described “the metaphor which Galton has used so well — and which may prove hereafter to be more than a metaphor.” Kellogg (1907), speaking of Galton's “familiar analogy” (p. 332), considered the polyhedron as an ideal illustration for the key non-Darwinian challenge of heterogenesis (salta­tional evolution). And de Vries (1909, p. 53) stated that Galton's polyhedron expressed his own view of variation “in a very beautiful way.”

  Galton introduced the metaphor of the polyhedron in his eugenic mani­festo and most influential book, Hereditary Genius (1869). In discussing “stability of types” in the closing chapter on “general considerations,” Galton presented his model in an overtly material, and petrological, form:

  The mechanical conception would be that of a rough stone, having, in consequence of its roughness, a vast number of natural facets, on any one of which it might rest in “stable” equilibrium. That is to say, when pushed it would somewhat yield, when pushed much harder it would again yield, but in a less degree; in either case, on the pressure being withdrawn, it would fall back into its first position. But, if by a powerful effort the stone is compelled to overpass the limits of the facet on which it has hitherto found rest, it will tumble over into a new position of sta­bility, when
ce just the same proceedings must be gone through as before, before it can be dislodged and rolled another step onwards. The various positions of stable equilibrium may be looked upon as so many typical attitudes of the stone, the type being more durable as the limits of its sta­bility are wider. We also see clearly that there is no violation of the law of continuity in the movements of the stone, though it can only repose in certain widely separated places (1884 edition, p. 369). [Page 345]

  Twenty years later, in Natural Inheritance (1889), the metaphor moved from an afterthought in the back of the book to the focal argument of an early chapter on “organic stability” (pp. 18-34). Galton now granted the im­age an abstract and formal geometry — as a polyhedron (based on a model that he actually built). He also supplied an illustration (reproduced as Fig. 5-1).

  It is a polygonal slab that can be made to stand on any one of its edges when set upon a level table ... The model and the organic structure have the cardinal fact in common, that if either is disturbed without trans­gressing the range of its stability, it will tend to re-establish itself, but if the range is overpassed it will topple over into a new position . . . Though a long established race habitually breeds true to its kind, subject to small unstable deviations, yet every now and then the offspring of these deviations do not tend to revert, but possess some small stability of their own. They therefore have the character of sub-types, always, how­ever, with a reserved tendency under strained conditions, to revert to the earlier type. The model further illustrates the fact that sometimes a sport may occur of such marked peculiarity and stability as to rank as a new type, capable of becoming the origin of a new race with very little assis­tance on the part of natural selection...