School of Historical and Philosophical Studies - Theses
Permanent URI for this collection
Search Results
1 - 2 of 2
-
ItemLogical principles of scientific inference( 1972)In this essay my aim is to explicate and vindicate the method of ampliative inference which constitutes the basis of scientific and, indeed, of all critical and disciplined reasoning.
-
ItemThe incommensurability of scientific theories( 1989)Kuhn and Fegerabend argue that successive or rival scientific theories may be incommensurable due to differences in the concepts and language they employ. The terms employed by such theories are unlike in meaning, and even reference, so they may fail to be translatable from one theory into the other. Owing to such semantical differences, statements from one theory neither agree nor disagree with statements from another theory with which it is incommensurable; so the content of such theories cannot be directly compared. As against the incommensurability thesis, the view which will be defended here is that theories are comparable provided they refer to at least some of the same things. In this we follow Scheffler, who notes that statements which vary in meaning may be incompatible if their terms have common reference. But Scheffler adopts a description theory of reference, which leads to excessive referential instability in the transition between conceptually disparate theories. So we follow Putnam instead in adopting a causal theory of reference, which allows stability of reference through conceptual change. However, the causal theory of reference is problematic in its own right, and cannot fully remove the problems raised by the incommensurability thesis. It must be modified to permit the reference of a term to be fixed in more than one way and to allow the possibility of reference change. It must also grant a role to descriptions in fixing the reference of theoretical terms. So while excessive reference change is avoided by adopting a causal theory of reference, the modifications prevent it from ruling out referential variance altogether. In addition, the modified causal theory of reference supports the thesis of translation failure between theories. For it may prove impossible to fix reference within the context of a theory in the same way as the reference of terms used in another theory is fixed. However, failure to translate does not entail content incomparability, for there may be relations of co-reference despite differences in how reference is fixed. Nor does it entail failure to communicate, for the meaning of a term may be understood even if the term cannot be translated into the specific language of a particular theory. The idea of translation failure between theories has been the subject of penetrating criticism by Putnam and Davidson, who argue that the very idea of an untranslatable language is incoherent. We will here defend the notion of translation failure against their arguments. The key elements of this defence are the points that understanding is independent of translation, and that the untranslatability in question is a limited translation failure between theoretical sub-languages within an encompassing background language. At times, the differences between incommensurable theories seem ontological, rather than merely semantical. There is often a hint of the idealist thesis that the world referred to by a theory depends upon the theory itself. It will be shown, however, that the incommensurability thesis is not an idealist rejection of the reality independent of theory. Weaker "constructivist" forms of idealism which grant the existence of a reality independent of theory but take the world referred to by a theory to be a construction will also be criticised.