"Compatibility, not incompatibility, is the issue. For pc set 'analysis' is incompatible with nothing, as the fact of its universal potential applicability already testifies. It begins not with observation of musical particularities but with a universe of possibilities. The comparison of any musical entity with such a universe yields an inexhaustible quarry of 'true facts' but no criterion of relevance. As long as not such criterion has been established… the endless stream of ostensible relations stemming from the pc survey can persuade us for a while that analysis is being accomplished. But in fact it is only a tabulation that can just as well be carried on in the presense of analysis as in its absence (hence the universal 'compatibility'). Nor is it really so innocuous as I may be making it seem, since in its anodyne effect (one never comes back from the fishing expedition empty-handed, there is always 'something to say' some 'finding' to report) it can deflect attention away from the task at hand, which is to formulate analytical methods, not concoct a universal solvent."