Saturday, May 02, 2009

Carroll on Gould

Anyone who has read much of this blog knows where my sympathies lie in the "Science Wars." I don't hate Richard Dawkins, but I find a few of his hobby horses to be annoying and/or dangerously misleading. And I pretty much despise the whole sociobiological project to colonize the social sciences and the humanities, be it by means of memes or by means of consillance, the whole project is something I admit is theoretically possible--like it is possible that human development was crucially influence by alien intervention--but almost always silly in practice. Why? Because the starting off point is always taking as granted a whole bunch of far from settled science on human development.

Recently, I've been commenting on a (fairly good) review of Denis Dutton's The Art Instinct over at Cognition and Culture and I've gotten into something of a dust up over (red flag!) Stephen Gould with Joseph Carroll, who has been a leading figure in the Literary Darwinism movement along with the biologist David Sloan Wilson and the Englsih PhD Jonathan Gottschall.

My own take is pretty much what it was--what is good in Dutton's book could have been written without the constant reference to our Pleistocene ancestors, and what requires them is highly doubtful.

Gould's name got dragged into this because Dutton rather dully and unneccessarily attacks him in his book. Gould the horrible, oppressive monster gets a lot of play in certain circles. Everyone else has opinions, Gould issues fiats from an undisclosed location to end free speech and scientific inquiry as we know it. Lucky for us dozens upon dozens of heroes are willing to stand up to this threat, disagree with Gould and tiresomely crow about how daring they are. Sometimes I wonder if Gould has anything to do with the Worldwide Jewish Conspiracy.

Anyhow, Gould's name arose, and immediately became the focus of dispute.

Hmm, let's see, theorists like Maynard Smith, E.O. Wilson, Conway Morris, and David Hull are won't to sling "bullshit." And critics who cite their opinion of Gould are likely to be "hotheads." My irony bone tingles.Panksepp and Panksepp are good and serious critics of orthodox or "narrow-school" evolutionary psychology. Other critics who fall into that category would include Kim Sterelny, Richerson and Boyd, David Sloan Wilson, Steven Mithen, Paul Griffiths, and Nicholas Wade. All of these commentators, like the critics of Gould previously mentioned, are different from Gould in that they are not intellectual charlatans. They are honest, straightforward thinkers. They aim to make sense, not to create confusion.
The Panksepps are honest, straightforward thinkers who cite Gould with approval and repeat many of his criticisms of Evolutionary Psychology. 

The confusion that many in the field of Evolutionary Psychology felt was, as far as I can see, mere denial. They just did not want to hear that they were often merely speculating, that the background knowledge that would make their endeavors useful didn't exist yet. That they were, as far as science goes, wasting their time.

[There follows in Carroll a long hostile gloss of two New York Review of Books pieces Gould wrote attacking Sociobiology and Evolutionary Psychology, but I don't really see the point of the gloss or what precisely is being advanced by Carroll through it.]

The two ideas for which Gould has generated the most publicity are “punctuated equilibrium” and “spandrels.”

Gould’s one other big idea is that of “spandrels” or non-adaptive structures. “The Spandrels of San Marco and the Panglossian Paradigm: A Critique of the Adaptationist Programme” is probably Gould’s best-known essay. . . .

The elements in these two ideas that are substantive and valid were integral parts of Darwinism before Gould formulated them, associated them spuriously with anti-adaptationist intimations, and popularized them with catchy phrases. Gould’s own distinctive contribution to these two concepts, insofar as they have consisted of ideas that were substantive and that were not already part of the Darwinian synthesis, have proven to be either compatible with mainstream adaptationist thinking, relatively unimportant, or simply wrong. . . . 

Gould's complaint was not generally about what pan-adaptionists were willing to concede was true, but rather what their practice reflected. His argument was that though EP may have said that spandrels and genetic drift and other change agents were possibilities, the possibility that they seemed to heavily favor in nearly all cases was adaptation.

In making spandrels into a biological metaphor, Gould blends two legitimate Darwinian concepts, but he spuriously represents this blended concept as an alternative or supplement to the idea of adaptation through natural selection. One of these legitimate Darwinian concepts is pleiotropy or multiple genic effects: what Darwin calls correlated growth. The other legitimate Darwinian concept is the idea that previously existing structures can be altered through natural selection to fulfill adaptive functions. Darwin offers as an example the swim bladder that in the course of evolution is transformed into a lung (2003, chap. 6, p. 214). The tetrapod body plan also caught Darwin’s attention (pp. 219-220) and has remained a favorite example among evolutionists. The forelimbs evolve from fins to legs, and from legs sometimes to wings and sometimes to flippers. Another favorite example, discovered after Darwin’s time, is that of the reptilian jaw bones that have been transformed into the mammalian ossicles--the bones of the inner ear. (See Young, 1992, pp. 185-186; Moore, 1993, pp. 176-177, 412-414.) For adaptations that use either previous adaptive structures or previous structures of no adaptive value, Gould and Vrba (1982) have invented the term “exaptation.” This term is a variant of a term that was previously current—“preadaptation”--and the concept is itself a commonplace in standard Darwinian theory. . . . 

Despite the confusions and ambiguities introduced through the architectural metaphor, none of the implications in the idea of spandrels is in any way contrary to standard adaptationist thinking. What Gould and Lewontin have attempted to do, though, is to use the metaphor to suggest, without quite saying it, that major features of complex functional structures have been produced independently of adaptive processes. Put this baldly, the claim is simply and obviously false, but unless it is put this way, the claim has no actual content that is not already part and parcel of standard Darwinian thinking. Since the time of his youthful foray into saltation, Gould himself has usually been careful, whenever he implies or suggests this false idea, also to say that he recognizes that complex functional designs result from adaptation, or that adaptation through natural selection is an “important” feature of the evolutionary process. The false and obfuscatory implications in the more radical understanding of “spandrels” are nonetheless its raison d’ĂȘtre, its chief purpose and function. It subserves the larger Gouldian program of minimizing in whatever way he can the general significance of adaptation through natural selection. 
In order to achieve their aim of minimizing the significance of adaptation through natural selection without clearly and decisively cutting themselves off from mainstream Darwinism, Gould and Lewontin are driven to the necessity of perpetual equivocation, and the equivocation is rendered all the more impenetrable by being commingled with a pseudo-concept produced by breaking a single, valid concept into two parts and representing these parts as antithetical. The single, valid concept is that of “selection,” and the two parts are “selective force” and “constraints.” We shall begin with the equivocation and then consider the pseudo-concept. Once again spuriously invoking Darwin as an antecedent for their own anti-adaptationism, Gould and Lewontin repudiate the idea that Darwin was himself “a radical selectionist at heart who invoked other mechanisms only in retreat, and only as a result of his age’s own lamented ignorance about the mechanisms of heredity” (1979, p. 589). “This view,” they declare, “is false.” But then they also declare, in the very next sentence, that “Darwin regarded selection as the most important of evolutionary mechanisms. As do we.” As do we. Strange, then, that the whole thrust of their essay should be toward the conclusion that “constraints restrict possible paths and modes of change so strongly that the constraints themselves become much the most interesting aspect of evolution” (p. 594). Or as they explain in the head note to the essay, “the constraints themselves become more interesting in delimiting pathways of change than the selective force that may mediate change when it occurs” (p. 581). Selection is the most important mechanism, but despite its importance, it is still not very interesting, somehow, not nearly so interesting as other things that are not so important.
The idea of a selective force operating independently of constraints--the idea of selection operating in a vacuum, independently of all actually existing conditions--is something like the idea of one hand clapping. When the idea of selection is placed in antithesis to the idea of constraints, it ceases altogether to be an intelligible idea. It becomes a pseudo-concept, a rhetorical term that is devoid of any conceptual content other than the confusion caused by the faulty way in which it is formulated. One might suppose that this feature of the concept--its lack of any content other than the confusion generated by the way it is formulated--would help to explain why it is so uninteresting, but it could hardly also explain why it is still “important.” Gould and Lewontin have here drifted into a very strange region of “thought,” a region much more familiar within the confines of postmodern literary theory than within those of evolutionary biology. Like Derrida or Foucault, Gould and Lewontin bring to bear sophisticated analytic and rhetorical skills, but these skills are oriented not to the production of clear and distinct ideas but to exactly the opposite, to the construction of pseudo-concepts that obstruct clear thinking.. . . . 

In his eagerness to minimize the significance of adaptation through natural selection, Gould is, in wish and emphasis, anti-Darwinian. But since, within the range of scientifically reputable evolutionary theory, there is no actual alternative to Darwinism--no alternative, that is, to adaptation through natural selection as an explanation for complex functional structure--Gould can never say fully what he wants to say. His plight recalls that of “Atticus” in Alexander Pope’s “An Epistle from Mr. Pope to Dr. Arbuthnot.” In Pope’s depiction, Atticus (Addison) wished to satisfy envy and spite without making himself vulnerable through open attack. He thus developed a proto-Gouldian rhetorical technique that enabled him to “Damn with faint praise, assent with civil leer, / And without sneering, teach the rest to sneer; / Willing to wound, and yet afraid to strike, / Just hint a fault, and hesitate dislike” (1969, ll. 201-204). . . . 

Among Darwin’s contemporaries, the one figure who most resembles Gould in his use of sophistical equivocation is the paleontologist Richard Owen (1804-92), who wished, on the one hand, to affirm that animal forms are determined by “archetypes” that are not related to one another by lineage and, on the other, to represent himself as having originated proto-Darwinian evolutionary ideas. Darwin responds to Owen’s equivocations in the historical sketch appended to the third edition of the Origin, and he there comes closer there to a snort of satirical contempt than he ever comes in responding to any other writer, even to Lamarck. “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” (2003, p. 84). Darwin himself operates in good faith, and his overriding assumption is that others do also, even when he fundamentally disagrees with them. In his Autobiography, he remarks, AI have almost always been treated honestly by my reviewers, passing over those without scientific knowledge as not worthy of notice” (1958, p. 125). Coming from a man who had received so many violently hostile reviews, this remark reflects a presumption of good faith so ingenuous in its benignity as to fall little short of the sublime. But Owen is so flagrantly and unmistakably not operating in good faith that even Darwin’s simplicity of good will is finally roused to an awareness of Owen’s deviousness and duplicity. One can only speculate how Darwin would have responded to Gould. He might well have wondered whether Gould is, as Maynard Smith characterizes him, merely confused, or, as Dawkins characterizes him, downright dishonest. To my own eye, it seems evident that Gould is not himself confused, though it is his purpose that his readers should be.
I quote this at length because the text itself really puzzles me. Somehow Carroll has seen around a corner of the Spandrels argument that Maynard Smith, who thought the piece a solid addition to the conference he was hosting, did not see? 

What the hell is the Owen comparison about? (I can't find anything to really nail YOU on, but you remind me quite a bit of Joseph Stalin . . .)

While Gould may have been guilty of exaggerating the innovativeness of some of his thinking, he certainly never made this claim for spandrels. the innovation of the spandrels argument was the illustration--and he and Lewontin say so--and the purpose was to re-emphasize non-adaptive change in the face of a then recent heavy emphasis on it--also this is all in the text itself. Perhaps having actually read the spandrels paper may have alleviated Carroll of some of this seeming confusion.

And having read Derrida and Foucault, I find Carroll's lumping of Gould and Lewontin into the same sentence with them laughable indeed. Lewontin is usually a perfectly clear writer. Gould is fairly clear, though sometimes a bit overwrought, but with a bit of effort he's far from obscure. Unlike your own passage above, I'd say. While grammatically unchallenging, a lot of its argument escapes me, and what I can perceive is just wrong.

The pointless lecture on constraints might be better illustrated by a lengthy citation of works in which Richard Dawkins talks at length about constraints on selection. He acknowledges they exist, but this is not his emphasis. The disagreement between Gould and Dawkins et al is about EMPHASIS not about innovation, or wholly new ways or wholly separate ways to approach evolution. Anyone who has read very much in the literature should know that.

And if Carroll still has problems with the spandrels metaphor, it is explained in a series of letters in response to Orr's review of Pinker in the New York Review.