In this essay, I criticize Laudan's view that methodological rules in science are best understood as hypothetical imperatives, for example, to realize cognitive aim A, follow method B. I criticize his idea that such rules are best evaluated by a naturalized philosophy of science which collects the empirical evidence bearing on the soundness of these rules. My claim is that this view yields a poor explanation of (1) the role of methodological rules in establishing the rationality of scientific practices, (2) the debates scientists have over rival methodological rules, and (3) the reasons on the basis of which scientists actually embrace, challenge, or transform methodological rules. A better explanation is provided by treating at least some methodological rules as foundational standards of theory-appraisal which define the criteria of scientific truth, knowledge, evidence, proof, explanations, etc., for a certain tradition or community of scientific practice. Such standards are in turn grounded or challenged on the basis of their coherence or incoherence with a wider body of scientific judgments. This view leads away from a naturalized philosophy of science back to its more classical epistemological conception.

The fact that the goals and methods of science, as well as its empirical conclusions, are subject to change, is shown to allow at once for: (a) the objectivity of warrant for knowledge claims; (b) the absence of a priori standards from epistemology; (c) the normative character of epistemology; and (d) the rationality of axiological innovation. In particular, Laudan's attempt to make axiological constraints undercut epistemic realism is confuted.

The prescriptive force of methodological rules rests, I argue, on the acceptance of scientific theories; that of the most general methodological rules rests on theories in the philosophy of science, which differ from theories in the several sciences only in generality and abstraction. I illustrate these claims by reference to methodological disputes in social science and among philosophers of science. My conclusions substantiate those of Laudan except that I argue for the existence of transtheoretical goals common to all scientists and concrete enough actually to have bearing on methodology. And I argue that Laudan is committed to such goals himself, willy nilly.

Normative naturalism is a view about the status of epistemology and philosophy of science; it is a meta-epistemology. It maintains that epistemology can both discharge its traditional normative role and nonetheless claim a sensitivity to empirical evidence. The first sections of this essay set out the central tenets of normative naturalism, both in its epistemic and its axiological dimensions; later sections respond to criticisms of that species of naturalism from Gerald Doppelt, Jarrett Leplin and Alex Rosenberg.

Rational actions reflect beliefs and preferences in certain orderly ways. The problem of theory is to explain which beliefs and preferences are relevant to the rationality of particular actions, and exactly how they are relevant. One distinction of interest here is between an agent's beliefs and preferences just before an action's time, and his beliefs and preferences at its time. Theorists do not agree about the times of beliefs and desires that are relevant to the rationality of action. Another distinction is between actions that would, in one sense or another, maximize expected utilities given relevant beliefs and desires, and actions, decisions for which would, in one sense or another, be stable. There is disagreement about the relevance of these things to the rationality of actions. Here, in the first part below, several possible positions on these issues are explained and compared. In the second part, I contrast perspectives on these issues, and comment on arguments that might be brought against the position I favor. In the third part, I restate and elaborate on this position.

A computational theory of induction must be able to identify the projectible predicates, that is to distinguish between which predicates can be used in inductive inferences and which cannot. The problems of projectibility are introduced by reviewing some of the stumbling blocks for the theory of induction that was developed by the logical empiricists. My diagnosis of these problems is that the traditional theory of induction, which started from a given (observational) language in relation to which all inductive rules are formulated, does not go deep enough in representing the kind of information used in inductive inferences.

As an interlude, I argue that the problem of induction, like so many other problems within AI, is a problem of knowledge representation. To the extent that AI-systems are based on linguistic representations of knowledge, these systems will face basically the same problems as did the logical empiricists over induction.

In a more constructive mode, I then outline a non-linguistic knowledge representation based on conceptual spaces. The fundamental units of these spaces are “quality dimensions”. In relation to such a representation it is possible to define “natural” properties which can be used for inductive projections. I argue that this approach evades most of the traditional problems.

I attempt to complement my earlier critiques of human sociobiology, by offering an account of how evolutionary ideas might legitimately be employed in the study of human social behavior. The main emphasis of the paper is the need to integrate studies of proximate mechanisms and their ontogenesis with functional/evolutionary research. Human psychological complexity makes it impossible to focus simply on specific types of human behavior and ask for their functional significance. For any of the kinds of behavior patterns that have occupied human sociobiologists, the underlying proximate mechanisms are very likely to be linked to a broad spectrum of types of behavior, and we cannot expect that natural selection will have acted directly on any individual element from this spectrum. I illustrate this general point with a specific example, considering the traditional sociobiological account of human incest-avoidance and outlining an alternative approach to the phenomena. The example is intended to show the possibility of a more rigorous and sophisticated successor to human sociobiology, which I call “human behavioral ecology”.

Several prominent philosophers have held that physical properties are dispositions. The aim of this paper is to establish the following conjunction: if the thesis that physical properties are dispositions is unsupplemented by controversial assumptions about dispositions, it entails a contradiction; and if it is so supplemented the resulting theory has the consequence that either many worlds which seem to be possible worlds are not possible worlds or some properties which seem to be identical are not identical. In this way it is shown that a dispositional account of physical properties is implausible.

This critical note concerns Harold Kincaid's “Reduction, Explanation and Empiricism” (this journal, December 1986). Kincaid criticizes methodological individualism on several grounds. The present note argues that Kincaid fails at least in his attempt to show that it is false that individualistic theory suffices to fully explain social phenomena. Kincaid's main reason for claiming that individualistic theory is insufficient is that it cannot adequately explain social kinds. The present note contends that an individualist can suitably reinterpret the social talk in question and so succeed in explaining social phenomena. In some cases his explanatory factors will presumably be direct counterparts to the holist's explanatory social kinds, whereas in some other cases they may not be—arguably need not be.

Tuomela (this issue, pp. 96–103) raises several objections to the analysis and critique of methodological individualism in my (1986). In what follows I reply to those criticisms, arguing, among other things, that: (1) the alleged reductions provided by Tuomela and others fail, because they either presuppose rather than eliminate social predicates or do not avoid the problem of multiple realizations; (2) supervenience does not guarantee that the social sciences are reducible, because merely describing supervenieence bases leaves numerous questions unanswered, and (3) the eliminativism that Tuomela favors verges on being self-refuting and is highly implausible short of a detailed empirical critique of the social sciences, something Tuomela does not provide and something there is little reason to think can be provided.

In this paper we investigate Accardi's claim that the “quantum paradoxes” have their roots in probability theory and that, in particular, they can be evaded by giving up Bayes' rule, concerning the relation between composite and conditional probabilities. We reach the conclusion that, although it may be possible to give up Bayes' rule and define conditional probabilities differently, this contributes nothing to solving the philosophical problems which surround quantum mechanics.

Suppose my utilities are representable by a set of utility assignments, each defined for atomic sentences; suppose my beliefs are representable by a set of probability assignments. Then each of my utility assignments together with each of my probability assignments will determine a utility assignment to non-atomic sentences, in a familiar way. This paper is concerned with the question, whether I am committed to all the utility assignments so constructible. Richard Jeffrey (1984) says (in effect) “no”, Isaac Levi (1974) says “yes”. I argue for “no”, and raise in passing a problem for Levi.