Bas C. Van Fraassen

Decades of debate about scientific realism notwithstanding, we find ourselves bemused by what different philosophers appear to think it is, exactly. Does it require any sort of belief in relation to scientific theories and, if so, what sort? Is it rather typified by a certain understanding of the rationality of such beliefs? In the following dialogue we explore these questions in hopes of clarifying some convictions about what scientific realism is, and what it could or should be. En route, we encounter some profoundly divergent conceptions of the nature of science and of philosophy.

For changing opinion, represented by an assignment of probabilities to propositions, the criterion proposed is motivated by the requirement that the assignment should have, and maintain, the possibility of matching in some appropriate sense statistical proportions in a population. This ‘tracking’ criterion implies limitations on policies for updating in response to a wide range of types of new input. Satisfying the criterion is shown equivalent to the principle that the prior must be a convex combination of the possible posteriors. Furthermore, this is equivalent to the requirement that prior expected values must fall inside the range spanned by possible posterior expected values. The tracking criterion is liberal; it allows for, but does not require, a policy such as Bayesian conditionalization, and can be offered as a general constraint on policies for managing opinion over time. Examples are given of non-Bayesian policies, both ones that satisfy and ones that violate the criterion. 1 Introduction2 Alternative Updating Policies3 Modelling the Situation for Normal Updating4 Tracking: A Criterion for Updating Policies5 Tracking: Precise Formulation and Relation to Convexity6 The Spanning Criterion7 Non-Bayesian Policies that Satisfy the Spanning and Tracking Criteria8 Policies that Violate the Spanning and Tracking Criteria AppendixAppendix

Thomason (1979/2010)'s argument against competence psychologism in semantics envisages a representation of a subject's competence as follows: he understands his own language in the sense that he can identify the semantic content of each of its sentences, which requires that the relation between expression and content be recursive. Then if the scientist constructs a theory that is meant to represent the body of the subject's beliefs, construed as assent to the content of the pertinent sentences, and that theory satisfies certain 'natural assumptions', then it implies that the subject is inconsistent if the beliefs include arithmetic. I challenge the result by insisting that the motivation for Thomason's principle (ii), via Moore's Paradox, leads to a more complex representation, in which stating the facts and expressing one's beliefs are treated differently. Certain logical connections among expressions of assent, and between expression and statement, are a matter of consequence on pain of pragmatic incoherence, not consequence on pain of classical logical inconsistency. But while this salvages the possibility that a modification of the above sort of representation could be adequate, Thomason's devastating conclusion returns if the scientist identifies himself as the subject of that representation, even when paying heed to the requirement of pragmatic coherence of the sort highlighted by Moore's Paradox

The physics and metaphysics of identity and individuality Content Type Journal Article DOI 10.1007/s11016-010-9463-7 Authors Don Howard, Department of Philosophy and Graduate Program in History and Philosophy of Science, University of Notre Dame, Notre Dame, IN 46556, USA Bas C. van Fraassen, Philosophy Department, San Francisco State University, 1600 Holloway Avenue, San Francisco, CA 94132, USA Otávio Bueno, Department of Philosophy, University of Miami, Coral Gables, FL 33124, USA Elena Castellani, Department of Philosophy, University of Florence, Via Bolognese 52, 50139 Florence, Italy Laura Crosilla, Department of Pure Mathematics, School of Mathematics, University of Leeds, Leeds, LS2 9JT UK Steven French, Department of Philosophy, University of Leeds, Leeds, UK Décio Krause, Department of Philosophy, Federal University of Santa Catarina, 88040-900 Campus Trindade, Florianópolis, SC Brazil Journal Metascience Online ISSN 1467-9981 Print ISSN 0815-0796

There are striking differences between the scientific theoretical description of the world and the way it seems to us. The consequent task of relating science to ’the world we live in’ has been a problem throughout the history of science. But have we made this an impossibility by how we formulate the problem? Some say that besides the successive world-pictures of science there is the world-picture that preceded all these and continues to exist by their side, elucidated by more humanistic philosophers. Wilfrid Sellars codified this conviction in his dichotomy of scientific image and manifest image. Others say that all our world-pictures are transient, evolve, conflict with and replace each other, undergo violent revolutions as well as periods of normal development, and may be incommensurable, allowing of no meaningful dialogue. All such formulations may themselves be tendentious metaphysics, full of false contrasts. Insistence on a radical separation between science and what we have apart from science, on the impossibility of accommodating science without surrender, may be a way of either idolizing or demonizing science rather than understanding it.

There are a number of dimensions to the realism-nominalism controversy. The topics of debate comprise: necessary connections and causality, dispositions and counterfactuals, space and time, the existence of abstract entities and mathematical objects, the existence of the theoretical entities of science. On all these except the last, Sellars takes a non-realist line: and on all these except the last, I agree with him to the extent that I presently have an opinion on them. But Sellars is a scientific realist, encapsulating this realism in the dictum: “to have good reason for holding a theory is ipso facto to have good reason for holding that the entities postulated by the theory exist”.

ABSTRACT For changing opinion, represented by an assignment of probabilities to propositions, the criterion proposed is motivated by the requirement that the assignment should have, and maintain, the possibility of matching in some appropriate sense statistical proportions in a population. This ‘tracking’ criterion implies limitations on policies for updating in response to a wide range of types of new input. Satisfying the criterion is shown equivalent to the principle that the prior must be a convex combination of the possible posteriors. Furthermore, this is equivalent to the requirement that prior expected values must fall inside the range spanned by possible posterior expected values. The tracking criterion is liberal; it allows for, but does not require, a policy such as Bayesian conditionalization, and can be offered as a general constraint on policies for managing opinion over time. Examples are given of non-Bayesian policies, both ones that satisfy and ones that violate the criterion. _1_ Introduction _2_ Alternative Updating Policies _3_ Modelling the Situation for Normal Updating _4_ Tracking: A Criterion for Updating Policies _5_ Tracking: Precise Formulation and Relation to Convexity _6_ The Spanning Criterion _7_ Non-Bayesian Policies that Satisfy the Spanning and Tracking Criteria _8_ Policies that Violate the Spanning and Tracking Criteria Appendix Appendix