In this essay, I argue that uneliminated idealizations pose a serious problem for scientific realism. I consider one method for “de-idealizing” models—robustness analysis. However, I argue that unless idealizations are eliminated from an idealized theory and robustness analysis need not do that, scientists are not justified in believing that the theory is true. I consider one example of modeling from the biological sciences that exemplifies the problem.

In several recent publications, Professor Adolf Grünbaum has inveighed against the conventionalism of writers like Einstein, Poincaré, Quine and especially Duhem. Specifically, Grünbaum has assailed the view that a single hypothesis can never be conclusively falsified. Grünbaum claims that the conventionalists’ insistence on the immunity of hypotheses from falsification is neither logically valid nor scientifically sound. Directing the weight of his argument against Duhem, Grünbaum launches a two-pronged attack. He insists, first, that conclusive falsifying experiments are possible, suggesting that Duhem's denial of such experiments is a logical non-sequitur. He then proceeds to show that, more than being merely possible, crucial falsifying experiments have occurred in physics. I do not intend to make a logical point against Grünbaum's critique so much as an historical and exegetical one. Put briefly, I believe that he has misconstrued Duhem's views on falsifiability and that the logical blunder which he discusses should not be ascribed to Duhem, but rather to those who have made Duhem's conventionalism into the doctrine which Grünbaum attacks. Whether there are any writers who accept the view he imputes to Duhem, or whether he is exploiting “straw-men” to give weight to an otherwise trivial argument is an open question. For now, I simply want to suggest that his salvos are wrongly directed against Duhem.

In this paper I criticize Cartwright's analysis of capacities and offer an alternative analysis. I argue that Cartwright's attempt to connect capacities to her condition CC fails because individuals can exercise capacities only in certain contexts. My own analysis emphasizes three features of capacities: 1) Capacities belong to individuals; 2) Capacities are typically not metaphysically fundamental properties of individuals, but can be explained by referring to structural properties of individuals; and 3) Laws are best understood as ascriptions of capacities.

The communist norm requires that scientists widely share the results of their work. Where did this norm come from, and how does it persist? I argue on the basis of a game-theoretic model that rational credit-maximizing scientists will in many cases conform to the norm. This means that the origins and persistence of the communist norm can be explained even in the absence of a social contract or enforcement, contrary to recent work by Michael Strevens but adding to previous work emphasizing the benefits of the incentive structure created by the priority rule.

Three major ways in which temporal asymmetries enter into scientific induction are discussed as follows:

1. An account is given of the physical basis for the temporal asymmetry of recordability, which obtains in the following sense: except for humanly recorded predictions and one other class of advance indicators to be discussed, interacting systems can contain reliable indicators of only their past and not of their future interactions. To deal with the exceptional cases of non-spontaneous “pre-records,” a clarification is offered of the essential differences in the conditions requisite to the production of an indicator having retrodictive significance (“post-record”), on the one hand, and of one having predictive significance (“prerecord” or recorded prediction), on the other. Purported counter-examples to the asymmetry of spontaneous recordability are refuted.

2. It is shown how in cases of asymmetric recordability, the associated retrodiction-prediction asymmetry makes for an asymmetry of assertibility as between an explanandum (or an explanans) referring to a future event and one referring to a past one. But it is argued that this epistemological asymmetry in the assertibility per se must be clearly distinguished from a logical asymmetry between the past and the future in regard to the inferability (deductive or inductive) of the explanandum from the explanans. And it is then contended that the failure to distinguish between an epistemological and a logical asymmetry vitiates the critiques that recent writers have offered of the Popper-Hempel thesis, which affirms symmetry of inferability as between predictive and post-explanatory arguments. In reply to Scriven, it is maintained that predictions based on mere indicators (rather than causes) do not establish an asymmetry in scientific understanding as between predictive arguments and post-explanatory ones.

3. As a further philosophical ramification of the retrodiction-prediction asymmetry, a set of sufficient conditions are stated for the correctness of philosophical mechanism as opposed to teleology.

In a review of my book “Wahrscheinlichkeitslehre”, Dr. Ernest Nagel has recently criticized some of my ideas on probability and induction. His review includes a good exposition of my ideas, and I have to thank him for his serious attempts to do justice to my results. He attacks, however, some very essential points of my theory. I may be allowed, therefore, to answer him as frankly and thoroughly as he attacks me.

Ontological questions in biology have typically focused on the nature of species: what are species; how are they identified and individuated? There is an analogous, but much neglected concern: what are characters; how are they identified and individuated? Character individuation is significant because biological systematics relies on a parsimony principle to determine phylogeny and classify taxa, and the parsimony principle is usually interpreted to favor the phylogenetic hypothesis that requires the fewest changes in characters. But no character individuation principle identified so far is adequate. For biological systematics we need a better way of conceiving characters.

This paper clarifies the historical nature of species by showing that species are path-dependent entities. A species’ identity is not determined by its intrinsic properties or its origin, but by its unique evolutionary path. Seeing that species are path-dependent entities has three implications: it shows that origin essentialism is mistaken, it rebuts two challenges to the species-are-historical-entities thesis, and it demonstrates that the identity of a species during speciation depends on future events.

This paper examines the so-called “gauge argument” sometimes used by physicists to motivate the introduction of gauge fields, here facilitated by an informal exposition of the fiber bundle formalism. The discussion suggests some preliminary ways of understanding the connection between gauge fields and interactions.

Darwinists classify biological traits either by their ancestry (homology) or by their adaptive role. Only the latter can provide traditional natural kinds, but only the former is practicable. Process structuralists exploit this embarrassment to argue for non-Darwinian classifications in terms of underlying developmental mechanisms. This new taxonomy will also explain phylogenetic inertia and developmental constraint. I argue that Darwinian homologies are natural kinds despite having historical essences and being spatio-temporally restricted. Furthermore, process structuralist explanations of biological form require an unwarranted assumption about the space of developmental possibility.

Criteria are given to characterize mature theories in contradistinction to developing theories. We lean heavily on the physical sciences. An established theory is defined as a mature one with known validity limits. The approximate truth of such theories is thereby given a quantitative character. Superseding theories do not falsify established theories because the latter are protected by their validity limits. This view of scientific realism leads to ontological levels and cumulativity of knowledge. It is applied to a defense of realism against recent attacks by Laudan.

Selective scientific realists disagree on which theoretical posits should be regarded as essential to the empirical success of a scientific theory. A satisfactory account of essentialness will show that the (approximate) truth of the selected posits adequately explains the success of the theory. Therefore, (a) the essential elements must be discernible prospectively; (b) there cannot be a priori criteria regarding which type of posit is essential; and (c) the overall success of a theory, or ‘cluster’ of propositions, not only individual derivations, should be explicable. Given these desiderata, I propose a “unification criterion” for identifying essential elements.

The main thrust of the paper involves a theoretical and philosophical analysis of the claim made in September 1999 that atomic orbitals have been directly imaged for the first time. After a brief account of the recent claims the paper reviews the development of the orbit and later orbital concepts and analyzes the theoretical status of atomic orbitals. The conclusion is that contrary to these claims, atomic orbitals have not in fact been observed. The non-referring nature of modern atomic orbitals is discussed in the context of Laudan's writings on realism, the success of theories, and whether or not scientific terms refer. I conclude that the failure to observe orbitals is a good prima facie case for divorcing the success of theories from the question of whether their central terms refer. The added relevance of this case is that it concerns a current and highly successful theory. Finally, the relevance of this ‘floating model’ to contemporary discussions on scientific models is briefly considered.

The need to align multiple experimental procedures and produce converging results so as to demonstrate that the phenomenon under investigation is real and not an artifact is a commonplace both in scientific practice and discussions of scientific methodology (Campbell and Stanley 1963; Wimsatt 1981). Although sometimes this is the purpose of aligning techniques, often there is a different purpose—multiple techniques are sought to supply different perspectives on the phenomena under investigation that need to be integrated to answer the questions scientists are asking. After introducing this function, I will illustrate it by considering three of the major techniques in cognitive neuroscience for linking cognitive function with neural structure.

I clarify the difference between pluralist and monist interpretations of levels of selection disputes. Lloyd has challenged my claim that a plurality of models correctly accounts for situations such as maintenance of the sickle-cell trait, and I revisit this example to show that competing theories don't disagree about the existence of ‘high-level’ or ‘low-level’ causes; rather, they parse these causes differently. Applying Woodward's theory of causation, I analyze Sober's distinction between ‘selection of’ versus ‘selection for’. My analysis shows that this distinction separates true causes from pseudocauses, but it also reveals that the distinction is irrelevant to the levels debate; it makes no sense to say true causes are at higher levels and not lower levels. The levels debate is not about separating real causes from pseudocauses; it's about finding useful ways to parse and disentangle causes.

In a recent paper, Malament (2004) employs a time reversal transformation that differs from the standard one, without explicitly arguing for it. This is a new and important understanding of time reversal that deserves arguing for in its own right. I argue that it improves upon the standard one. Recent discussion has focused on whether velocities should undergo a time reversal operation. I address a prior question: What is the proper notion of time reversal? This is important, for it will affect our conclusion as to whether our best theories are time-reversal symmetric, and hence whether our spacetime is temporally oriented.

The operations of deep networks are widely acknowledged to be inscrutable. The growing field of Explainable AI (XAI) has emerged in direct response to this problem. However, owing to the nature of the opacity in question, XAI has been forced to prioritise interpretability at the expense of completeness, and even realism, so that its explanations are frequently interpretable without being underpinned by more comprehensive explanations faithful to the way a network computes its predictions. While this has been taken to be a shortcoming of the field of XAI, I argue that it is broadly the right approach to the problem.

Whereas much of the literature in the social epistemology of scientific knowledge has focused either on scientific communities or research groups, we examine the epistemic significance of scientific/intellectual movements (SIMs). We argue that certain types of SIMs can play an important epistemic role in science: they can remedy epistemic injustices in scientific practices. SIMs can counteract epistemic injustices effectively because many forms of epistemic injustice require structural and not merely individual remedies. To illustrate our argument, we discuss the case of indigenous studies.