This paper argues, first, that recent studies of experimentation, most notably by Deborah Mayo, provide the conceptual resources to describe scientific discovery's early stages as error-probing processes. Second, it shows that this description yields greater understanding of those early stages, including the challenges that they pose, the research strategies associated with them, and their influence on the rest of the discovery process. Throughout, the paper examines the phenomenon of “chemical hormesis” (i.e., anomalous low-dose effects from toxic chemicals) as a case study that is important not only for the biological sciences but also for contemporary public policy. The resulting analysis is significant for at least two reasons. First, by elucidating the importance of discovery's earliest stages, it expands previous accounts by philosophers such as William Wimsatt and Lindley Darden. Second, it identifies the discovery process as yet another philosophical topic on which the detailed studies of the “new experimentalists” can shed new light.

We attempt to defend the species-as-individuals hypothesis by examining the logical role played by the binomials (e.g., “Homo sapiens,” “Pinus ponderosa”) in biological discourse about species. Those who contend that the binomials can be properly understood as functioning in biological theory as singular terms opt for an objectual account of species and view species as individuals. Those who contend that the binomials can in principle be eliminated from biological theory in favor of predicate expressions opt for a predicative account of species and view species as kinds. We contend that biologists' talk about species is talk about species as individuals, and we conclude that the most plausible account of species is an objectual account.

This article presents a new framework for the analysis of experimental control. The framework highlights different functions for experimental controls in the realization of an experiment: experimental controls that serve as tests and experimental controls that serve as probes. The approach to experimental control proposed here can illuminate the constitutive role of controls in knowledge production, and it sheds new light on the notion of exploratory experimentation. It also clarifies what can and what cannot be expected from reviewers of scientific journal articles giving feedback on experimental controls.

The Intergovernmental Panel on Climate Change has developed a novel framework for assessing and communicating uncertainty in the findings published in its periodic assessment reports. But how should these uncertainty assessments inform decisions? We take a formal decision-making perspective to investigate how scientific input formulated in the IPCC’s novel framework might inform decisions in a principled way through a normative decision model.

This article considers the temporal dimension of data processing and use and the ways in which it affects the production and interpretation of knowledge claims. I start by distinguishing the time at which data collection, dissemination, and analysis occur (Data time, or Dt) from the time in which the phenomena for which data serve as evidence operate (Phenomena time, or Pt). Building on the analysis of two examples of data reuse from modeling and experimental practices in biology, I then argue that Dt affects how researchers (1) select and interpret data as evidence and (2) identify and understand phenomena.

Because statements like ‘This object is soluble in aqua regia’ involve the causal modalities, we call them causal dispositional statements. Now while this involvement has long been recognized, no thorough examination of its exact nature has ever been made. One purpose of this paper is to begin such an examination. In Sec. 2 we will suggest an analysis of causal dispositional statements, and in Sec. 3 we will discuss some philosophic issues to which this analysis is relevant.

Sober and Lewontin's critique of genic selectionism is based upon the principle that a unit of selection should make a context-independent contribution to fitness. Critics have effectively shown that this principle is flawed. In this paper I show that the context independence principle is an instance of a more general principle for characterizing causes,called the contextual unanimity principle. I argue that this latter principle, while widely accepted, is erroneous. What is needed is to replace the approach to causality characterized by the contextual unanimity criterion with an approach based on the concept of causal mechanism. After sketching such an approach, I show how it can be used to shed light on the units of selection problem.

I defend the interpretation of the Aharonov-Bohm effect originally advanced by Aharonov and Bohm, i.e., that it is caused by an interaction between the electron and the vector potential. The defense depends on taking the fiber bundle formulation of electrodynamics literally, or almost literally.

In the curve fitting problem two conflicting desiderata, simplicity and goodness-of-fit pull in opposite directions. To solve this problem, two proposals, the first one based on Bayes's theorem criterion (BTC) and the second one advocated by Forster and Sober based on Akaike's Information Criterion (AIC) are discussed. We show that AIC, which is frequentist in spirit, is logically equivalent to BTC, provided that a suitable choice of priors is made. We evaluate the charges against Bayesianism and contend that AIC approach has shortcomings. We also discuss the relationship between Schwarz's Bayesian Information Criterion and BTC.

Skipper and Millstein analyze natural selection and mechanism, concluding that natural selection is not a mechanism in the sense of the new mechanistic philosophy. Barros disagrees and provides his own account of natural selection as a mechanism. This discussion identifies a missing piece of Barros's account, attempts to fill in that piece, and reconsiders the revised account. Two principal objections are developed: one, the account does not characterize natural selection; two, the account is not mechanistic. Extensive and persistent variability causes both of these difficulties, so further attempts to describe natural selection as a mechanism are also unlikely to succeed.

Four philosophical problems—predication, speech acts, rules, and innate ideas— are discussed in the light of their implications for psychological and linguistic research. The discussion of predication concerns both form and use. With respect to form, it is argued that our lexical memory is organized according to a predicate-argument formula that underlies the subject-predicate form of our sentences. With respect to use, it is argued that the illocutionary force of the sentence as a speech act must be taken into account. Both the formation and the use of such verbal constructions are normally characterized by systems of rules, but there is no clear account of what a rule is or how it might operate to control behavior, and this problem is especially difficult when, as in language, the person's knowledge of the rules is implicit. The innate basis for our human ability to acquire linguistic rules is considered and the problem of innateness is redefined around the conjecture that there are innate, language-specific mechanisms unique to human beings. The problem of investigating such language-specific mechanisms psychologically, however, is quite difficult at the present time.

In this paper I provide a novel argument against the claim that classical genetics is being reduced to molecular genetics. Specifically, I demonstrate that reductionists must subscribe to the unargued and problematic thesis that molecular genetics is ‘independent’ of classical genetics. I also argue that several standard antireductionist positions can be faulted for unnecessarily conceding the Independence Thesis to the reductionists. In place of a ‘tale of two sciences’, I offer a ‘heroic’ stance that denies classical genetics is being reduced, yet sees classical and molecular genetics as fundamentally unified.

I analyze a number of the quantum no-signalling proofs (Ghirardi et al. 1980, Bussey 1982, Jordan 1983, Shimony 1985, Redhead 1987, Eberhard and Ross 1989, Sherer and Busch 1993). These purport to show that the EPR correlations cannot be exploited for transmitting signals, i.e., are not causal. First, I show that these proofs can be mathematically unified; they are disguised versions of a single theorem. Second, I argue that these proofs are circular. The essential theorem relies upon the tensor product representation for combined systems, which has no physical basis in the von Neumann axioms. Historically, the construction of this representation scheme by von Neumann and Weyl built no-signalling assumptions into the quantum theory. Signalling between the wings of the EPR-Bell experiments is unlikely but is not ruled out empirically by the class of proofs considered.

Several quantitative techniques for choosing among data models are available. Among these are techniques based on algorithmic information theory, minimum description length theory, and the Akaike information criterion. All these techniques are designed to identify a single model of a data set as being the closest to the truth. I argue, using examples, that many data sets in science show multiple patterns, providing evidence for multiple phenomena. For any such data set, there is more than one data model that must be considered close to the truth. I conclude that, since the established techniques for choosing among data models are unequipped to handle these cases, they cannot be regarded as adequate.

Professionals often believe that they must “exercise judgment” in making decisions critical to other people's lives. The relative superiority (established in roughly 150 studies) of statistical prediction rules (SPR's) to intuitive judgment for combining incomparable sources of information to predict important human outcomes leads us to question this personal input belief. Some professionals hence use SPR's to “educate” intuitive judgment, rather than replace it. In psychology in particular, such amalgamation is not justified. If a well-validated SPR that is superior to professional judgment exists in a relevant decision making context, professionals should use it, totally absenting themselves from the prediction.

The traditional view of evolutionary theory asserts that we can usefully understand natural selection, drift, mutation, migration, and the system of mating as forces that cause evolutionary change. Recently, Denis Walsh and Robert Brandon have objected to this view. Walsh argues that the traditional view faces a fatal dilemma and that the force analogy must be rejected altogether. Brandon accepts the force analogy but argues that drift, rather than the Hardy-Weinberg law, is the best candidate for a zero-force law. Here I defend the traditional view against these objections.

My thesis is that biology is most plausibly regarded as a universal, as distinct from a provincial, science. First, I develop the general notion of a provincial science, formulate three criteria for applying the concept, and present brief examples illustrating their use. Second, I argue that a consideration of population genetics as a characteristic example of a basic biological theory strengthens the prior presumption that biology is not a provincial science. Finally, I examine two arguments to the effect that biology is a provincial science. The first concerns biology's exclusively terrestrial evidence base, the second the logical character of its laws. I introduce considerations that weaken the persuasiveness of the first argument and then show that the second one rests upon a false premise and so should be rejected.

In 1966, Richard Levins argued that there are different strategies in model building in population biology. In this paper, I reply to Orzack and Sober's (1993) critiques of Levins and argue that his views on modeling strategies apply also in the context of evolutionary genetics. In particular, I argue that there are different ways in which models are used to ask and answer questions about the dynamics of evolutionary change, prospectively and retrospectively, in classical versus molecular evolutionary genetics. Further, I argue that robustness analysis is a tool for, if not confirmation, then something near enough, in this discipline.

This paper studies the theory of uniqueness of scales of measurement, and in particular, the theory of meaningfulness of statements using scales. The paper comments on the general theory of meaningfulness adopted by Luce in connection with his work on dimensionally invariant numerical laws. It comments on Luce's generalization of the concept of meaningfulness of a statement involving scales to a concept of meaningfulness of an arbitrary relation relative to the defining relations in a relational structure. It is argued that in studying the concept of meaningfulness, it is necessary to consider invariance under endomorphisms, not just automorphisms. The difference between the endomorphism and automorphism concepts of meaningfulness is studied. Luce's primary result, that automorphism meaningfulness is preserved under isomorphism, is extended to the result that endomorphism meaningfulness is preserved under homomorphism.

When, in recent years, philosophers of science attend to the physical sciences, their activity centers on the analysis and clarification of the methods and theories of these sciences. Thus philosophers have made remarkable contributions to our understanding of mechanics, the relativity theory, the quantum theory, probability, and geometry, as well as to the foundations of mathematics. More general discussions about theory construction, explanation, and concept formation always are illustrated and illuminated by reference to specific theories within science. In our generation, most philosophers of science have taken for granted that physical science is descriptive, empirical, functional, and—leaving aside controversial interpretations of quanta—deterministic.