Abstract
Three lines of argument are employed to show that Glymour's position on the epistemology of geometry is probably not as strong theoretically as the position of the underdeterminists whom he attempts to refute. The first argument centers on Glymour's implicit use of a realist position on intertheoretic reference, similar to that employed by Boyd and other realists. Citations are made to various portions of Glymour's work, and the relationship between the imputed theory of reference and Glymour's position spelled out. The second line of argument refutes Glymour's contention that his criteria for choosing among theories are strongly based and not a priori, and a third line of argument rephrases Glymour's original position in a way which more clearly shows its implicit base and manifests as well the vagueness from which the formulation of the position suffers. It is concluded that a thorough examination of Glymour's deoccamization yields conclusions similar to those of the underdeterminists. * This paper is the result of work begun in a seminar on Philosophy of Science held at Rutgers University in the late 1970's. Work was continued at the University of California, Santa Barbara under the generosity of Grant DOE (OSERS) G0083/03651. I would like to thank colleagues from both institutions, and I owe a special debt of gratitude to an anonymous reviewer for the journal.