What sort of subject is the philosophy of science? It would be difficult to find any sort of agreement on the answer to this question. There are a large number of physicists who have promoted the idea that the subject is a kind of cosmic journalism: any new major discoveries in physics warrant an up-to-the-minute, catch-as-catch-can analysis of the boundaries of scientific knowledge. On the other hand, the Oxonian sirens of ordinary language tell us that any careful scrutiny of technical scientific matters is not proper philosophical activity. I want to argue that the philosopher of science need be neither a journalist of science nor merely an acute man of common sense eternally restricted to contemplating the general meaning of such notions as those of mind, free will, cause and determinism.

Using Gebharter’s representation, we consider aspects of the problem of discovering the structure of unmeasured submechanisms when the variables in those submechanisms have not been measured. Exploiting an early insight of Sober’s, we provide a correct algorithm for identifying latent, endogenous structure—submechanisms—for a restricted class of structures. The algorithm can be merged with other methods for discovering causal relations among unmeasured variables, and feedback relations between measured variables and unobserved causes can sometimes be learned.

This article gives an explicit presentation of Newtonian gravitation on the backdrop of Maxwell space-time, giving a sense in which acceleration is relative in gravitational theory. However, caution is needed: assessing whether this is a robust or interesting sense of the relativity of acceleration depends on some subtle technical issues and on substantive philosophical questions over how to identify the space-time structure of a theory.

It is well known that a degree-of-belief function P is coherent if and only if it satisfies the probability calculus. In this paper, we show that the notion of coherence can be extended to higher order probabilities such as P(P(h) = p) = q, and that a higher order degree-of-belief function P is coherent if and only if it satisfies the probability calculus plus the following axiom: P(h) = p iff P(P(h) = p) = 1. Also, a number of lemmata which extend an incomplete probability function to a complete one are established.

In “Concepts Are Not a Natural Kind” (2005), I argued that the notion of concept in psychology and in neuropsychology fails to pick out a natural kind. Piccinini and Scott (2006, in this issue) have criticized the argument I used to support this conclusion. They also proposed two alternative arguments for a similar conclusion. In this reply, I rebut Piccinini and Scott's main objection against the argument proposed in “Concepts Are Not a Natural Kind.” Moreover, I show that the two alternative arguments developed by Piccinini and Scott are not promising for supporting the conclusion that concepts are not a natural kind.

This paper is concerned with two closely related problems: the first is the general question of border-line cases; the second is a suggested identification of the notions of ambiguity and vagueness. In the first part of the paper I propose to discuss border-line cases in the following way: I shall say what is meant by “border-line cases,” discuss their genesis, enumerate and evaluate the different methods of resolving such cases, and make a brief comment or two the bearing, if any, that such cases have on logical theory. In the second part it will be suggested that the notion of vagueness can be reduced to the status of a special but important case of ambiguity; I hold that such a reduction would effect both an economy and a clarification.

The distinction between analog and digital representation is reexamined; it emerges that a more fundamental distinction is that between symbolic and analog simulation. Analog simulation is analyzed in terms of a (near) isomorphism of causal structures between a simulating and a simulated process. It is then argued that a core concept, naturalistic analog simulation, may play a role in a bottom-up theory of adaptive behavior which provides an alternative to representational analyses. The appendix discusses some formal conditions for naturalistic analog simulation.

In a recent article, Wayne (1995) offers a critique of Horwich's (1982) Bayesian explication of the confirmational significance of evidential diversity. Presently, I argue that Wayne's reconstruction of Horwich's account is not faithful to Horwich's original presentation. Because Wayne's reconstruction is uncharitable, his criticisms turn out to be off the mark. I try to provide a more faithful and charitable rendition of Horwich's account. Unfortunately, even this more charitable reconstruction of Horwich's position seems—at best—to provide an incomplete explication of the confirmational significance of evidential diversity.

There are few, perhaps no known, exact, true, general laws. Some of the work of generalization is carried by ceteris paribus generalizations. I suggest that many models continue such work in more complex form, with the idea of ceteris paribus conditions thought of as extended to more general conditions of application. I use the term regularity guide to refer collectively to cp-generalizations and such regularity-purveying models. Laws in the traditional sense can then be thought of as idealizations, which idealize away from the conditions of application of regularity guides. If we keep clearly in mind the status of laws as such idealizations, problems surrounding traditional topics—such as lawlikeness, corresponding counterfactuals and modality—no longer look to be intractable.

Although cases are central to the epistemic practices utilized within clinical medicine, they appear to be limited in their ability to provide evidence about causal relations because they provide detailed accounts of particular patients without explicit filtering of those attributes most likely to be relevant for explaining the phenomena observed. This paper uses a series of recent case reports to explore the role of cases in causal attribution in medical diagnosis. It is argued that cases are brought together by practitioners to generate causal attributions and testable predictions using a manipulability view of causation.

Several argue that truth cannot be science’s sole epistemic goal, for it would fail to do justice to several scientific practices that advance understanding. I challenge these arguments but only after making a small concession: science’s sole epistemic goal is not truth as such; rather, its goal is finding true answers to relevant questions. Using examples from the natural and social sciences, I then show that scientific understanding’s epistemically valuable features are either true answers to relevant questions or a means thereof.

The dark matter problem in astrophysics exposes an underappreciated weakness in the evidential warrant for General Relativity (GR). The “dark matter double bind” entails that GR gets no differential evidential support from dynamical phenomena occurring at scales larger than our solar system, as compared to members of a significant class of rival gravitation theories. These rivals are each empirically indistinguishable from GR for phenomena taking place at solar system scales, but make predictions that may differ radically from GR's at larger scales. Thus the typical confidence in the universal applicability of GR is insufficiently warranted in the present evidential context.

The attempt to draw a clear distinction between Philosophy and the empirical sciences can almost be taken as the defining trait of the analytic movement in contemporary philosophical thought. The empirical science that has most frequently threatened to swallow up questions of particular interest to philosophers since the time of Descartes has been psychology. Characteristic, then, of analytic philosophy has been the rejection of what it terms psychologism, that is to say, the mistake of identifying philosophical categories with those of psychology, whether introspective or behavioristic. It is clear that to launch an attack on psychologism, thus conceived, presupposes that one has a list of philosophical categories which one is able to identify as such; and this in turn presupposes an ability to sketch, at least in a general way, a distinctly philosophical account of these concepts, although a systematic account along non-psychologistic lines may be a distant and ill-defined goal. The analytic movement in philosophy has gradually moved towards the conclusion that the defining characteristic of philosophical concepts is that they are formal concepts relating to the formation and transformation rules of symbol structures called languages. Philosophy, in other words, tends to be conceived of as the formal theory of languages. From this standpoint, consequently, psychologism is conceived of as the psychological treatment of concepts which are properly understood as formal devices defining a mode of linguistic structure.

Teller argued that violations of Bell's inequalities are to be explained by interpreting quantum entangled systems according to ‘relational holism’, that is, by postulating that they exhibit irreducible (‘inherent’) relations. Teller also suggested a possible application of this idea to quantum statistics. However, the basic proposal was not explained in detail nor has the additional idea about statistics been articulated in further work. In this article, I reconsider relational holism, amending it and spelling it out as appears necessary for a proper assessment, and application, of the position.

The title of this symposium, “Formal Simplicity as a Weight in the Acceptability of Scientific Theories,” to some people might seem to suggest that we are to be making positive proposals about how the concept of simplicity could be defined for formalized languages, defined so as to figure in a formalized theory of confirmation. I must confess at the start that I do not have any such ambitious object in view. I now feel, indeed, that premature formalizations have little power to illuminate the philosophically interesting questions which cluster round the problem of the role of simplicity in scientific thinking. So in this short paper I wish merely to present some elementary considerations, not very novel ones, concerning the role which simplicity seems to me to play in our non-demonstrative reasoning about matters of empirical fact. Although many writers have sought to analyze the logical character of such reasoning, little unanimity has been attained in their over-all views; thus it is that though the considerations which I wish to present are elementary, they are not wholly uncontroversial. But because these matters are elementary in nature, I do feel it appropriate in discussing them to use homely examples of uncomplicated kinds, rather than elaborate examples drawn from the more sophisticated reaches of scientific theory; in doing so, I am taking it for granted that scientific inference as regards its logical character is fundamentally a refinement of everyday thinking, rather than a procedure of some essentially different nature.

Competition between scientific hypotheses is not always a matter of mutual exclusivity. Consistent hypotheses can compete to varying degrees either directly or indirectly via a body of evidence. We motivate and defend a particular account of hypothesis competition by showing how it captures these features. Computer simulations of Bayesian inference are used to highlight the limitations of adopting mutual exclusivity as a simplifying assumption to model scientific reasoning, particularly due to the exclusion of hypotheses that may be true. We end with a case study demonstrating the subtleties involved in hypothesis competition in scientific practice.

By reference to Maxwell's kinetic theory, one feature of hypothetico-deductivism is defended. A scientist need make no inference to a hypothesis when he first proposes it. He may have no reason at all for thinking it is true. Yet it may be worth considering. In developing his kinetic theory there were central assumptions Maxwell made (for example, that molecules are spherical, that they exert contact forces, and that their motion is linear) that he had no reason to believe true. In this paper I develop a position that explains why they were worth considering, and that rejects the retroductive position that a hypothesis is worth considering when, if true, it would explain the observed data.