A coherent treatment of the flow of ideas throughout Darwin's works, this volume presents a unified theoretical system that explains Darwin's investigations, evaluating the literature from a historical, scientific, and philosophical perspective.

The desire to minimize the number of individual new entities postulated is often referred to as quantitative parsimony. Its influence on the default hypotheses formulated by scientists seems undeniable. I argue that there is a wide class of cases for which the preference for quantitatively parsimonious hypotheses is demonstrably rational. The justification, in a nutshell, is that such hypotheses have greater explanatory power than less parsimonious alternatives. My analysis is restricted to a class of cases I shall refer to as (...) additive. Such cases involve the postulation of a collection of qualitatively identical individual objects which collectively explain some particular observed phenomenon. Especially clear examples of this sort occur in particle physics. 1 Introduction 2 Particle physics: a case study 3 Three kinds of simplicity 4 Explanatory power 5 Explanation and non-observation 6 Parsimony and scientific methodology 7 Conclusions. (shrink)

When Kangas suggested in 1986 that wildlife reserve designs could be much smaller than previously thought, community ecologists attacked his views on methodological grounds (island biogeographical theory is beset with uncertainties) and on conservation grounds (Kangas seemed to encourage deforestation and extinction). Kangas' defenders, like Simberloff, argued that in a situation of biological uncertainty (the degree/type of deforestation-induced extinction), scientists ought to follow the epistemologically conservative course and risk type-II error (the risk of not rejecting a null hypothesis that is (...) false), rather than type-I error. (This is the risk of rejecting a null hypothesis that is true). Kangas' opponents, like Noss, argued that, in a situation of scientific uncertainty, scientists ought to risk type-I, rather than type-II, error. This essay argues that there are different types of rationality appropriate to science and applied science and, therefore, in cases of applied science (like conservation biology), the more conservative course of action is for scientists to risk type-I error. The essay argues further that, on grounds of scientific rationality, Kangas, Simberloff, and others were correct in risking type-II error, but that, on grounds of decision-theoretic rationality, Noss, Waide, and others were correct in risking type-I error. (shrink)

Traditional analyses of the curve fitting problem maintain that the data do not indicate what form the fitted curve should take. Rather, this issue is said to be settled by prior probabilities, by simplicity, or by a background theory. In this paper, we describe a result due to Akaike [1973], which shows how the data can underwrite an inference concerning the curve's form based on an estimate of how predictively accurate it will be. We argue that this approach throws light (...) on the theoretical virtues of parsimoniousness, unification, and non ad hocness, on the dispute about Bayesianism, and on empiricism and scientific realism. * Both of us gratefully acknowledge support from the Graduate School at the University of Wisconsin-Madison, and NSF grant DIR-8822278 (M.F.) and NSF grant SBE-9212294 (E.S.). Special thanks go to A. W. F. Edwards.William Harper. Martin Leckey. Brian Skyrms, and especially Peter Turney for helpful comments on an earlier draft. (shrink)

The present edition provides a detailed and accessible discussion ofhis theories and adds an account of the immediate responses to the book on publication.

When data are limited, simple models of complex ecological systems tend to wind up closer to the truth than more complex models of the same systems. This greater proximity to the truth, or verisimilitude, leads to greater predictive success. When more data are available, the advantage of simplicity decreases, and more complex models may gain the upper hand. In ecology, holistic models are usually simpler than reductionistic models. Thus, when data are limited, holistic models have an advantage over reductionistic models, (...) with respect to verisimilitude and predictive success. I illustrate these points with models designed to explain and predict the numbers of species on islands. (shrink)

In 1986-1987, a number of ecologists were involved in a dispute over design of wildlife reserves and species losses resulting from deforestation. The battle was played out largely in the pages of the Ecological Society of America Bulletin. The most focused aspect of the controversy began in August 1986 when P. C. Kangas gave a paper at the meetings of the Fourth International Congress of Ecology, held in Syracuse, New York.Using data on trees in Costa Rica and the “objective approach” (...) of species-area curves, Kangas argued that extinction rates due to deforestation in the tropics are actually much lower than most other researchers have alleged (Kangas 1986, p. 194). Ecologists in the audience disagreed with Kangas’ claim that extinction rates were indeed lower, complained that his questionable results would be used to justify more rapid and widespread deforestation, and claimed to have “sharply trounced his [Kangas’] arguments” (Noss 1986, pp. 278-279). (shrink)

The desire to minimize the number of individual new entities postulated is often referred to as quantitative parsimony. Its influence on the default hypotheses formulated by scientists seems undeniable. I argue that there is a wide class of cases for which the preference for quantitatively parsimonious hypotheses is demonstrably rational. The justification, in a nutshell, is that such hypotheses have greater explanatory power than less parsimonious alternatives. My analysis is restricted to a class of cases I shall refer to as (...) additive. Such cases involve the postulation of a collection of qualitatively identical individual objects which collectively explain some particular observed phenomenon. Especially clear examples of this sort occur in particle physics.1Introduction2Particle physics: a case study3Three kinds of simplicity4Explanatory power5Explanation and non‐observation6Parsimony and scientific methodology7Conclusions. (shrink)

The Mathematical Principles of Natural Philosophy.

The problem of simplicity involves three questions: How is the simplicity of a hypothesis to be measured? How is the use of simplicity as a guide to hypothesis choice to be justified? And how is simplicity related to other desirable features of hypotheses -- that is, how is simplicity to be traded-off? The present paper explores these three questions, from a variety of viewpoints, including Bayesianism, likelihoodism, and the framework of predictive accuracy formulated by Akaike (1973). It may turn out (...) that simplicity has no global justification -- that its justification varies from problem to problem. (shrink)

Research programmes are sets of problems preferred on epistemic grounds and including preferred heuristics for inquiry. Charles Lyell's research programme for biogeograpy includes the problem of explaining the distribution of species constrained by laws governing locomotion and containment of species. Included in the programme are laws governing the supernatural introduction of replacement species. Wallace and Darwin derected arguments against the putative intelligibility of this aspect of Lyell's programme before discovering natural selection, and their defence, at this time of natural laws (...) governing the introduction of new species is interesting as a defence of a noval programme of research, not a defence of a proposed theroy. These arguments are presented and an epistemological foundation is offered for them. 1 The author is indebted to a referee for helpful comments on an earlier draft of this paper. (shrink)

This paper adopts a symmetrical approach tocontroversies over R.H. MacArthur and E.O. Wilson'sequilibrium model of island biogeography, in order toshow how different interpretations of the model dependupon different philosophical understandings of theapplication of models and theories. In particular,there are quite distinct domains to which the modelcould apply; in addition, some equivocation amongthese domains is important to the model's success.Therefore, apparently inconsistent interpretations,interpretations that fit into roughly instrumentalist,realist and rationalist conceptions of science, may bemutually supporting in practice. Descriptions ofscientific practice, then, (...) should not adjudicate amongthese interpretations, but should instead recognizeways in which successful models translate amongdomains, and in so doing can become realistic,instrumentally successful, or rationally established.As complex social objects, models can afford complexrepresentational relations. (shrink)