David W. Harker

Accommodation and Prediction: The Case of the Persistent Head

A not unpopular thesis, when it comes to the confirmation of scientific theories, is that data which were used in the construction of a theory afford poorer support for that theory than data that played no role. Some compelling thought experiments have been offered in favour of this view, not as proof but rather to add some intuitive plausibility. In this paper I consider such thought experiments and argue that they do not support the thesis; the perceived importance of prediction over accommodation, at least in these cases, is illusory. 1. Introduction2. Background assumptions3. Strong thesis4. Weak thesis5. Conclusions Introduction Background assumptions Strong thesis Weak thesis Conclusions

ERIC CHRISTIAN BARNESThe Paradox of Predictivism

If a scientific theory is constructed with the explicit intention of accommodating certain data, there is some temptation to regard the resulting theory as receiving little if any confirmational support from those same data. The intentions of the theorist, it would thus appear, are relevant to our evaluations of theories. Yet we seem equally tempted to suppose that theories are confirmed to a greater or lesser extent by the available data alone; theorists’ intentions concerning what data a theory was designed to accommodate are, given this intuition, irrelevant to matters of confirmation. We have an apparent tension. Suppose we define the novel predictions of a theory as those phenomena which are entailed by the theory but were not used in its construction. Our dilemma can now be couched in terms of the relative confirmatory importance of novel predictions. Other things being equal, either the verified novel predictions of a theory do or do not provide more compelling support in favour of a theory than other types of evidence. The reasons outlined above persuade Eric Barnes to describe the situation as paradoxical, a paradox that provides the title for his book and the motivation for some of its principal themes. Although concerned foremost with the nature, role and status of predictive success, Barnes also ranges over aspects of the realism/antirealism debate, the problem of old evidence as it is usually associated with Bayesian accounts of confirmation, and, as illustration of the superiority of his own predictivist account, a nice analysis of Mendeleev’s verified predictions that were based on his periodic table. A central theme of the book is that philosophers of science have too often assumed epistemic individualism, a view that assumes rational agents evaluate theories without attending to others’ judgments concerning those theories. Barnes embraces, defends and relates to predictivism a Brit. J. Phil. Sci. 62 (2011), 219–223

A Likely Explanation: IBE as a Guide to Better (but not more probable) Hypotheses

Abstract Several friends of inference to best explanation have claimed in recent work that explanatory virtues, such as consilience, simplicity and increased precision, play an important heuristic role in assigning probabilities to available hypotheses and that it is this role that justifies continued efforts to investigate the scope, nature and epistemic value of the inference rule. In this paper I argue that understanding explanatory virtues as a guide to probability assignments creates a critical dilemma for advocates of IBE that has not previously been made sufficiently explicit and which has significant implications for the prospects of the rule. I conclude that the viability of IBE requires that explanatory virtues be related to a non-probabilistic conception of success.