John Norton has recently argued that Newtonian gravitation theory (at least as applied to cosmological contexts where one envisions the possibility of a homogeneous mass distribution throughout all of space) is inconsistent. I am not convinced. Traditional formulations of the theory may seem to break down in cases of the sort Norton considers. But the difficulties they face are only apparent. They are artifacts of the formulations themselves, and disappear if one passes to the so-called “geometrized” formulation of the theory.

It has recently been argued that the success of the connectionist program in cognitive science would threaten folk psychology. I articulate and defend a “minimalist” construal of folk psychology that comports well with empirical evidence on the folk understanding of belief and is compatible with even the most radical developments in cognitive science.

Scientists employ a variety of procedures to eliminate degrees of freedom from computationally and/or analytically intractable equations. In the process, they often construct new models and discover new concepts, laws and functional relations. 1 argue these procedures embody a central notion of reduction, namely, the containment of one structure within another. However, their inclusion in the philosophical concept of reduction necessitates a reevaluation of many standard assumptions about the ontological, epistemological and functional features of a reduction. On the basis of the reevaluation, 1 advocate a continuum of reduction which proceeds from the eliminative to the constructive. The metaphysical aspects of theory use in constructive reductions are sketched.

One of our more fundamental beliefs is that causal chains are continuous in time: we believe that every influence from the past upon the future runs through the present. I argue that this tenet, given certain data, can force conceptual changes upon us. I attempt to formulate a heuristic for discovery, based as explicitly as possible upon this tenet, and illustrate it by means of several examples, one of which is Mendel's discovery of genes.

P. Kitcher's unification theory of explanation appears to endorse a reductionistic view of scientific explanation that is inconsistant with scientific practice. In this paper, I argue that this appearance is illusory. The existence of multiply realizable generalizations enable the unification theory to also count many high-level accounts as explanatory.

I argue that a certain version of physicalism, which is viewed by both its admirers and its detractors as non-reductionist, in fact entails two claims which, though not reductionist in the currently most popular sense of ‘reductionist’, conform to the spirit of reductionism sufficiently closely to compromise its claim to be a comprehensively non-reductionist version of physicalism. Putatively non-reductionist versions of physicalism in general, I suggest, are likely to be non-reductionist only in some senses, but not in others, and hence to disappoint those who wish to be physicalists but still to remain soft and cuddly non-reductionists.

It is shown that it is impossible for an observer to distinguish all present states of a system in which he or she is contained, irrespective of whether this system is a classical or a quantum mechanical one and irrespective of whether the time evolution is deterministic or stochastic. As a corollary, this implies that it is impossible for an observer to measure the EPR-correlations between himself or herself and an outside system. Implications of the main result are discussed for how we have to conceive of universally valid theories.

1. Introduction. David Malament (1995) has described a natural and satisfying resolution of the traditional problems of Newtonian cosmology—natural in the sense that it effects the escape by altering Newtonian gravitation theory in a way that leaves its observational predictions completely unaffected. I am in full agreement with his approach. There is one part of his account, however, over which Malament has been excessively modest. The resolution requires a modification to Newtonian gravitation theory. Malament presents the modification as so straightforward as to be automatic. This trivializes the crucial postulate, which I shall call the “relativity of acceleration.” It is a significant physical statement in its own right and requires careful justification. Moreover the postulate proved easy to overlook for decades of discussion of the paradox. It really only becomes natural from the perspective of the newer geometric methods Malament exploits. There the postulate has become a commonplace. My purpose here is to develop the following:

• While Newtonian cosmology can be repaired satisfactorily, in its traditional form it remains deeply troubled. These troubles can be expressed most vividly as the paradoxical contradictions indicated below. They persist in both the integral and differential formulations of Newtonian gravitation theory. (Section 2)

• Malament's careful geometric treatment is necessarily dense. By taking some liberties with precision, his core result can be expressed in a far simpler form. (Section 4)

• Attempts to avoid the resolution Malament describes do lead to disaster. Therefore this episode can be inverted and used as the strongest extant argument for the relativity of acceleration in Newtonian gravitation theory. (Section 5)

Given the personalist's latitudinarian conception of rationality, what is progress toward wisdom? An answer is in C. I. Lewis's concept of the “congruence” of propositions, propositions so related that the antecedent probability of any one of them will be increased if the remainder can be assumed. This effect can be modelled in the probability calculus with due attention to the temporal sequencing of our learning of contingent propositions without ever becoming certain of them, as Jeffrey proposes. A diachronic bootstrapping effect is obtained for Ockham's razor and premises of arguments about a god's existence. As a theory's probability rises with increased evidence, the probability of our earlier evidence rises too.

D. Miller's demonstrations of the language dependence of truthlikeness raise a profound problem for the claim that scientific progress is objective. In two recent papers (Barnes 1990, 1991) I argue that the objectivity of progress may be grounded on the claim that the aim of science is not merely truth but knowledge; progress thus construed is objective in an epistemic sense. In this paper I construct a new solution to Miller's problem grounded on the notion of “approximate causal explanation” which allows for linguistically invariant progress outside an epistemic context. I suggest that the notion of “approximate causal explanation” provides the resources for a more robust theory of progress than that provided by the notion of “approximate truth.”

I discuss some aspects of the epistemological distinction between laws of nature and accidental uniformities. In order that the exposition be self-contained I briefly provide a taxonomy proposed in another work for statements that appear in a scientific theory. Once this taxonomy has been presented I attempt to prove two very different types of accidental uniformities: hard and soft. The distinction is fundamental because the latter have frequently been confused with laws of nature. I try to justify why I believe that these statements are accidental uniformities and not laws of nature.

1. Introduction. One major concern in cognitive psychology is explaining cognitively motivated behavior. However, attempts to specify the nature of the behavior that psychology is to explain have proved to be somewhat controversial. The aim of this paper is to conduct a preliminary investigation into the kinds of behavior psychology is (or should be) concerned with.

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.

Despite influencing the social sciences since the 1930s, S. S. Stevens' “operationist” philosophy of science has yet to be adequately understood. I reconstruct Stevens' operationism from his early work and assess the influence of various views (logical positivism, behaviorism and the “operational viewpoint” of P. W. Bridgman, among others) on Stevens. Stevens' operationism emerges, on my reconstruction, as a naturalistic methodological directive aimed at agreement, founded in turn on the belief that agreement is constitutive of science, the scientific community, and objectivity. Further, I show that operationism is historically and philosophically independent of the views mentioned above.

This article provides an overview of the historical and philosophical context from which originated G. Bachelard's concept of “phenomenotechnique”. It analyzes why phenomenotechnique is crucial for science studies. By incorporating the concept of phenomenotechnique into Hacking's and Galison's models of science, I argue that we can avoid the radicalism of both while also preventing the analysis of scientific practices from collapsing into the interpretive frames mandated by social constructivists.

Despite a vast philosophical literature on the epistemology of mathematics and much speculation about how, in principle, knowledge of this domain is possible, little attention has been paid to the psychological findings and theories concerning the acquisition, comprehension and use of mathematical knowledge. This contrasts sharply with recent philosophical work on language where comparable issues and problems arise. One topic that is the center of debate in the study of mathematical cognition is the question of innateness. This paper critically examines the controversy.

An amended bootstrapping can avoid Christensen's counterexamples. Earman and Edidin argue that Christensen's examples to bootstrapping rely on his failure to analyze background knowledge. I add an additional condition to bootstrapping that is motivated by Glymour's remarks on variety of evidence. I argue that it avoids the problems that the examples raise. I defend the modification against the charge that it is holistic, and that it collapses into Bayesianism.

As Laudan and Fine show, and Boyd concedes, the attempt to infer the truth of scientific realism from the fact that it putatively provides the best explanation of the instrumental success of science is circular, since what is to be shown is precisely the legitimacy of such abductive inferences. Hacking's “experimental argument for scientific realism about entities” is one of the few arguments for scientific realism that purports to avoid this circularity. We argue that Hacking's argument is as dependent on inference to the best explanation (IBE), and therefore as weak, as the other realist arguments.

This paper investigates the adequacy of evidential relevance relations proposed by Glymour and others. These accounts incorporate, as a necessary condition, what I call the Positive Instance Condition (PIC): the evidence statement and auxiliary assumptions entail a “positive instance” of the hypothesis. I argue that any account which incorporates PIC as a necessary condition while allowing “bootstrap testing” is doomed to fail. A nonbootstrapping evidential relevance relation of similar form is proposed, and it is argued that, in addition to avoiding published counter examples, this new relation meets two general requirements which, if not met, would undermine the ability of any account that incorporates PIC to accord with our intuitions of evidential relevance.

I argue that Earman and Norton's familiar “hole argument” raises questions as to whether GTR is a deterministic theory only given a certain assumption about determinism: namely, that to ask whether a theory is deterministic is to ask about the physical situations described by the theory. I think this is a mistake: whether a theory is deterministic is a question about what sentences can be proved within the theory. I show what these sentences look like: for interesting theories, a harmless bit of infinitary logic puts in an appearance.