UC Berkeley

Many of us think that truth is prior to science: though it belongs to the nature of science to pursue truth, the nature of truth does not involve science. Frege denies this. He thinks that to be true is to play a role in the characteristic goals of science, such as explanation.

For Frege and the Neo-Kantian philosophers in his philosophical milieu, the point of this view is that no truth is trivial: every one makes a contribution to a deep understanding of the world. This allows them to maintain that truth is a value, whose significance is rivalled only by that of the Good and the Beautiful. This view leads Frege to a normative conception of logic, on which the logical laws are the ones that tell us how we ought to judge if we are to achieve our highest scientific goals. Since the nature of truth itself is to be found in such goals, these laws tell us about truth itself.

This makes possible an epistemology for logical axioms that has frustrated readers by its apparent inconsistency: Frege seems both

to rule out arguing for logical axioms, and himself to offer arguments for them. But he only means to rule out arguments that derive the axioms from other truths that we are already justified in accepting. His own arguments, by contrast, derive them from a goal: they show us that the axioms must be true if that goal is to be reached. The goal in question is the construction of a logical system; but because such a system is needed in every science, the goal is none other than truth itself.

Frege's discussions of truth and logic are conducted independently of any claims about language, which sits badly with his reputation as the herald of the ``linguistic turn,'' who proclaims the philosophy of language to be the foundation of all philosophy. This calls for a re-evaluation of the role of language in Frege's philosophy. I argue that Frege's philosophy of language is a grand technological project in linguistic engineering: he does not aim to describe language, but to change it so that we can reach the truth more effectively. Reading Frege this way helps us understand the most puzzling passages in his early work, and, I hope, points us to a resolution of the outstanding puzzles in his later philosophy of language as well.