Consider statements like ``Machines will never be able to think'' or ``You cannot turn a cat into a dog''. Many people will find such propositions evident, but we have learned to be cautious. 100 years from now things might look very different; it seems unduly dogmatic to insist on todays opinions as unrevisable points of reference, disallowing future breakthroughs in computer science or biology. The scientific outlook presupposes and enhances a high degree of curiosity and it seems a good idea to generally keep one's meanings flexible so as to master break-downs of everyday understanding induced by unforseen circumstances, science or technology. Such intuitions are at the heart of Quine's philosophy. According to him we find ourselves enmeshed within a web of belief that joins together past and present attitudes, cognitive procedures and evidential tests. The strands show various degrees of coherence and interconnectedness and there are no general rules on how to proceed in case something unexpected happens. Sometimes it is ignored, sometimes it is explained away and sometimes it induces more or less extensive changes in our predictive and behavioural patterns. This seems to be the lesson drawn from the Vienna Circle's battle against metaphysics.
Philosophers like Putnam, Kripke and Davidson have, it is true, come up with accounts of meaning, necessity and reality that seriously challenge Quine's indeterminacy of translation and ontological relativity, re-introducing elements of a priori presuppositions and (quasi-)transcendental arguments into philosophical discourse. But - compared to Quine's unrelenting insistence on naturalistic first-level talk - those attempts have a free-floating, speculative touch. Quine refuses to engage in elaborate philosophical constructions designed to block the permanent drift of meaning. His is a truly radical position, admirably suggestive and provocative, to be challenged on its own ground. The untenability of a priori truth is suggested by profound experiences that cannot easily be refuted by arguments purporting to show how some more or less exotic sentences could, after all, be held immune from change.1 If the untenability thesis is to be shaken a more down to earth strategy seems to be called for. It would have to be shown that Quine's radical move entails consequences that are highly unattractive even according to its own standards.