CHOOSING SELECTION: The Revival of Natural Selection in Anglo-American Evolutionary Biology, 1930-1970

Article excerpt

1

Introduction

When did biologists accept Charles Darwin's theory of evolution? One might argue that his theory, in its original form, was never accepted by more than a handful of biologists. The theory that was accepted in the twentieth century was based on a different understanding of natural selection, of the variations on which it acts, and on the source of those variations. Nevertheless, by 1970, Darwin was given credit for proposing a theory that was essentially correct when the necessary corrections were made. In particular, biologists came to believe that natural selection acting on small mutations, combined with Mendelian principles of heredity, is the major cause of evolutionary adaptation. Additional factors, such as genetic drift, geographical isolation, polyploidy, and so forth, may be involved in some cases (especially in speciation) and at some levels, but they would accomplish little by themselves. As modified in the twentieth century, Darwin's theory, sometimes (misleadingly) called neo-Danvinism,1 is generally believed by historians of biology to have been established in the 1940s. It was the core of an "evolutionary synthesis," which brought together previously separate disciplines such as genetics, zoology, botany, and paleontology into a unified science of biology. At the same time, it could be considered a constriction rather than a synthesis, because it excluded many possible causes of evolution (Provine 1988). There is now a growing scholarly literature on the history of this synthesis.2

Why did biologists accept the modern version of Darwin's theory? I discuss in this book the reception by evolutionists and other biologists of what I will call the natural selection hypothesis (NSH): the hypothesis that natural selection, with an ample supply of variation in heritable characters, is not only the major process involved in evolution (with the help of geographical isolation or polyploidy in some cases), but also that Lamarckian effects, random genetic drift, and macromutations have essentially no evolutionary significance. The NSH was accepted by a bare majority of evolutionary biologists for only a short time (in the 1950s and 1960s). In particular I ask, what were the reasons they gave for adopting that hypothesis and for rejecting the argument that drift plays a significant role in evolution?

Although I have generally used the concepts and terminology of evolutionary biologists in the time period under discussion, these concepts and terms may be confusing to a modern reader. Thus, John Endler argues that natural selection should be considered a process rather than a cause. It does not "act" - rather, it is the result of heritable differences and occurs when variation among individuals arises due to mutation or other causes.3

This study is part of a program to compare the reasons why scientists in different fields accept (or in some cases reject) proposed new theories. Is it true that "the reception of neo-Darwinism was colored by prevailing empiricist currents of thought" (Depew and Weber 1985b, 240)? If so, what specific empirical evidence was most persuasive? What weight did scientists give to the confirmation of predictions, supposedly the essential desideratum of the hypothetico-deductive scientific method, and according to Karl Popper, the basis of the criterion for demarcating science from pseudoscience?4

That question is especially appropriate in the case of evolutionary biology, since Popper s own critique of Darwinism provoked a long-running debate among philosophers and biologists about whether biological theories should be expected to make testable predictions, and whether more generally, the methodological standards for biology should be different from those of the physical sciences. The creation of "philosophy of biology " - as distinct from traditional "philosophy of science," allegedly biased toward physics-resulted in part from this debate. …