On Miracles: A Defence of Explanationist Realism
Item statusRestricted Access
“No Miracles” arguments have been a live part of the realism debate for several decades, but there has been a tendency to misrepresent or conflate them. In this thesis I identify several varieties of explanationist realism, and argue that one of them (introduced in Boyd (1973)) has the potential to overcome some of the most persistent anti-realist challenges to convergent realism. I focus on a couple of particularly enduring objections to “No Miracles Realism” (NMR); the Pessimistic Meta-Induction (PMI) and the epistemological challenges raised by van Fraassen’s alternative view of science - constructive empiricism. In response to Laudan, I demonstrate how an emphasis on methodology might permit the realist to rebut the PMI, offering a solution that both coheres with episodes in the history of scientific progress and escapes the difficulties threatening other purported solutions. In addition, I argue that a couple of promising accounts of verisimilitude have been developed in the years since Laudan’s (1981), and demonstrate how one of these in particular shows promise in addressing his “downward path”. Subsequently, I look at van Fraassen’s numerous objections to explanationist realism. These include the Darwinian explanation for the success of science, several complaints against Inference to the Best Explanation (IBE), and a general challenge to the form of Boyd’s programme. I argue that, in each instance, the realist offers a more plausible account of scientific practice and its success, and that the constructive empiricist struggles to provide a descriptively adequate account of such practice and/or its success. Ultimately, I consider whether the different priorities of the realist and the constructive empiricist can be illuminated by the voluntarist background that informs constructive empiricism, and how this might impact upon my arguments. I conclude that, irrespective of one’s epistemological preference (for naturalism or voluntarism), the work undertaken here should be considered significant.