Darwin saw many wonderful things on the Beagle; nature challenged him, broadened his view, and instilled flexibility. Darwin returned to England with the tools of conversion, but still as a creationist, however suffused with doubts and questions (Sulloway, 1982a; Gruber and Barrett, 1974; Schweber, 1977; Kohn, 1980; Desmond and Moore, 1991; Browne, 1995). As for the Galapagos, he had missed the story of the finches entirely, because he had been fooled by their convergences and had not recognized the underlying taxonomic [Page 193] unity (Sulloway, 1982b; Gould, 1985c). Darwin had been informed that the tortoises differed from island to island, but had failed to appreciate the significance of this claim. Naturalists then believed, falsely, that the Galapagos tortoises were not indigenous but had been transported in recent memory by Spanish buccaneers and placed on the Galapagos as a source of food for revictualizing ships. Thus, Darwin must have reasoned, if the tortoises had only been on the Galapagos for two or three centuries, differences among islands could not be consistent or meaningful. The Beagle had provisioned itself with several tortoises, stored live in the hold as meat on the hoof, so to speak. In a modern version of fiddling while Rome burned, Darwin partook of the feasts but made no plea for conservation when his shipmates then pitched the carapaces overboard.
Darwin became an evolutionist by returning to England and immersing himself in the scientific culture of London — by arguing with colleagues, by reading and pondering (mostly in the library of the Atheneum Club), by seeking good advice (learning from ornithologist John Gould, for example, that those diverse Galapagos birds were all finches). He exploited a broad range of humanistic Western culture in pursuing his struggle for intellectual reform in natural history. He read Plato, Milton, and Wordsworth. He constructed the theory of natural selection, as argued in the last chapter, in conscious analogy with the laissez-faire theories of Adam Smith and the Scottish economic school. Darwin, without the impetus and challenge of this intellectual environment, might have become a country parson, with a beetle collection maintained by an ecclesiastical sinecure as the remnant of a childhood passion for natural history.
In this enlarged perspective on the origin of Darwin's evolutionary views, the importance of his precursors becomes greatly enhanced. Lamarck and Chambers* do not figure as irrelevancies to be ignored, or (even worse) as impediments [Page 194] to be overcome; they represent an essential part of his context for study, and they played a major role in shaping the radical and different character of Darwin's distinctive theory.
I believe that Lamarck had a far greater influence on Darwin than tradition has allowed (a point advanced by other historians of science as well — see Corsi, 1978; Mayr, 1972, p. 90). I base this claim on Darwin's own contact with Lamarck's works, his private reactions as recorded in letters, and the eventual content of his theory. I don't claim that Darwin devised natural selection [Page 195] as a conscious point-by-point contrast or refutation of Lamarck, but I suspect that Darwin clearly recognized what he liked least in Lamarck and strove to formulate a theory of opposite import.
Darwin said little about Lamarck in his published works, with no explicit reference to Lamarck's evolutionary views in the first edition of the Origin, and grudging praise in the historical preface added to later editions. But we know that he studied Lamarck intensely, and didn't like what he read. Darwin owned a copy of the 1830 printing of the Philosophie zoologique (see Hull, 1985, p. 802), and read the book while making heavy annotations at least twice. More important, perhaps, Lamarck had provided Darwin's introduction to evolution via Lyell's fair but critical exegesis in the Principles of Geology.
Lyell's characterization becomes particularly interesting because he emphasizes, in his masterful prose, the very two points that Darwin would strive most mightily to correct. First, Lyell castigates Lamarck for making assertions without a shred of direct evidence. Note that Lyell directs his scorn not at the palpable forces of lateral adaptation, but at claims for the origin of new organs as increments of complexity wrought by the forces of progress:
We point out to the reader this important chasm in the chain of the evidence, because he might otherwise imagine that we had merely omitted the illustrations for the sake of brevity, but the plain truth is, that there were no examples to be found; and when Lamarck talks “of the effects of internal sentiment,” “the influence of subtle fluids,” and the “acts of organization,” as causes whereby animals and plants may acquire new organs, he gives us names for things, and with a disregard to the strict rules of induction, resorts to fictions, as ideal as the “plastic virtue,” and other phantoms of the middle ages (Lyell, 1832, p. 8).
Second, and more important for my argument, Lyell gives a crisp and accurate account of Lamarck's hierarchical view of evolutionary causality, emphasizing the contrast between the regular cause of progress, and the disrupting force of adaptation. The passage, worth quoting in extenso, probably represents Darwin's first contact with this invalid style of hierarchical theory:
Nature is daily engaged in the formation of the elementary rudiments of animal and vegetable existence, which correspond to what the ancients termed spontaneous generations . . . These are gradually developed into the higher and more perfect classes by the slow, but unceasing agency of two influential principles: first, the tendency to progressive advancement in organization, accompanied by greater dignity in instinct, intelligence, etc.; secondly, the force of external circumstances, or of variations in the physical condition of the earth, or the mutual relations of plants and animals ... Now, if the first of these principles, the tendency to progressive development, were left to exert itself with perfect freedom, it would give rise, says Lamarck, in the course of ages, to a graduated scale of being, where the most insensible transition might be traced from the simplest to the most compound structure, from the humblest to the most exalted degree [Page 196] of intelligence. But in consequence of the perpetual interference of the external causes before mentioned, this regular order is greatly interfered with, and an approximation only to such a state of things is exhibited by the animate creation, the progress of some races being retarded by unfavorable, and that of others accelerated by favorable, combinations of circumstances. Hence, all kinds of anomalies interrupt the continuity of the plan, and chasms, into which whole genera or families might be inserted, are seen to separate the nearest existing portions of the series (Lyell, 1832, pp. 13-14).
Darwin's public silence (or mild approbation) is belied by his consistently negative attitude towards Lamarck, as recorded in private letters extending from the 1840's to post-Origin years. In 1844, he wrote to Hooker on the dearth of available evolutionary writing: “With respect to books on this subject, I do not know of any systematical ones except Lamarck's, which is veritable rubbish” (in F. Darwin, 1887, volume 2, p. 29). The most interesting post-Origin references occur in letters to Lyell, who criticized Darwin for not giving Lamarck sufficient credit. In responding to Lyell's first reaction to the Origin, Darwin wrote on October 11, 1859 (in F. Darwin, 1887, volume 2, p. 215): “You often allude to Lamarck's work; I do not know what you think about it, but it appeared to me extremely poor; I got not a fact or idea from it.” Perhaps, but I suspect that Darwin got many a concept against it.
Darwin's longest statement, a testy comment directed against Lyell's repeated designation of Lamarck as a source (though mitigated at the end by Darwin's usual humor), conveys special insight in Darwin's stated rationale for rejecting Lamarck's theory so firmly:
Lastly, you refer repeatedly to my view as a modification of Lamarck's doctrine of development and progression. If this is your deliberate opinion there is nothing to be said, but it does not seem so to me. Plato, Buffon, my grandfather before Lamarck, and others, propounded the obvious view that if species were not created separately they must have descended from other species, and I can see nothing else in common between the “Origin” and Lamarck. I believe this way of putting the case is very injurious to its a
cceptance, as it implies necessary progression, and closely connects Wallace's and my views with what I consider, after two deliberate readings as a wretched book, and one from which (I well remember my surprise) I gained nothing. But I know you rank it higher, which is curious, as it did not in the least shake your belief. But enough, and more than enough. Please remember you have brought it down on yourself!! (in F. Darwin, 1887, volume 2, pp. 198-199).
Note the basis of Darwin's critique — “very injurious ... as it implies necessary progression.” In other words, Darwin dismisses the higher-level cause of Lamarck's hierarchy. Darwin's own theory, of course, rested on the complete sufficiency, by extrapolation, of the lower-level force of adaptation in [Page 197] Lamarck's system (as produced by the different mechanism of natural selection).
Lamarck's hierarchical theory set a context (in opposition) for Darwin's distinctive single-level theory of extrapolation, based on uniformitarian assumptions, from the palpable and small-scale cases of adaptation that surround us to all evolutionary changes at all scales of time and magnitude. Natural selection does not emerge from the raw observation of nature, but as a complex idea embedded both in observation and in Darwin's voracious study and trenchant analysis of contemporary ideas in biology and general culture. Lamarck's hierarchical theory formed an important, though not widely recognized, part of the mix, distilled by Darwin to extract a theory that would change the world.
Darwin directed his anti-hierarchical theory against Lamarck's old and invalid concept of hierarchy — different and opposed causes at distinct levels. Darwin labored mightily to encompass the entire domain of evolutionary causation within a single level — natural selection working on organisms. He knew what he wanted to do, and he pursued and extended the logic of his argument relentlessly. Most of his supporters (including Wallace and Huxley) never understood the subtle logic of the single-level theory. Among the few who did, Weismann also made a strenuous effort to bring the system to completion. I find the strongest historical support for modern versions of hierarchy (same causes working in different ways at various levels, in direct contrast with Lamarck's notion of disparate causes in opposition), in the intense intellectual struggle carried out by the two greatest selectionists of the 19th century — Weismann for lower levels, and Darwin himself for upper levels — to bring the nonhierarchical theory of selection to completion and sufficiency. Both men, as we shall see in the subsequent sections of this chapter, struggled valiantly, but could not prevail. (Chapter 5 will then discuss (in the saltationist context of his version) the other major hierarchical system of evolutionary thought before the Modern Synthesis — the fascinating and subtle early 20th century theory of de Vries on reintroducing selection at the species level after denying its central importance at Darwin's own level of the organism.) Call it the bad penny that keeps cropping up, or the pearl of great price always found within, but hierarchy seems unavoidable. Could the basic reason for this persistence find an explanation in something so lovely, and so beautifully simple, as truth-value?
No Allmacht without Hierarchy: Weismann
on Germinal Selection
THE ALLMACHT OF SELECTION
In 1893, Herbert Spencer, who had a word (many of them) and a thought for nearly everything,* published a long critique in the Contemporary Review — [Page 198] “The Inadequacy of 'Natural Selection.'” He strongly supported the Lamarckian principle of use and disuse with inheritance of acquired characters and, while not denying the importance of Darwin's principle, railed against the exclusivity claimed for natural selection by August Weismann and his school, variously labelled as “strict,” or “ultra” or “neo” Darwinism. Weismann quickly rose to the challenge, choosing for his title a phrase that would become a motto for his approach. He called his rebuttal, in German, and parrying Spencer directly: Die Allmacht der Naturzuchtung — a title rendered by the English translator as “The All-Sufficiency of Natural Selection” (although I would prefer “Omnipotence,” or the literal “All-Might”).
This exchange (Spencer, 1893a and b; Weismann, 1893) became the focal point and most widely cited set of documents in the great debate between “neo-Darwinism” and “neo-Lamarckism,” perhaps the hottest subject in evolutionary theory at the end of the 19th century (see Kellogg, who wrote, 1907, p. 134 — “The best known part of the general debate was that carried on directly by Weismann and Spencer in the Contemporary Review.” These terms, as so often noted, bear little relation to the chief concerns of the name-bearers. Neo-Lamarckians bypassed Lamarck's central concept of materialistic progress and focused on a theory of heredity that Lamarck espoused as the folk wisdom of his day, not as anything distinctive in his system. Neo-Darwinism referred to the panselectionism of Weismann and Wallace, an attitude explicitly and pointedly rejected by Darwin, who gave selection pride of place (hence the association), but granted other forces (including “Lamarckism”) important, if lesser, roles in evolutionary change).
Passions ran high; I own Weismann's annotated copies of Spencer's articles, and his anger drips off the pages (Fig. 3-4). The two warriors thrusted and parried on both high and low roads, mixing some good arguments about the structure of evolutionary explanation with ad hominem charges of incompetence. Weismann (1893, p. 317) disparaged Spencer for being merely a philosopher, and not a true scientist: “ [I] can only explain Mr. Spencer's ignoring such cogent instances by supposing that, as a philosopher, he is unacquainted with the facts by personal observation, and that therefore they appear less weighty to him than to a naturalist; for I would not for a moment suppose that he purposely evades the difficulties which face his opinion, as is the manner of popular orators and advocates — and alas! even of some scientists.” Spencer, in his touché (1893, p. 23), replied, not entirely without justice as we shall see, that Weismann had hidden poor arguments under the cloak of authority as a practicing scientist: “Now it is doubtless true that as a naturalist he may claim for his 'opinion' a relatively great weight. Still, in pursuance of the method of science, it seems to me that something more than an opinion is required as the basis for a far-reaching theory.” [Page 199]
This chapter explores the hints, inklings, and tentative formulations of hierarchical selection theory developed and published during Darwin's century. As a primary goal, I wish to bring to light the buried and forgotten discovery made by all strict Darwinians: that they could not carry through the logic of organismal selection to render all evolution without a crucial assist from selection at other levels. Hierarchy theory only became a major and explicit theme in evolutionary thought during the late 20th century, but a secret of history reveals that none of the great thinkers who struggled, with uncompromising respect for logic, to establish a general theory based on organismal selection alone could ever make the argument work without an appeal, sometimes in frustration, to hierarchy.
Alfred Russel Wallace and August Weismann stand out as the two principal “neo-Darwinians” of the late 19th century, the men most strongly dedicated to the Allmacht of selection. They therefore become the test cases for my assertion that hierarchy cannot be avoided. I shall bypass Wallace, though he fits my claim that no pure selectionist could avoid hierarchy, because I find no
3-4. Weismann's personal copy (see his signature in upper right hand corner) of Spencer's reply to their first round of polemics. The two marginal comments on page 12 read (in translation of Weismann's German): “Impermissibly weak!” And (ironically): “As if that were certain!” (Author's collection.)
[Page 200]
evidence that he ever clearly conceptualized the issue of levels in selection. Wallace felt entirely comfortable with selection on all levels (see Kottler, 1985) and never seemed to grasp either the logic of Darwin's central commitment to the organismal level, or the problems involved in claiming that selection on other units (particularly higher “individuals”) could be effective in the face of strong selection at the organismal level. Wallace maint
ained such an unshakable and primary commitment to the ubiquity of good design that he unhesitatingly invoked higher levels to preserve an argument for active selection whenever a focus on organisms raised the specter of nonadaptation (notably in his uncritical advocacy of species selection for sterility in interspecific crosses, rather than accepting Darwin's argument for infertility as a side consequence of accumulated differences in two diverging populations — see pp. 131–132).
But August Weismann represents the ideal test case for my assertion. Once he had declared war on “Lamarckian” inheritance, Weismann dedicated his professional life to promoting the Allmacht of selection. He grasped the logic of Darwin's argument in all its details and extensions. He recognized the centrality of selection on organisms, and he struggled to make Darwin's single-level theory work for all phenomena of evolution. His famous 1893 paper on the Allmacht of selection presents, as its central theme, an explicit defense for exclusivity of the organismal level — or “personal selection” in his terms.* Later, and largely in response to strong arguments made by Spencer, Weismann admitted that he could not rely on personal selection alone. He could continue to promote Allmacht only by recognizing another level of “germinal selection” for subcellular components of the germplasm.
Moreover, Weismann gradually extended the theory of germinal selection, from an ad hoc aid for personal selection (in the original formulation of 1895 and 1896) to a fully articulated theory of hierarchy replete with notions of independence and conflict between levels (1904 version). Finally, Weismann came to regard hierarchical selection as the linchpin and completion of his entire theory (see pp. 221–224) — though we have forgotten his cogent arguments, and usually depict him as the champion of conventional, organismic selection. Weismann's intellectual journey, his relentless probings and frequent reformulations, leading finally (and perhaps inexorably) to a full theory of hierarchy, provide an object lesson in the logic of evolutionary argument, [Page 201] and the needs imposed by completeness and coherence once we abandon the myopia of regarding “organized adaptive complexity” (Dawkins, 1986) as the only focus for evolutionary explanation (with all else arising by extrapolation therefrom).
The Structure of Evolutionary Theory Page 33