There is currently no viable alternative to the Bayesian analysis of scientific inference, yet the available versions of Bayesianism fail to do justice to several aspects of the testing and confirmation of scientific hypotheses. Bayes or Bust? provides the first balanced treatment of the complex set of issues involved in this nagging conundrum in the philosophy of science. Both Bayesians and anti-Bayesians will find a wealth of new insights on topics ranging from Bayes’s original paper to contemporary formal learning theory.In (...) a paper published posthumously in 1763, the Reverend Thomas Bayes made a seminal contribution to the understanding of "analogical or inductive reasoning." Building on his insights, modem Bayesians have developed an account of scientific inference that has attracted numerous champions as well as numerous detractors. Earman argues that Bayesianism provides the best hope for a comprehensive and unified account of scientific inference, yet the presently available versions of Bayesianisin fail to do justice to several aspects of the testing and confirming of scientific theories and hypotheses. By focusing on the need for a resolution to this impasse, Earman sharpens the issues on which a resolution turns. John Earman is Professor of History and Philosophy of Science at the University of Pittsburgh. (shrink)

Science Without Numbers caused a stir in 1980, with its bold nominalist approach to the philosophy of mathematics and science. It has been unavailable for twenty years and is now reissued in a revised edition with a substantial new preface presenting the author's current views and responses to the issues raised in subsequent debate.

This paper criticizes the attempt to found the epistemological doctrine that all theories are evidentially underdetermined on the thesis that all theories have empirically equivalent rivals. The criticisms focus on the role of auxiliary hypotheses in prediction. It is argued, in particular, that if auxiliaries are underdetermined, then the thesis of empirical equivalence is undecidable. The inference from empirical equivalence to the underdetermination of total theories would seem to survive the criticisms, because total theories do not require auxiliaries to yield (...) observational consequences. It is shown that, nevertheless, underdetermination cannot be established for total theories. (shrink)

The instrumentalist argument from the underdetermination of theories by data runs as follows: (1) every theory has empirically equivalent rivals; (2) the only warrant for believing one theory over another is its possession of a greater measure of empirical virtue; (3) therefore belief in any theory is arbitrary. In this paper, I examine the status of the first premise. Several arguments against the universal availability of empirically equivalent theoretical rivals are criticized, and four algorithms for producing empirically equivalent rivals are (...) defended. I conclude that the case for the first premise of the argument from underdetermination is very strong. The disposition of the argument itself depends on the fate of the second premise. (shrink)

The central contention of this book is that second-order logic has a central role to play in laying the foundations of mathematics. In order to develop the argument fully, the author presents a detailed description of higher-order logic, including a comprehensive discussion of its semantics. He goes on to demonstrate the prevalence of second-order concepts in mathematics and the extent to which mathematical ideas can be formulated in higher-order logic. He also shows how first-order languages are often insufficient to codify (...) many concepts in contemporary mathematics, and thus that both first- and higher-order logics are needed to fully reflect current work. Throughout, the emphasis is on discussing the associated philosophical and historical issues and the implications they have for foundational studies. For the most part, the author assumes little more than a familiarity with logic comparable to that provided in a beginning graduate course which includes the incompleteness of arithmetic and the Lowenheim-Skolem theorems. All those concerned with the foundations of mathematics will find this a thought-provoking discussion of some of the central issues in the field today. (shrink)

One construal of convergent realism is that for each clear question, scientific inquiry eventually answers it. In this paper we adapt the techniques of formal learning theory to determine in a precise manner the circumstances under which this ideal is achievable. In particular, we define two criteria of convergence to the truth on the basis of evidence. The first, which we call EA convergence, demands that the theorist converge to the complete truth "all at once". The second, which we call (...) AE convergence, demands only that for every sentence in the theorist's language, there is a time at which the theorist settles the status of the sentence. The relative difficulties of these criteria are compared for effective and ineffective agents. We then examine in detail how the enrichment of an agent's hypothesis language makes the task of converging to the truth more difficult. In particular, we parametrize first-order languages by predicate and function symbol arity, presence or absence of identity, and quantifier prefix complexity. For nearly each choice of values of these parameters, we determine the senses in which effective and ineffective agents can converge to the complete truth on an arbitrary structure for the language. Finally, we sketch directions in which our learning theoretic setting can be generalized or made more realistic. (shrink)

The underdetermination of theory by evidence must be distinguished from holism. The latter is a doctrine about the testing of scientific hypotheses; the former is a thesis about empirically adequate logically incompatible global theories or "systems of the world". The distinction is crucial for an adequate assessment of the underdetermination thesis. The paper shows how some treatments of underdetermination are vitiated by failure to observe this distinction, and identifies some necessary conditions for the existence of multiple empirically equivalent global theories. (...) We consider how empiricists should respond to the possibility of such systems of the world. (shrink)

This book is devoted to a different proposal--that the logical structure of the scientist's method should guarantee eventual arrival at the truth given the scientist's background assumptions.

Constructive Empiricism, the view introduced in The Scientific Image, is a view of science, an answer to the question “what is science?” Arthur Fine’s and Paul Teller’s contributions to this symposium challenge especially two key ideas required to formu- late that view, namely the observable/unobservable and accept- ance/belief distinctions. I wish to thank them not only for their insightful critique but also for the support they include. For they illuminate and counter some misunderstandings of Constructive Empiricism along the way. That (...) leaves me free to focus on those two main challenges. The three of us share a good deal of common history. So it is perhaps only remarkable, and not astonishing, that we now share a common leaning to Pragmatism in philosophy. Of us three I am clearly themost conservative in this respect, especially as pertaining to truth, reference, and belief.2 Arthur Fine showed very nicely how Constructive Empiricism could have been conceived under the canopy of Dewey’s Instrumentalism. Much of it could appear as a Corollary to that sort of Instrumentalism, I agree. But in fact Iwould not be happy to land in that general Pragmatist position. (shrink)

then essentially characterized the hypotheses that mechanical scientists can successfully decide in the limit in terms of arithmetic complexity. These ideas were developed still further by Peter Kugel [4]. In this paper, I extend this approach to obtain characterizations of identification in the limit, identification with bounded mind-changes, and identification in the short run, both for computers and for ideal agents with unbounded computational abilities. The characterization of identification with n mind-changes entails, as a corollary, an exact arithmetic characterization of (...) Putnam's n-trial predicates, which closes a gap of a factor of two in Putnam's original characterization [12]. (shrink)