In lieu of an abstract, here is a brief excerpt of the content:

1 1 EVOLUTIONARY BIOLOGY AS A HISTORICAL SCIENCE When we regard every production of nature as one which has had a long history; when we contemplate every complex structure and instinct as the summing up of many contrivances, each useful to the possessor, in the same way as any great mechanical invention is the summing up of the labor, the experience, the reason, and even the blunders of numerous workmen; when we thus view each organic being, how far more interesting . . . does the study of natural history become! CHARLES DARWIN, ON THE ORIGIN OF SPECIES, 1859, PP. 485–486 One of the great goals of modern science is to understand biological diversity: where it comes from, how it evolves, and what maintains it. It has fallen to the field of evolutionary biology to try to answer these questions. In attempting to do so, evolutionary biology does not fit the everyday view of science in which hypotheses are put forward and subjected to experimental test. The reason is obvious. The scale in space and time is simply too large. It would be wonderful to be able to do an experiment on, for example, the role of interspecific competition as a driving force in evolutionary diversification. Just get an island archipelago, seed it with an ancestral finch population, and let nature take its course. Then get another archipelago, seed with the same type of finches, but add an overabundance of resources so that resources are not limiting and competition does not occur. Replicate the treatments a few times (say, four archipelagoes flush with resources, four without), come back in several million years, and, voilá, the hypothesis has been tested. Too bad we can’t do this. In trying to understand how and why evolutionary diversification has occurred, we’re stuck with studying a phenomenon that has occurred over large spatial scales through the course of thousands to millions of years. For this reason, evolutionary biology is more akin to a social science—history—than it is to laboratory based sciences like chemistry (Cleland, 2002; Mayr, 2004) Lacking time machines, both losos_ch01.qxd 4/11/09 8:43 AM Page 1 historians and evolutionary biologists must draw inferences from a variety of different sources and approaches in their attempts to understand the past2,3 . I like to compare studying evolutionary diversification to a detective story:4 something happened in the past, and it is our job to build the best case to explain whodunit (or, at least, whathappenedtoit). In doing so, there usually is no smoking gun, no decisive experiment or single piece of evidence (Turner, 2005). Rather, we must gather as much data, from as many different sources, as possible. Then we must weave together these data to present the best explanation of what happened.5 As in a court case, the more consistent and corroborative the data, the more compelling the case (for a generally congruent, but slightly different, view, see Cleland [2002]; also see Pigliucci [2006]). Such explanations, of course, are more than mere stories; they are the hypotheses that guide further work. Each time we learn something new, each time we bolster our case a bit more, new hypotheses are suggested that await subsequent testing. The better supported an explanation is, the less likely it will be that a single new piece of data will discredit it. Nonetheless, given that we are trying to explain what happened in the past, we can never know for sure what happened,6 and it is always possible that additional data will change our thinking. SYNTHESIZING DATA FROM THE PRESENT AND FROM THE PAST Ideally, we would like to know what processes occurred in the past and how these processes shaped the diversity we see today. This is where building the best detective case comes in. We can’t directly study the processes operating in the past (Cracraft, 1981). But we can study processes in the present, and we can even observe their outcome over short evolutionary timescales. What we can study in the past is the pattern: the history of change through time. Depending on the quality of the historical record, we can infer, with a greater or lesser degree of confidence, what happened. The key, then, is to extrapolate from our understanding of the relatively short term outcomes of ongoing processes to explain the patterns of change in the past. 2 • E V O L U T I O N A R Y B I O L O G Y A...

Share