We owe to Charles Darwin and Alfred Russel Wallace not the discovery of evolution but the creation of a powerful causal explanation for the small and large facts about nature that fascinate us as children and as adults. The principle of natural selection provides at least part of an explanation for a myriad of traits ranging from the shape of an orchid blossom to the visual acuity of a hawk’s eye. Darwin’s 1859 presentation of the principle in On the Origin of Species is a synthesis of observation and explanation that should amaze any curious reader. Wallace’s (1870a) discovery of natural selection is no less amazing, especially given the more difficult circumstances that surrounded his work (see Berry 2002). Their discovery is a human achievement of the first order, which should transcend political and social divisions. But it is best to understand the world as an “is” and not as a “should.” Most people believe some sort of divine explanation for the kind of facts about nature that I just mentioned; if they are aware of Darwin’s and Wallace’s discovery, they dismiss it as false. Few, if any, of these people will read this essay. You, the reader, are likely aware of the truth and importance of Darwin’s and Wallace’s discovery or are open to their discovery being correct. Even so, in order to understand how the testing of evolution has evolved, it is useful to examine the nature of Darwin’s and Wallace’s achievement, especially because it is often misunderstood even by those who recognize that it is correct.

In this context, it is worth comparing the “before” and the “after” of their discovery. To understand the “before,” it is most meaningful to consider claims about nature that at least appear to be based on substantial assemblages of facts (as opposed to claims as to the “fact” of, say, divine creation that were simply appeals to faith and referenced few, if any, facts). With this restriction, all claims we consider are arguably science, inasmuch as there is marshaling of evidence (as opposed to just an appeal to belief). Given this restriction, one finds a wide variety of claims that range from the claim that the facts are explained by divine creation (Paley 1802) to those that invoke natural causes.

Summary

Our understanding of the evolution of sex ratios has advanced substantially in recent decades, in part due to the important work of Hamilton (1967) on ‘extraordinary sex ratios’. However, important aspects of the biology have largely remained unstudied, including the mating dynamics and structure of natural populations, the role of the individual in producing observed sex ratios, and the nature of sex ratio control. Also little studied is how tests of sex ratio models should be structured so as to provide maximum insight. Perhaps ignoring these facets of the biology has aided progress in the past, but these gaps in the study of sex ratios are now blindspots, which hinder understanding. Further progress in the evolutionary analysis of sex ratios requires their elimination. This can be accomplished only by direct investigations of mating dynamics, population structure and the behaviour of individuals.

Introduction

Every great scientific theory is a partial lie about nature. This is not a claim about scientific fraud, as the issue is not one of honesty in the usual sense. Instead, it is a claim about how theories succeed (Cartwright 1984). A theory can succeed by explaining facts correctly. It can also succeed by making facts appear irrelevant or unimportant, sometimes correctly but sometimes incorrectly. Such facts sometimes resurface, but sometimes they are never seen again (Lewontin 1991). To this extent, the lie arises because nature is not described, as much as it is created from some facts while others are eliminated and left unexplained.

Scientific controversies have a bad reputation these days. Instead of being viewed as an engine of scientific progress, they are seen by many scientists and others as clashes of unresolvable ideological differences in the way the world is perceived and understood. One of us long ago, as a know-it-all Harvard graduate student in conversation with other such graduate students, often referred to this or that controversy as an example of the “dialectic of polarized fanatics,” with the absolute but unspoken know-it-all understanding that there was little, if anything, useful taking place. More meaningfully perhaps, controversies are often said to generate “more heat than light” on a given scientific issue and, to this extent, to do little to advance understanding. To some observers, the debate over adaptationism in evolutionary biology has this character. From this perspective, the various views about the power of natural selection to shape phenotypes are so different that proponents of each talk past each other and often do not even acknowledge the potential validity of alternative views. Of course, one can still view such steadfastness as essential for the progress of science instead of as indicating an unproductive narrow-mindedness. One need not think that the debate over adaptationism is unresolvable or agree with either view as to the nature of ideological differences to appreciate the importance of refining our understanding of the meaning of adaptation and of optimality models.

In the Philosophical Investigations, Wittgenstein (1968, 94) describes a man who is unconvinced by a claim he reads in a newspaper, so he buys another copy of the same newspaper to double-check. The point of the joke is that it isn't just the quantity of evidence in support of a claim that matters, but the quantity of independent evidence (Sober 1989).

The issue of independent evidence is of central importance to hypothesis testing in evolutionary biology. Suppose you wanted to test the hypothesis that long fur is an adaptation to cold climate and short fur is an adaptation to warm climate. You look at 20 bear species; 10 live in a cold climate and have long fur, and 10 live in a warm climate and have short fur. Is there any reason to think that the data do not confirm the adaptive hypothesis? One worry is that the species in each group resemble each other merely because they inherited their fur length from a common ancestor of the group (and that the temperatures experienced by ancestors and descendants are similar). This influence of ancestor on descendant is often called phylogenetic inertia (e.g., see Harvey and Pagel 1991).

If phylogenetic inertia and natural selection are alternative explanations of the data, how are these competing hypotheses to be evaluated? The following quotation exemplifies a common idea about this problem: