Stuctural realism versus entity realism: a critical evaluation
The two partial scientific realist positions, called structural realism and entity realism make conflicting claims about our knowledge ofthe unobservable world. They appear to be mutually exclusive and create a tension in the scientific realist camp. Tn this dissertation I have attempted to reconcile the two as follows. In the first chapter, I state the problem and set the objectives of the study. The second chapter gives an overview of the scientific realism debate, focusing on the epistemological dimension. I bring out the challenges facing scientific realism. The third chapter surveys the development of structural realism and explains the stance put forward by John Worrall in defense of scientific realism. I examine his arguments and show that they are valid only where the structural part of a theory can be expressed mathematically in classical physics. Worrall has appealed to history of science to argue that we cannot know the unobservable theoretical entities. I utilize the same case-study from optics to argue that the unobservable entity ether was never an accepted and empirically adequate theoretical entity. The fourth chapter presents the arguments put forward by two entity realists - Nancy Cartwright and Ian Hacking. They have argued that the theoretical entities, which cause a phenomenon, exist; and by establishing the causal link, we can know them. Further, the entities, which can be interfered-with and manipulated, can be known and this knowledge is independent of the theories. I have argued that knowledge of unobservable causes is parasitic on theories putating them. Entity realists' arguments are valid only where causal links can be theorized and then established experimentally. Interference and manipulation reveal certain aspects of the nature of : the entities and are not possible for all types of unobservable entities. The fifth chapter brings out the tension created by structural realism and entity realism. After analyzing their arguments, I show that they do not apply to the whole enterprise of science. In some areas of the enterprise, beliefs about unobservables are justified by mathematization whereas in some other areas by manipulation and by showing causal connection. Thus the two schools of thought do not contradict each other and can coexist. The tension created by the two partial realist positions, is thus shown to be a pseudo- problem and dissolved.