This paper investigates what molecular biology has done for our understanding of the gene. I base a new account of the gene concept of classical genetics on the classical dogma that gene differences cause phenotypic differences. Although contemporary biologists often think of genes in terms of this concept, molecular biology provides a second way to understand genes. I clarify this second way by articulating a molecular gene concept. This concept unifies our understanding of the molecular basis of a wide variety of phenomena, including the phenomena that classical genetics explains in terms of gene differences causing phenotypic differences.

Most biologists use the Linnaean system for constructing classifications of the organic world. The Linnaean system, however, has lost its theoretical basis due to the shift in biology from creationist and essentialist tenets to evolutionary theory. As a result, the Linnaean system is both cumbersome and ontologically vacuous. This paper illustrates the problems facing the Linnaean system, and ends with a brief introduction to an alternative approach to biological classification.

I adopt a cladistic view of species, and explore the possibility that there exists an equally valuable cladistic view of organismic traits. This suggestion seems to run counter to the stress on functional views of biological traits in recent work in philosophy and psychology. I show how the tension between these two views can be defused with a multilevel view of biological explanation. Despite the attractions of this compromise, I conclude that we must reject it, and adopt an essentially cladistic conception of biological traits.

Because of the problems associated with ecological concepts, generalizations, and proposed general theories, applied ecology may require a new “logic” of explanation characterized neither by the traditional accounts of confirmation nor by the logic of discovery. Building on the works of Grünbaum, Kuhn, and Wittgenstein, we use detailed descriptions from research on conserving the Northern Spotted Owl, a case typical of problem solving in applied ecology, to (1) characterize the method of case studies; (2) survey its strengths; (3) summarize and respond to its shortcomings; and (4) investigate and defend its underlying “logic”.

When classical mechanics is seen as the short-wavelength limit of quantum mechanics (i.e., as the limit of geometrical optics), it becomes clear just how serious and all-pervasive the measurement problem is. This formulation also leads us into the Bohm theory. But this theory has drawbacks: its nonuniqueness, in particular, and its nonlocality. I argue that these both reflect an underlying problem concerning information, which is actually a deeper version of the measurement problem itself.

To date, both sides in the conventionality of simultaneity debate grant that transformations from “standard” to “nonstandard” coordinates are possible without any empirically significant effects. However, it is argued here that the very possibility of defining nonstandard coordinates vanishes if one represents special relativity, not by real four-vectors (as has been the case so far in the debate), but by complex spinors as used in the representation of half-integer spin. Thus, in the topologically simplest representation of the Lorentz group, the transformation over which the debate has been fought does not exist and one of the most basic foundations of the conventionalist thesis is challenged.

This paper explores a new reason for preferring a model-theoretic approach to understanding the nature of scientific theories. Identifying the models in philosophers' model-theoretic accounts of theories with the concepts in cognitive scientists' accounts of categorization suggests a structure to families of models far richer than has commonly been assumed. Using classical mechanics as an example, it is argued that families of models may be “mapped” as an array with “horizontal” graded structures, multiply hierarchical “vertical” structures, and local “radial” structures. These structures promise important implications for how scientific theories are learned and used in actual scientific practice.

This paper presents a drastically revised version of the theory of causality, based on analyses of causal processes and causal interactions, advocated in Salmon (1984). Relying heavily on modified versions of proposals by P. Dowe, this article answers penetrating objections by Dowe and P. Kitcher to the earlier theory. It shows how the new theory circumvents a host of difficulties that have been raised in the literature. The result is, I hope, a more satisfactory analysis of physical causality.