Today's empiricism and transcendentalism both reject metaphysics, but each appears sometimes to the other as actually engaging in the rejected metaphysics. From an empiricist standpoint, transcendentalism seems to grant too much to the knowing subject whereas from a transcendentalist standpoint, empiricism seems to concede too much to realism. The challenge posed for empiricism is to explain how it could make sense, within an empiricist stance, to say that there could be things that are not describable (in our language in use) and hence not knowable, let alone known. The remaining ‘common sense’ realism, that I acknowledge in response, can – I submit – be clearly distinguished from any metaphysical version vulnerable to transcendentalist critique.