Theory-Change And The Logic Of Enquiry: New Bearings In Philosophy Of Science

Review of Metaphysics, The, Sept, 1999 by Christopher Norris

"It seems to me," Marcus writes, "that much of the discussion these past few years concerning the apparent breakdowns of substitutivity principles in intensional contexts and its presumably devastating results for logic and mathematics are [sic] largely terminological."(10) And again, more pointedly: "the apparent difficulties of interpreting such [modal/intensional] systems are not genuine, but analogous to a rejection of non-Euclidean geometry because it allows parallel lines to meet."(11) As it happens both Husserl and Bachelard are much concerned with the processes of thought by which advances come about in fields such as mathematics, geometry, and physics, advances that can only be achieved--so they argue--through a break with hitherto accepted ideas of self-evident, commonsense, or a priori truth. What is at issue here, as Marcus brings out very clearly, is the question how far such processes of thought should figure in a properly logical account of the context of justification. For some analytic philosophers, Quine preeminent among them, there is just no place for intensionalist (let alone intentionalist) modes of talk when trying to make logical sense of such matters. For others, including Marcus, there is a firm line to be drawn between intensional approaches (which are strictly indispensable if one wishes to deploy modal, counterfactual, or other such causal-explanatory devices) and intentional approaches which supposedly open the way to all manner of psychologistic vagaries. Thus, on her account, logic can perfectly well accommodate most of the items that Quine objects to ("knows that," "is aware that," and in particular "is necessary that") without fear of referential opacity. However it requires that those qualifying phrases be properly construed, that is to say, in accordance with a modal-logical or strictly intensionalist approach which finds no place for such inscrutable matters as the discovery procedures or the processes of thought by which conjectural beliefs are transformed into adequate knowledge.

II

It is this dimension of enquiry that was firmly closed off by the analytic turn against Husserlian phenomenology and indeed against all those later continental developments perceived as not matching up to the highest, most rigorous standards of logical accountability. My case is that Anglo-American philosophy was thereby deprived of some useful epistemo-critical resources, not least for overcoming the various problems bequeathed by logical empiricism, problems acknowledged--in their different ways--by Quine and Marcus. More specifically, the two projects of Husserl and Bachelard can be seen to address crucial issues that were mostly disregarded by thinkers in the mainstream analytical tradition, but which have often surfaced to vex that tradition in the form of such internal disputes as the rift over modal/intensional conceptions of meaning, logic, and truth. Hence the lately emergent strain of post-analytic philosophy, conceived--very often with reference to Quine--as an outcome of just these intractable problems in the doctrine of logical empiricism. I have argued elsewhere that this reactive movement of thought can easily swing right across to the opposite extreme, that is, toward a strong hermeneutical approach (derived from Wittgenstein and Heidegger) or a full-blown Rortian textualist position which finds no room for such old-fashioned notions as reality or truth.(12) Indeed, Quine's argument in "Two Dogmas" is itself a very striking example of the way that a thoroughly physicalist or naturalized epistemology can be made to square with an equally thoroughgoing doctrine of meaning holism, one that takes the entirety of scientific knowledge or belief at any given term as the unit of empirical significance. And from here it is no large step to those varieties of hermeneutic-textualist doctrine according to which interpretation goes all the way down, so that any results turned up through direct observation or empirical research must themselves be treated as interpretive constructs, or as making sense only against some background horizon of communal belief. For if indeed, as Quine writes, "it is nonsense, and the root of much nonsense, to speak of a linguistic component and a factual component in the truth of any individual statement," then equally it is the source of much confusion--or so I shall argue--to push this case for meaning holism to the point where "truth" becomes entirely a matter of contextual definition.(13)