We generalize the Kolmogorov axioms for probability calculus to obtain conditions defining, for any given logic, a class of probability functions relative to that logic, coinciding with the standard probability functions in the special case of classical logic but allowing consideration of other classes of "essentially Kolmogorovian" probability functions relative to other logics. We take a broad view of the Bayesian approach as dictating inter alia that from the perspective of a given logic, rational degrees of belief are those representable (...) by probability functions from the class appropriate to that logic. Classical Bayesianism, which fixes the logic as classical logic, is only one version of this general approach. Another, which we call Intuitionistic Bayesianism, selects intuitionistic logic as the preferred logic and the associated class of probability functions as the right class of candidate representions of epistemic states (rational allocations of degrees of belief). Various objections to classical Bayesianism are, we argue, best met by passing to intuitionistic Bayesianism—in which the probability functions are taken relative to intuitionistic logic—rather than by adopting a radically non-Kolmogorovian, for example, nonadditive, conception of (or substitute for) probability functions, in spite of the popularity of the latter response among those who have raised these objections. The interest of intuitionistic Bayesianism is further enhanced by the availability of a Dutch Book argument justifying the selection of intuitionistic probability functions as guides to rational betting behavior when due consideration is paid to the fact that bets are settled only when/if the outcome bet on becomes known. (shrink)

The fact that simplicity has been linked with induction by many philosophers of science, some of whom have proposed or supported criteria of “inductive simplicity,” means that the problem must be given some serious attention. I take “inductive simplicity” as a title, however, only by way of concession to these historical treatments, since it is precisely the burden of my paper to show that there is no such thing. So much for the conclusion. I shall spend the remainder of my (...) time arguing for it. (shrink)

Kolmogorov''s axiomatization of probability includes the familiarratio formula for conditional probability: 0).$$ " align="middle" border="0">.

With clarity and grace, he also reveals the limited truth of some of the "common sense" assumptions which make it difficult for us to appreciate its full ...

In these inspiring lectures David Bohm explores Albert Einstein’s celebrated _Theory of Relativity_ that transformed forever the way we think about time and space. Yet for Bohm the implications of the theory were far more revolutionary both in scope and impact even than this. Stepping back from dense theoretical and scientific detail in this eye-opening work, Bohm describes how the notion of relativity strikes at the heart of our very conception of the universe, regardless of whether we are physicists or (...) philosophers. (shrink)

A common view among statisticians is that convergence (which statisticians call consistency) is a necessary property of an inference rule or estimator. In this paper, this view is challenged by appeal to an example in which a rule of inference has a likelihood rationale but is not convergent. The example helps clarify the significance of the likelihood concept in statistical inference.

There were already confusions in the Middle Ages with the reading of Aristotle on negative terms, and removing these confusions shows that the four traditional Syllogistic forms of statement can be readily generalised not only to handle polyadic relations (for long a source of difficulty), but even other, more measured quantifiers than just ‘all’, ‘some’, and ‘no’. But these historic confusions merely supplement the main confusions, which arose in more modern times, regarding the logic of singular statements. These main confusions (...) originate in the inability of the mainline modern tradition to supply the ‘logically proper names’ which alone have the right to replace individual variables; an inability which has resulted in the widespread, but erroneous replacement of individual variables with ordinary proper names, i.e. names for contingent beings, in many if not most contemporary logic texts. The paper includes the exhibition and grammatical characterisation of the logically proper names that are required instead, specifying just how they differ syntactically from ordinary proper names. It also shows how ontologically significant is the distinction, since not only do logically proper names refer to necessarily existent objects (showing there are no ‘empty domains’ for Classical Logic to fail to apply to), but also thereby central features of Realism become considerably clarified. (shrink)

There is a long tradition in formal epistemology and in the psychology of reasoning to investigate indicative conditionals. In psychology, the propositional calculus was taken for granted to be the normative standard of reference. Experimental tasks, evaluation of the participants’ responses and psychological model building, were inspired by the semantics of the material conditional. Recent empirical work on indicative conditionals focuses on uncertainty. Consequently, the normative standard of reference has changed. I argue why neither logic nor standard probability theory provide (...) appropriate rationality norms for uncertain conditionals. I advocate coherence based probability logic as an appropriate framework for investigating uncertain conditionals. Detailed proofs of the probabilistic non-informativeness of a paradox of the material conditional illustrate the approach from a formal point of view. I survey selected data on human reasoning about uncertain conditionals which additionally support the plausibility of the approach from an empirical point of view. (shrink)

Aristotle draws two sets of distinctions in Metaphysics 9.2, first between non-rational and rational capacities, and second between one way and two way capacities. He then argues for three claims: [A] if a capacity is rational, then it is a two way capacity [B] if a capacity is non-rational, then it is a one way capacity [C] a two way capacity is not indifferently related to the opposed outcomes to which it can give rise I provide explanations of Aristotle's terminology, (...) and of how [A]-[C] should be understood. I then offer a set of arguments which are intended to show that the Aristotelian claims are plausible. \\\ [Nicholas Denyer] In De Caelo 1: 11-12 Aristotle argued that whatever is and always will be true is necessarily true. His argument works, once we grant him the highly plausible principle that if something is true, then it can be false if and only if it can come to be false. For example, assume it true that the sun is and always will be hot. No proposition of this form can ever come to be false. Hence this proposition cannot be false. Hence it is necessarily true, and so too is anything that follows from it. In particular, it is necessarily true that the sun is hot. Moreover, if the sun not only is and always will be hot, but also always has been, then it follows by similar reasoning that the sun not only cannot now fail to be hot, but also never could have failed. Anything everlastingly true is therefore, in the strictest sense of the term, necessarily true. (shrink)

Various formulations of the principle of simplicity in science are examined and rejected in favor of Goodman's proposal, the essence of which is to concentrate attention upon the predicates that form the extralogical basis of any given theory and to provide measures for comparing the relative structural simplicity of different sets of such predicates. The postulational basis of Goodman's method is set out and explained, together with some important amendments and additions, and a number of theorems are proved, with whose (...) aid the simplest theory to account for a certain corpus of scientific phenomena is readily determinable. (shrink)

Confirmation functions are generally thought of as probability functions. The well known difficulties associated with the probabilistic confirmation functions proposed to date indicate that functions other than probability functions should be investigated for the purpose of developing an adequate basis for confirmation theory. This paper deals with one such function, the likelihood function. First, it is argued here that likelihood is not a probability function. Second, a proof is given that, in the limit, likelihood can be used to determine which (...) of two observationally distinct hypotheses is true. Finally, a demonstration is given that, in the presence of a finite amount of experimental information, likelihood can serve as a good estimator of truth. (shrink)