Trizio, Emiliano
(2001)
The Duhem thesis, the Quine thesis and the problem of meaning holism in scientific theories.
MPhil thesis, London School of Economics and Political Science.
Abstract
Through a detailed analysis of Duhem's writings some light is cast on the relations between holism, underdetermination and theory-ladenness of experimentation. The latter, which results from the need to interpret theoretically what is actually observed during an experiment, plays a key role in Duhem's analysis of the relation between observation and theory. I will argue that the theory-ladenness of experimentation on one hand provides a general argument for the holistic character of theory testing, and on the other renders problematic the thesis that theories are underdetermined by empirical evidence. A tension is found between Duhem's claim that the aim of theory is to save the phenomena and his analysis of the interpretative role of theory in experiments. I suggest how to overcome this difficulty by showing in what sense we can say that theory saves theory-laden phenomena. After stressing the differences between the Duhemian and the Quinean variants of holism, I argue that Quine fails to take into account the importance of the theory- ladenness of experimentation and the implications of Duhem's thought: Quine shares with the Logical Empiricists the belief that it is possible to detach from theories their empirical content. His acceptance of holism has simply the effect of restricting the attribution of empirical content only to conjunctions of many theoretical statements. I analyse and criticise the two notions of empirical content that Quine has developed. Furthermore I argue that there is no general theory-free expression of the experiential implications of a theory, for theories are logically connected to observable events only within local contexts defined theoretically and brought about by the activities of experimenters. Finally I suggest that, in the light of these considerations, the implications resulting from the possibility of rival incommensurable traditions of research should be discussed, rather than Quine's dilemma concerning empirically equivalent systems of the world.
Actions (login required)
 |
Record administration - authorised staff only |
Download statistics
Download statistics