Scientists’ Bookshelf

Darwin Tried and True Robert J. Richards

WHAT DARWIN GOT WRONG. Jerry Fodor and Massimo Piattelli-Palmarini. xxii + 264 pp. Farrar, Straus and Giroux, 2010. $26.

n November of 1859, when Charles ality on the part of nature; he preferred Herbert Darwin excitedly held in his hands the first Spencer’s phrase survival of the fittest. I published copy of On the Origin of Species, In the 1930s and 1940s, with the synthesis of virtually no reputable naturalist in England, modern genetics and Darwinian theory—pre- on the Continent or in America accepted the pared by such leading neo-Darwinists as Ernst proposition that species had altered over Mayr, Sewall Wright, Theodosius Dobzhansky, vast periods of time. The conception itself R. A. Fisher, Julian Huxley and J. B. S. Hal- was not new: Jean-Baptiste de Lamarck had dane—the device of natural selection was ce- advanced it at the beginning of the 19th mented into the foundation for most subsequent century, as had Darwin’s own grandfather work in evolutionary biology. Nonetheless, sev- Erasmus Darwin at the end of the 18th. Even eral of the 19th-century objections to natural Immanuel Kant entertained the hypothesis, selection have been reinvented and advanced although he rejected it for lack of evidence. by Jerry Fodor, a distinguished philosopher of By the time Darwin died in 1882, however, mind, and Massimo Piattelli-Palmarini, a cog- hardly a naturalist of standing could be found nitive psychologist, in their book What Darwin who denied the transmutation of species, so Got Wrong. In the view of the authors, Darwin utterly convincing was the argument of his and the neo-Darwinians got almost everything great book. wrong, because they founded their theories on In the Origin, Darwin gathered evidence of the vacuous concept of natural selection. various sorts: from biogeography, from embry- Fodor and Piattelli-Palmarini divide their po- ology, from systematics, from anatomy, from lemic into three parts. In the first, they maintain behavior and especially from the experience of that most current developments in evolutionary breeders in transforming their stocks. Virtually theory show that two conjoined neo-Darwinian all of these areas of evidence were already well propositions are invalid: that phenotypic traits known to naturalists. Darwin, however, wove are the simple products of selection by an ex- these strands together into what he called “one ternal environment, and that such selection long argument,” the force of which depended operates without any internal constraints on on his causal device of natural selection. With- variability. In the second part, they argue that out natural selection as its unifying theme, his the idea of Darwinian selection smuggles into argument would have unraveled into its hetero­ the operations of nature the assumption of in- geneous threads. tentional discrimination, based as natural selec- Darwin’s triumph in the 19th century oc- tion is on the model of the breeder’s selecting curred in the face of historical irony. Although for certain traits in organisms. In the third part, naturalists came to agree that species altered they suggest that accounts of species descent over time, many had objections to the cause ought to be regarded not as scientific but as his- Darwin posited. The astronomer and philos- torical explanations. The belief that underlies opher John Herschel called natural selection all three sections is that “Darwin’s theory of “the law of the higgledy-piggledy,” as if chance natural selection is fatally flawed.” The authors could shape the design of organisms. The biolo- have created a three-headed Hydra, with each gist St. George Jackson Mivart maintained that head distractedly chewing off the other two. natural selection was only an extrinsic, nega- The first part of their argument seems to tive factor and that descent required a positive, be principally the work of Piattelli-Palmarini, intrinsic principle. Even Alfred Russel Wallace, who has some training in biology, whereas the cofounder of the theory of evolution, thought second is most likely the creation of Fodor, that Darwin’s term natural selection was too an- who has mounted comparable conceptual ar- thropomorphic, suggesting as it did intention- guments against natural selection in a couple

238 American Scientist, Volume 98 © 2009 Sigma Xi, The Scientific Research Society. Reproduction with permission only. Contact [email protected]. of recent essays. In the first part, the strategy is Also Reviewed in This Issue to survey a fair amount of work in evolution- ary biology and then to show that this work demonstrates the operation of internal con- 242 A NUCLEAR WINTER’S TALE: Science and Politics in the 1980s. straints on the range of genetic and phenotypic By Lawrence Badash. Reviewed by Sean L. Malloy. In this intri-­ variability, thus undermining the assumption, cately detailed history, Badash examines both the scientific origins attributed to the neo-Darwinians, that nature of the concept of nuclear winter and the debate over its relevance to exogenously selects from a completely filled nuclear policy and the Cold War arms race space of morphological possibilities. For example, some species of the plant ge- 246 SCIENCE AS A CONTACT SPORT: Inside the Battle to Save nus Achillea display “norms of reaction” that Earth’s Climate. By Stephen H. Schneider • STORMS OF MY are not simply monotonically related to gradu- GRANDCHILDREN: The Truth about the Coming Climate ated changes in the environment: Clones of ­Catastrophe and Our Last Chance to Save Humanity. By some varieties of the plant are tall at the foot of James Hansen • CLIMATE COVER-UP: The Crusade to Deny mountains and also at higher levels but shorter ­Global Warming. By James Hoggan with Richard Littlemore. Re- at intermediate levels—thus expressing endog- viewed by Dimitri Zenghelis. Schneider, Hansen and Hoggan all enous constraints on phenotypic expression. say that their efforts to communicate the urgency of the problem of The authors bring to the bar two articles (one climate change to policy makers and the general public have met with published in 1981, the other in 1993) discussing powerful resistance coordinated by organizations seemingly bent on this phenomenon, but they seem completely sowing confusion. The public is indeed confused—and the conse-­ unaware that the classic studies of Achillea these quences, Zenghelis warns, could be tragic articles cite were actually conducted more than half a century ago by researchers who fully 250 THE AGE OF WONDER: How the Romantic Generation Dis- embraced natural selection and the emerging covered the Beauty and Terror of Science. By Richard Holmes. neo-Darwinian synthesis (Jens Clausen, David ­Reviewed by Frederick Gregory. This is a wonderfully engaging D. Keck and William M. Hiesey, Experimental narrative, says Gregory; Holmes’s biographical storytelling shows the Studies on the Nature of Species: III. Environmental impact of science on the hearts as well as the minds of Humphry Davy, Responses of Climatic Races of Achillea [Carnegie Joseph Banks, Mungo Park, and William and Caroline Herschel Institution of Washington, 1948]). This is a typi- cal example of the authors’ misunderstanding 252 THE EERIE SILENCE: Renewing Our Search for Alien Intelligence. of the history of evolutionary biology. By Paul Davies. Reviewed by Gregory P. Laughlin. We’ve been A more egregious case occurs when they fa- searching for extraterrestrial intelligence for 50 years now but haven’t vorably quote Ernst Mayr as protesting against found it. Could it be the case that civilizations inevitably destroy them-­ what he has referred to as “beanbag genet- selves before they evolve to the point of being able to colonize the galaxy? ics”—the view that genes singly determine phenotypic traits. Mayr, one of the architects 254 FIRST PEOPLES IN A NEW WORLD: Colonizing Ice Age America. of neo-Darwinism, is thus called on to oppose By David J. Meltzer. Reviewed by Douglas K. Charles. When did neo-Darwinism. Fodor and Piattelli-Palmarini the first humans arrive in the Americas? What route—or routes—did quote at length from a 1999 study by Geoffrey they take? Was there more than one wave of settlers? What did they B. West, James H. Brown and Brian J. Enquist do when they got here? Meltzer explains why archaeology has not yet (Science, 284:1677–1679) concerning “invari- been able to provide very precise answers to such questions ant scaling laws,” which constrain phenotypic expression quite independently of natural se- 255 A BRILLIANT DARKNESS: The Extraordinary Life and Mysteri- lection; they apparently do not realize that Ju- ous Disappearance of Ettore Majorana, the Troubled Genius of lian Huxley was a pioneer of this kind of allo­ the Nuclear Age. By João Magueijo. Reviewed by David Kaiser. metric study in his Problems of Relative Growth In addition to examining theories about what became of Majorana (1932) and that he then integrated these very when he disappeared permanently in 1938, Magueijo explains why concepts of allometric constraint into his Evo-­ Majorana’s ideas about the neutrino are once again on the minds of lution: the Modern Synthesis (1944), a classic physicists, now that evidence has begun to accumulate that the elusive work of neo-Darwinism. neutrino has some tiny but nonzero mass My point is that none of the major contribu- tors to the neo-Darwinian synthesis subscribed 256 ISAAC NEWTON ON MATHEMATICAL CERTAINTY AND METH- to the assumption against which Fodor and OD. By Niccolò Guicciardini. Reviewed by Paolo Mancosu. Piattelli-Palmarini pit more recent studies. Guicciardini reconstructs Newton’s mathematics and his concep-­ When Mayr threw the beanbag at popula- tion of mathematical methodology. Convinced that only geometrical tion geneticists like Haldane, the latter agreed proofs can be considered certain, Newton favored using geometrical that mathematical genetics made unrealistic, techniques to inject certainty into natural philosophy; he was an anti-­ simplifying assumptions in order to make modernist in that he considered the techniques of the ancients to be a computation more tractable; he did not deny paradigm of correct methodology that actual organisms have all sorts of internal constraints. Even Darwin himself recognized that traits are often linked—that is, internally constrained; one of his “laws of variation” www.americanscientist.org © 2009 Sigma Xi, The Scientific Research Society. Reproduction 2010 May–June 239 with permission only. Contact [email protected]. in the Origin is precisely that of “cor- reciting studies about the endogenous tive values as an attractor and hin- relation of growth,” of which he gave mechanisms that help to explain evo- drance in the presence of predators.) many examples. lutionary change and branching, they But to perform this discrimination, the The logic of this first part of the book preface their account with this warn- authors maintain, nature would have chews right through the evidentiary ing: “In fact, we don’t know very well to be sensitive to the counterfactual support that the studies cited might how evolution works. Nor did Dar- situation: namely that, for instance, appear to offer. Consider the article win, and nor (as far as we can tell) if hearts were to pump blood silently by West and colleagues, which Fodor does anybody else.” This remark is they would nonetheless have fitness and Piatelli-Palmarini obviously think perhaps more indicative of a supercil- value in the various environments in is decidedly probative. The very first ious attitude than of a considered epis- which vertebrates find themselves. line of that article reads, “Evolution temology. If taken literally, it would “So, thinking of selection as an inten- by natural selection is one of the few mean that a great deal of time has tional process is one way to bring into universal principles in biology.” Virtu- been wasted in rehearsing studies by play the counterfactuals that we need ally every recent study that Fodor and scientists whom the authors believe to make the distinctions that we need Piattelli-Palmarini cite likewise has to be ignorant about the very subjects in order to individuate phenotypic natural selection as the fundamental for which they’ve been cited. The first traits.” But, again, natural selection principle of the science—the principle head of the authors’ argument lies not doesn’t have intentions. that, in the second part of their book, merely severed, but chewed into indi- The authors helpfully suggest that they claim to be utterly muddled. In gestible gristle. maybe there are natural laws of selec- other words, they draw their evidence In the second part of the book, the tion. Laws certainly can support coun- from scientists whom they believe to authors argue that Darwin, along terfactuals. I am entitled to claim that be confused about the fundamental with virtually every other evolution- should I attempt to light my cigarette mechanisms of evolution. (This is no ary biologist, has been misled by the in the absence of oxygen, it would not less true of the concept of “constraint supposed analogy between artificial light. That claim is supported by the on,” a matter taken up below.) selection and natural selection. When law that oxygen is necessary for com- Moreover, the authors are frequent- breeders select for traits, say finer wool bustion. But laws aspire to generality, ly able to explain the work they cite in lambs or more desirable marbling in whereas the adaptiveness of pheno- only by using the concept of natural cattle—or, as in Darwin’s own breed- typic traits is context-dependent, say selection. Hence, to explain the stud- ing efforts with pigeons, the color pat- Fodor and Piattelli-Palmarini. It may ies that they believe undermine neo- terning of the ancestor—they intend to be the case that in a given environment ­Darwinism, they must use a concept produce a certain trait. But nature, as large creatures have the advantage, at the heart of neo-Darwinism that the authors constantly remind us, has but “it simply isn’t true, for example, they suppose to be without any sci- no intentions. Yet, in the natural situ- that being big is in general better for entific merit. Or do they? They write, ation, as they point out, a given trait fitness than being small.” Because en- rather offhandedly: “Natural selection that is supposedly selected for might vironments and conditions are always is real, of course (when properly con- have other features not assumed to in flux, no trait will inevitably win in strued).” Yet in the second part of the be selected for but which come along competition with other traits: Some- book they attempt to demonstrate that as “free riders.” To use their example, times bigness is an advantage for a “selectionism cannot be true” and is when the neo-Darwinist maintains creature, sometimes smallness. Hence thus not construable at all. Natural se- that hearts were selected for pump- there are no laws of necessary connec- lection, then, seems to be real precisely ing blood, he ignores that they also tion that govern filtering by natural in the sense that it is not. have the concomitant feature of mak- selection; hence counterfactuals can- The authors finally make a confes- ing pumping noises. Thus to say that not be supported; hence “selection sion: “It’s only fair to acknowledge blood pumping was the trait selected for” cannot be the causal process by that the majority of biologists whom for ascribes to nature an intentional which species are altered over time. we have cited here, including several act that discriminates between the two In reading through all this, I was of the discoverers of these quite intri- correlated traits. The human breed- reminded of John Dewey, who began cate levels of endogenous regulation, er might intentionally select for one his philosophic career as a Hegelian still today endorse natural selection as of two such linked traits, but nature but said he finally came to realize that the determinant par excellence of the cannot. “Selecting for” necessarily as- a system of thought can be internally course of evolution.” “Fair” is hardly sumes intentional power, which only coherent and still be crazy. What Dar-­ the term I’d apply to this admission. human nature has. win Got Wrong, at least across the three Fodor and Piattelli-Palmarini are con- Fodor and Piattellini-Palmarini al- parts, doesn’t even have the virtue of ceding that what they take as evidence low that the neo-Darwinist might as- being consistent. If “selection for” at- against neo-Darwinism is regarded by sume that the causal matrix formed tributes to nature an intentionality that the scientists they cite as simply the by the environment might yet filter it cannot have, then “constraint on,” normal development of evolutionary through one of the coextensive traits the favored conception of the first part theory within the boundaries of neo- that has superior fitness value while of the book, must also operate under Darwinism. simply allowing the free rider with the tainted assumption. What part of the work of these im- no or even negative value also to pass To constrain expression of either ge- portant scientists would the authors through. (The fitness value of the pea- netic or phenotypic variability means have us believe? Apparently none of cock’s tail in the presence of females that, contrary to fact, if the constraint it. Although they spend half the book might outweigh its concomitant nega- were removed, expression would have

240 American Scientist, Volume 98 © 2009 Sigma Xi, The Scientific Research Society. Reproduction with permission only. Contact [email protected]. a different pattern. But, as the authors metaphysical necessity,” but they do in the narrative had not occurred, insist, only intentional systems can express well-founded generalizations neither would the event of interest, be sensitive to contrary-to-fact condi- (such as that vertebrates need oxygen at least not in the form that it did. If tions. Hence “constraint on” cannot be to survive) that can indeed support the historian could not defend such ascribed to nature as a property that counterfactuals. The second head of counterfactuals, then it would be im- helps explain transmutation—even the authors’ polemic hangs upside possible to assess his or her narrative though such ascription forms the con- down by a thread. as “reasonable” or “plausible.” clusion of the first part of their book. In the third part of the book, the Historical accounts depend on evi- The concepts “selection for” and authors suggest that if evolutionary dence and generalizations derived “constraint on” do indeed have inten- accounts of the transmutation of spe- from observation, and evolutionary tional properties, because they express cies cannot be scientific—because explanations have the added advan- judgments made not by nature but by their supposed causal explanations tage of experiment. These well-honed human beings—intentional judgments are based on a vacuous concept—such techniques of inquiry form the basis by biologists to the effect that in partic- accounts may be better regarded as for the kinds of discriminations that ular environmental situations certain historical narratives—that is, chains of Fodor and Piattelli-Palmarini both features of traits are causally relevant. “causally sufficient conditions,” which deny and grant the laborers in these Quite routinely, for example, medi- are invoked ex post facto, to explain, fields. The authors thus orchestrate a cal experts attribute the evolution of for example, “why the dinosaurs be- medley of contradictions that can de- drug-resistant strains of bacteria to the came extinct.” What else do accounts light only the ears of creationists and excessive use of antibiotics in hospi- of natural selection purport to be but proponents of intelligent design. tals—or in cattle feedlots. No hospital causally sufficient explanations of the In the legendary meeting of the Brit- workers or cattlemen intend to select evolution—or extinction—of species ish Association in 1860, Bishop Samuel for drug-resistant bacteria, although via adaptations or the failure thereof? Wilberforce attacked the Darwinian de- their actual intentions obviously play a There goes the third head. fender Thomas Henry Huxley with this causal role. Scientists understand quite The authors, in a denigrating mode, barb: Did Mr. Huxley claim his descent well how selection operates in these in- claim that a historical account cannot from a monkey through his grandmoth- stances; indeed, they are able to breed be supported by counterfactuals, as er’s or his grandfather’s side? Huxley drug-resistant bacteria experimentally if evidence and generalizations were reputedly whispered to a friend: “The precisely in the way these organisms unknown to the historian. History Lord has delivered him into my hands.” are selected for in the “wild,” thereby and thus evolution are both, they say, Huxley retorted that he would rather confirming the natural selection of “just one damned thing after another.” have a monkey as his ancestor than be drug resistance. Yet they concede that many historical connected with a man who used great Had Fodor and Piattelli-Palmarini narratives, that is, causally sufficient gifts to obscure the truth. read the first chapter of the Origin, accounts, are “reasonable” and “plau- they would have seen that Darwin sible.” Unless this is an utterly empty argues there not so much that artifi- concession, they must allow what the Robert J. Richards is Morris Fishbein Professor of the cial selection is a model for natural historian takes for granted: namely, that History of Science and Medicine; professor of history, philosophy and psychology; and director of the Fish-­ selection as that it is exactly the same he or she, on the basis of evidence and bein Center for the History of Science and Medicine thing. Darwin regarded the breeder’s supported generalizations, can uphold at the University of Chicago. His most recent book is intention, correctly I believe, as simply the relevant counterfactuals—counter- The Tragic Sense of Life: Ernst Haeckel and the another environmental condition— factuals to the effect that if the signifi- Struggle over Evolutionary Thought (University one that rarely has a predictable out- cant antecedent conditions mentioned of Chicago Press, 2008). come, as he discovered when he tried to breed fancy pigeons back to their original ancestral colors. Darwin thus POLICY directly demonstrated natural selec- tion at work. And we do the same in the case of drug resistance. Weathering Nuclear War There are, of course, many instances in which the biologist can only hazard Sean L. Malloy a guess as to what is being selected for; but at other times, experiments A NUCLEAR WINTER’S TALE: Science and Politics in the 1980s. Lawrence Badash. xvi and systematic observation can sift out + 403 pp. The MIT Press, 2009. $40. the likely properties being selected. In the well-known case of sickle-cell ane- Nuclear Winter’s Tale, by Lawrence even one interested in nuclear weapons mia, researchers are fairly confident Badash, is an intricately detailed or the history of science, is likely to pick that the property of malarial resis- A history of a subject that, by the up. Scientists and historians of science, tance was selected for, although obvi- author’s own admission, never managed meanwhile, will likely be wary of the ously not the free-riding trait of poor to capture the sustained attention of the amount of space that Badash devotes oxygen transport when the alleles are media, the public or policy makers— to the bureaucratic and organizational homozygous. These connections be- even at the height of its visibility in the wrangling over the subject of nuclear tween traits and conditions may not early to mid-1980s. Thus it is not the sort winter. Any book that contains chap­ have the canonical status of “laws of of book the average educated reader, ters titled “Report after Report” and

242 American Scientist, Volume 98 © 2009 Sigma Xi, The Scientific Research Society. Reproduction with permission only. Contact [email protected].