Interpretation and Objectivity

Velocity, we know, is frame relative. That is no threat to its objectivity, for we can agree on the velocity of a body relative to a given frame, whatever frames we ourselves variously occupy. Nor is objectivity in science threatened by the proliferation of competing research programmes, so long as there is intertheoretic agreement that a certain outcome corroborates my theory to this (possibly negative) degree and yours to that. But if the incommensurability thesis is correct and theory infects our terms all the way down, the objectivity of science is under threat. For those divided by theory have no common vocabulary in which to describe experimental outcomes, and no way to agree about the relative confirmation of the theories that divide them. If we allow ourselves a generalized notion of point of view that subsumes spatiotemporal, theoretical and other perspectives, our two cases suggest objectivity as the invariance of judgement across points of view-under suitably standardized conditions of judgement of course. ' If we grant the incommensurability theorist's argument, corroboration fails the test of objectivity. Velocity passes, not withstanding its relativity to the very thing-frame of reference thatjudgements of (frame-relative) velocity are said to be invariant across.2 Indeed, by relativizing we may transform a highly non-objective notion into an objective one, though what we get is not always very interesting. Unrelativized judgements of taste are notoriously variable across points of view, but we can all agree that rancid yak butter is deliciousfor Fred. Deliciousness, as we suspected, is not objective; deliciousness for Fred is-if anyone cares. My question is whether objectivity as invariance of judgement has application within the discipline of interpretation.3 I conclude that its prospects are poor, even after liberal applications of the relativization strategy. After clarifying the scope and method of the project a little, I start with a theory about what interpretation is, and about the appropriate method for the interpreter to adopt. That package I call interpretive instrumentalism, and it is from the per-