van Fraassen's constructivist empiricist account of theories makes an epistemic distinction between entities that can and cannot be observed with the naked eye. A belief about the correctness of a theoretical description of an entity that is observable with the naked eye can be warranted by a theory. In contrast, no theory can warrant a belief about the correctness of a description of an unobservable entity.

I argue that we ought to instead adopt a view that takes account of the fact that some entities that cannot be observed with the naked eye can nevertheless be observed on the basis of the same physical principle as those entities that can be. This suggests that there is a distinction different from van Fraassen's that might do the work van Fraassen intends his to do, but a distinction that is principled. Understanding why this is so suggests that his distinction is grounded merely in human chauvinism.

It is widely thought that the size of a probability makes no difference to the quality of a probabilistic explanation. I argue that explanatory practice in statistical physics, past and present, belies this claim. The claim has gained currency only because of an impoverished conception of the nature of probabilistic processes and an unwarranted assumption that all probabilistic explanations have a single form.

There is no set Δ of probability axioms that meets the following three desiderata:

• (1) Δ is vindicated by a Dutch book theorem;

• (2) Δ does not imply regularity (and thus allows, among other things, updating by conditionalization);

• (3) Δ constrains the conditional probability q(·, z) even when the unconditional probability p(z) (= q(z, T)) equals 0.

• This has significant consequences for Bayesian epistemology, some of which are discussed.

No one has a well developed solution to Duhem's problem, the problem of how experimental evidence warrants revision of our theories. Deborah Mayo proposes a solution to Duhem's problem in route to her more ambitious program of providing a philosophical account of inductive inference and experimental knowledge. This paper is a response to Mayo's Error Statistics (ES) program, paying particular attention to her response to Duhem's problem. It turns out that Mayo's purported solution to Duhem's problem is very significant to her project, for the epistemic license claimed by ES and the philosophical underpinnings to her account of experimental knowledge depend on this solution. By introducing the partition problem, I argue that ES fails to solve Duhem's problem and therefore fails to provide an adequate account of experimental knowledge.

Kant's views on the epistemological status of physical science provide an important example of how a philosophical system can be applied to understand the foundation of scientific theories. Michael Friedman has made considerable progress towards elucidating Kant's philosophy of science; in particular, he has argued that Kant viewed Newton's law of universal gravitation as necessary for the possibility of experiencing what Kant called true motion, which is more than the mere relative motion of appearances but is different from Newton's concept of absolute motion. In this context, Friedman has provided an account of how Kant must have viewed Newton's supposed derivation of universal gravitation from Kepler's laws, based on, among other things, Kant's claim that Newton really needed to make extra assumptions in order to derive universal gravitation. In this paper, I argue that Friedman's account is incomplete for three reasons. First, Friedman has overlooked an important aspect of how Newton's third law is applied in the relevant sections of the Principia; as a result, Friedman's account partially misconstrues the relation between the planetary phenomena and the theory of universal gravitation. Second, his account fails to account for Kant's apparent belief that Kepler's laws are only empirically-based rules, even though they seem to be necessary for the derivation of universal gravitation and hence also necessary for Kant's own definition of true motion. Third, Friedman has overlooked some remarks by Kant that indicate that Kant thought the crucial properties of universal gravitation could be known without reference to the empirically determined motions of the planets and hence seemingly without any help from Newton.

This paper explores the relationship between psychology and neurobiology in the context of cognitive science. Are the sciences that constitute cognitive science independent and theoretically autonomous, or is there a necessary interaction between them? I explore Fodor's Multiple Realization Thesis (MRT) which starts with the fact of multiple realization and purports to derive the theoretical autonomy of special sciences (such as psychology) from structural sciences (such as neurobiology). After laying out the MRT, it is shown that, on closer inspection, the argument is either circular or self-undermining—the argument either assumes the very autonomy it seeks to demonstrate or the concluded autonomy is contradicted by the theoretical interdependence invoked by the premises of the argument. Next, I explore a concrete example of multiple realization in the explanation of animal behavior: the convergent evolution of jamming avoidance behaviors in three genera of weakly electric fish. Contrary to the image painted by the MRT, the work on these animals involves a high degree of interaction between the various levels of investigation. The fact that our understanding of electric fish behavior involves functional theories and multiple realization without the kind of disunified science that is supposed to follow from such a situation indicates that the mere fact of multiple realization cannot be the basis for an autonomous psychology.

Non-reductive physicalists have made a number of attempts to provide the relation of supervenience between levels of properties with enough bite to analyze interesting cases without at the same time losing the relation's acceptability for the physicalist. I criticize some of these proposals and suggest an alternative supplementation of the supervenience relation by imposing a requirement of robustness which is motivated by the notion of structural stability familiar from dynamical systems theory. Robust supervenience, I argue, captures what the non-reductive physicalist wants from supervenience; most importantly, it provides a natural background for reconstructing the notion of (diachronic) property emergence in a way acceptable to physicalists.

Evolutionary game theoretic accounts of justice attempt to explain our willingness to follow certain principles of justice by appealing to robustness properties possessed by those principles. Skyrms (1996) offers one sketch of how such an account might go for divide-the-dollar, the simplest version of the Nash bargaining game, using the replicator dynamics of Taylor and Jonker (1978). In a recent article, D'Arms et al. (1998) criticize his account and describe a model which, they allege, undermines his theory. I sketch a theory of evolutionary explanations of justice which avoids their methodological criticisms, and develop a spatial model of divide-the-dollar with more robust convergence properties than the models of Skyrms (1996) and D'Arms et al. (1998).

In a recent essay in this journal, John Maynard Smith (2000) argues that the often expressed idea that the genome is the repository of meaningful information is not merely a heuristically useful metaphor. Instead, he contends, it is a central idea in contemporary microbiology. While I am in general agreement with Maynard Smith on this issue, his account suffers, I believe, from using an inappropriate concept of ‘information.’ One result of this is that the concept of genomic information becomes burdened by a conceptual load that the biological facts cannot support. When, however, the appropriate concept of information is used, then these burdens are shed, and the result is an account that is in substantial agreement with (what I consider to be) the heart of Maynard Smith's thesis.