Eliminative reasoning seems to play an important role in the sciences, but should it be part of our best theory of science? Statistical evidence, prevalent across the sciences, causes problems for eliminative inference, supporting the view that probabilistic theories of confirmation provide a better framework for reasoning about evidence. Here I argue that deductive elimination has an important inferential role to play in science, one that is compatible with probabilistic approaches to evidence. Eliminative inferences help frame testing problems, an essential step that determines the context for evaluating statistical evidence. I illustrate this process with examples from molecular evolutionary biology.

We critically examine Denis Walsh's latest attack on the causalist view of fitness. Relying on Judea Pearl's Sure-Thing Principle and geneticist John Gillespie's model for fitness, Walsh has argued that the causal interpretation of fitness results in a reductio. We show that his conclusion only follows from misuse of the models, that is, (1) the disregard of the real biological bearing of the population-size parameter in Gillespie's model and (2) the confusion of the distinction between ordinary probability and Pearl's causal probability. Properly understood, the models used by Walsh do not threaten the causalist view of fitness.

In a recent paper, Brandon and Nijhout argue against genic selectionism—the thesis, roughly, that evolutionary processes are best understood from the gene's-eye point of view—by presenting a case in which genic models of selection allegedly make predictions that conflict with the (correct) predictions of higher-level genotypic selection models. Their argument, if successful, would refute the widely held belief that genic models and higher-level models are predictively equivalent. Here, I argue that Brandon and Nijhout fail to demonstrate that the models make incompatible predictions.

The received view of dynamical explanation is that dynamical cognitive science seeks to provide covering-law explanations of cognitive phenomena. By analyzing three prominent examples of dynamicist research, I show that the received view is misleading: some dynamical explanations are mechanistic explanations and in this way resemble computational and connectionist explanations. Interestingly, these dynamical explanations invoke the mathematical framework of dynamical systems theory to describe mechanisms far more complex and distributed than the ones typically considered by philosophers. Therefore, contemporary dynamicist research reveals the need for a more sophisticated account of mechanistic explanation.

In a previous article, we have demonstrated by a general phase space argument that a Maxwellian Demon is compatible with statistical mechanics. In this article, we show how this idea can be put to work in the prevalent model of the Demon, namely, a particle-in-a-box, used, for example, by Szilard and Bennett. In the literature, this model is used in order to show that a Demon is incompatible with statistical mechanics, either classical or quantum. However, we show that a detailed phase space analysis of this model illustrates that a Maxwellian Demon is compatible with statistical mechanics.

The received view in philosophy of biology is that there is a well-understood, philosophically rigorous account of information—causal information. I argue that this view is mistaken. Causal information is fatally undermined by misinterpretations and conflations between distinct independent accounts of information. As a result, philosophical arguments based on causal information are deeply flawed. I end by briefly considering what a correct application of the relevant accounts of information would look like in the biological context.

Although many philosophers have employed the distinction between “direct” and “indirect” roles for values in science, I argue that it merits further clarification. The distinction can be formulated in several ways: as a logical point, as a distinction between epistemic attitudes, or as a clarification of different consequences associated with accepting scientific claims. Moreover, it can serve either as part of a normative ideal or as a tool for policing how values influence science. While various formulations of the distinction may (with further clarification) contribute to a normative ideal, they have limited effectiveness for regulating how values influence science.

Leuridan argued that mechanisms cannot provide a genuine alternative to laws of nature as a model of explanation in the sciences, and he advocates Mitchell's pragmatic account of laws. I first demonstrate that Leuridan gets the order of priority wrong among mechanisms, regularity, and laws, and then make some clarifying remarks about how laws and mechanisms relate to regularities. Mechanisms are not an explanatory alternative to regularities; they are an alternative to laws. The existence of stable regularities in nature is necessary for either model of explanation: regularities are what laws describe and what mechanisms explain.