During the past decade there has been a very effective campaign against any explanation of remembering whose basic concept is that of a causally mediating trace. This paper attempts to provide such an explanation by presenting an explicit deductive argument for the existence of the memory trace. The conclusion is shown to follow from reasonable, empirical assumptions of which the most interesting is a spatio-temporal contiguity thesis. Set-theoretic techniques are used to provide a framework of analysis and probabilistic definitions of some causal notions, as that of a causal chain, are presented.

A recent version of the causal theory of time makes crucial use of a concept of the genidentity of events when it attempts to define temporal betweenness in terms of empirical, physical properties. By presenting and discussing an apparent counterexample it is argued that the role of genidentity in an empirical theory of time is problematic. In particular, it may be that the temporal behavior of objects is used to decide which events are genidentical, and, if so, the definition of temporal betweenness is circular. On the other hand, though there are strategies for avoiding the charge of circularity, in certain hypothetical situations the definition may yield inconsistent temporal orders. Then, the definition would have to be supplemented by a choice of temporal orders, and this choice may introduce an element of conventionality into the causal theory of time.

It is argued that a better theory neither (I) proves better at enabling us to realize our goals, nor (II) enables us to make more accurate predictions than a worse theory. (I) fails because it, tacitly, erroneously assumes, in talking of our goals, that individual preferences for theories can be aggregated into a social preference ordering; (II) fails because it cannot distinguish between important and unimportant predictions. Neither of these failures can be patched up by appealing to the notion of a true theory. The conclusion is that we as yet possess no adequate understanding of the relationships among theoretical progress, goal realization, and predictive accuracy.

Ramsey's device [2], for eliminating theoretical terms from a theory assumes that the theory has only finitely many axioms. Paul Berent [1], has recently suggested a modification of Ramsey's method that is designed to cover the infinite case. But, his modified method is quite unlike Ramsey's in that it involves an ascent from the original theory to a suitable metatheory. Thus, it is perhaps of some interest to note that there is an extension of Ramsey's method that covers the infinite case and at the same time is much closer in spirit to his original suggestion.

In concluding section 2, chapter VI of part II of [6], Duhem claimed:

1. (I) ... the physicist can never subject an isolated hypothesis to experimental test, but only a whole group of hypotheses ... ([6], p. 187)

2. (II) ... when the experiment is in disagreement with his predictions, what he learns is that at least one of the hypotheses constituting this group is unacceptable and ought to be modified; but the experiment does not designate which one should be changed'. ([6], p. 187)

According to our intuitions, certain generalizations are better confirmed by positive instances than others. In order to characterize the difference between these generalizations, I have proposed in [3] to investigate the generalizing behavior of living beings. Such an investigation makes it possible to classify into different categories the generalizations that are intuitively confirmed by their positive instances and those that are not intuitively confirmed by such instances. One important aspect of my treatment, however, has been shown to be unsatisfactory: sentences that are logically equivalent may belong to different categories. In this paper, I shall first give a short summary of the relevant parts of my previous paper, and then I shall try to improve it in order to avoid the undesirable result.

The works of Paul K. Feyerabend, Norwood Russell Hanson and Thomas S. Kuhn have come to occupy a central place in the annals of contemporary philosophy of science. Some of their contemporaries, ([1], [14], [20]), tend to regard them as the vanguard of a new “revolutionary” intellectual movement. Reacting against the views of their positivist predecessors, they embrace and propagate the idea that “pervasive presuppositions” are fundamental to scientific investigations. Thus, Feyerabend thinks that, “... scientific theories are ways of looking at the world; and their adoption affects our general beliefs and expectations, and thereby also our experiences and our conception of reality” ([5], p. 29). This is in stark contrast to the positivist view that the aim of science is the systematization of experience that exists independently of any scientific theories ([12], p. 178). This new view of scientific theories also involves a “radical” conception of the nature of theoretical change. Rejecting the positivist notion of any constant element (experience, observation language, etc.) through such change, Kuhn regards a basic theoretical change as a conceptual revolution and “wants to say that after a revolution scientists are responding to a different world” ([14], p. 110). Hanson uses the phrase ‘theory-loaded’ to give expression to a view of the semantic content of observation statements that follows from this general position ([11], pp. 19 ff.).

Rom Harré's proposed Copernican Revolution in the philosophy of science is a very ambitious undertaking. It challenges established views, proposes a radically new model for scientific explanation, and forces a rethinking of the foundations of the field. In his treatment of the natural sciences, Harré rejects all deductivist accounts of scientific explanation (whether these accounts are fashioned by rationalists or empiricists) basically on the grounds that such accounts seriously distort the methods of explanation actually operative in science. In the social sciences Harré, in collaboration with Secord, rejects mechanistic, positivistic, and behavioristic accounts of human behavior [5]. In place of these discarded interpretations he substitutes a new realism which stresses the role that models play in supplying accounts of observed patterns of phenomena in terms of the things and processes whose interactions are responsible for the manifest phenomena. This revolution, if successfully carried out, would indeed be a valuable contribution to the philosophy of science.

By observing a black swan, one ascertains an interesting zoological fact. The fact involves two items, a specific object and the property of being a black swan, and consists in the former instantiating the latter. The object is in itself of little interest. Our environment abounds in objects, and knowledge is hardly enhanced by one of them being merely picked out for consideration. The same goes for the property. Properties are at least as plentiful as objects and one does not get to know anything by merely focussing attention upon one of them. What is of interest is the combination of the two items, i.e. the circumstance that the object instantiates the property. And this is what the said fact consists in.