Re-Thinking Scientific Realism: Structure and Beyond

This thesis examines the contemporary scientific realism debate, with a special focus on the various forms of structural realism. It comprises three parts. The first part prefaces the work with a scrutiny of the principal arguments of the classic debate. The global and local explanationist arguments are critically analysed, and juxtaposed with the so-called experimental arguments for realism. It is argued that focusing on explanatory considerations does not serve the justificatory task the realist faces, but a local analysis of ampliative reasoning can nevertheless secure a level of realist commitment to a significant class of unobservables. This part also defends the anti-realist argument of Pessimistic Induction against two challenges that take it to be fallacious. The second part looks at the main structural realist suggestions for an image of theoretical knowledge that harmonises with our best understanding of the current and past science. It concludes that both of these—epistemic Ramsey sentence realism in the syntactic-axiomatic framework, and the ontological structural realism in the semantic framework—are incomplete and inadequate responses to the anti-realist challenge. In addition to providing a comparative review of the various structuralist motivations and intuitions, this part contributes to the literature by clarifying the often referred to, but also by and large misunderstood, problem of unintended models faced by the Ramseyfying realist (‘Newman’s problem’). The third part begins by looking at the notion of approximate truth in detail, in order to argue that the traditional ‘standard’ realist alternative to structural realism is not the only alternative. Paying due attention to the explanatory requirements for the realist argument yields an informally articulated notion of explanatory approximate truth which gives rise to a fresh realist perspective: eclectic realism is realism about success-fuelling multiply realised properties. This part concludes the thesis by developing and defending this position, by conducting a detailed case study of the historical theory change from Fresnel’s ether to Maxwell’s electromagnetic theory of optics.