The Ubiquity of Selection: Problems with David
Ford's Critique
Post of the Month: July 1998
by Loren King
After reviewing a selection of ambiguities and discontinuities in the fossil record, David Ford remarks that
"a look at the peer-reviewed literature reveals that the fossil record does not contain confirmation of the theory of natural selection's prediction, and in fact presents the opposite of what the theory predicts."
Ford thinks that the key "prediction" of natural selection is the gradual emergence of evolutionary novelties, phenotypic changes that allow an organism to perform a new function. This may be a prediction of evolutionary theory, but Ford's argument is critically undermined by serious confusions about what Darwin meant, and what scientists have been up to in the years since the Origin of Species first appeared.
At the outset I want to point out the sillines of Ford's claim that, since he has not been provided on talk.origins with citation after citation, therefore there are no peer-reviewed findings of evolutionary transitions in the fossil record. This is just foolish, equivalent to me denying the existence of the Aztec or Maya civilizations because no one on sci.archeology would give me references to relevant studies and findings; not doing one's own homework is hardly a convincing basis for argument. "Does the fossil record tell lies?" Ford asks, and responds: "Darwin believed so. I for one do not." This when he has already tipped his hat that -- aside from some selective reading of Ernst Mayr, Niles Eldredge, Steven Jay Gould and a couple of other writers -- Ford has not bothered to consult the extensive literatures on paleontology and natural history (let alone molecular genetics and evolutionary biology)!
But forgive Ford for not wanting to become conversant with such an enormous volume of literature! Instead, let's consider the merits of Ford's argument that Darwin's reasoning about natural selection is fatally flawed.
First, Ford argues that Darwin's analogy between artifical and natural selection is overdrawn: there is, Ford argues, observed limits to phenotypic variation within species. We observe this when we try to transform species through our own selection. Nature, Ford implies, is faced with similar deep constraints on phenotypic form. Second, Ford argues that, in the absence of intelligent selection, there is no "end product" being sought, and so Darwin's analogy between artificial and natural selection is critically overdrawn. Third, Ford suggests that the number of negative variations across generations ought to make evolutionary progress difficult if not impossible; on this point Ford invokes the oft-heard refrain that mutation alone is an inadequate basis for the "good" variations required for evolutionary progress.
But Ford's arguments betray a striking confusion about Darwin's central thesis, and a related ignorance of subsequent work in ecology, mathematical biology and molecular genetics.
First, there is no evidence for the deep genetic constraints on phenotypic form that Ford claims we observe in nature and the lab. This is perhaps the creationists' strongest claim, i.e. their theory of fixed kinds: although organisms may exhibit adaptive variation (thus allowing so-called "microevolution"), this variation is constrained within fixed kinds (thus making so-called "macroevolution" impossible). But creationists have never actually tested this controversial conjecture; they merely utter it as a suspicion, much as Ford does. And vague suspicions do not stand alone as valid scientific claims.
Second, the so-called "end product" of natural selection is in fact definable, i.e. differential success of heritable traits and/or behaviours. As George Williams puts it: "the essence of the genetical theory of natural selection is a statistical bias in the relative rates of survival of alternatives (genes, individuals, etc.)" (Adaptation and Natural Selection: a Critique of Some Current Evolutionary Thought [Princeton, 1966], p. 22). That's it. No overarching purposes are required. The process of selection is a function of heritability and differential success in reproduction. There is no need for intelligent selection.
And third, mutation is not the only, or even the most salient source of variation within organisms. Much work has been devoted of late to exploring possible sources of variation, and all of it is compatible with a Darwinian mechanism of natural selection. And the argument that most variations are harmful is simply confused. While it is true that most mutations are either irrelevant or detrimental to the fitness of an organism, it is not the case that mutation is the only source of variation upon which natural selection can act.
In short, Ford's argument is undermined by a poor grasp of Darwinian and post-Darwinian ideas, and an insufficient examination of empirical workthat has been done since Darwin's time.
Loren
-------------------------------------- Loren King lking@mit.edu http://web.mit.edu/lking/www/home.html
First posted 20 July 1998
Home Page | Browse | Search | Feedback | Links