HE GROUNDS for scepticism about interart analogies are extensive. One is the record of self-contradiction and silliness in transferring art historical terms: the best we can say about the casual speculations even of Bernard Fehr, Wylie Sypher and Mario Praz is that they leave the hypothesis of analogy substantially untested.1 A more fundamental ground is C. S. Lewis's reason for not joining the search for "meanings" in historical periods: that the content of the past may not have a meaning. Even if it has, "we can never know that content. The greater part of the life actually lived . . . consists of minute particulars and uncommunicated, even incommunicable, experiences which escape all record. .... On such a basis it seems to me impossible to reach the sort of knowledge which is implied in the very idea of a 'philosophy' of history." 2 Imagine what future cultural historians may say about our own period. Must they not tidy up our confused and cluttered scene? But if in one way the historian knows too little, in another he knows too much for period classifications. Anyone who studies literature finds it to consist mostly of exceptional cases. Metaphysical conceits, for example, are not so very common in Metaphysical poetry; whereas striking instances occur elsewhere, such as Jonson's image of the "brave infant of Saguntum." And if this is true even of schools of poetry, when we turn to whole periods and megaperiods we may well find-as indeed Praz himself inadvertently put its-that "all human things exist in every period of history."3 However, this is not to say, with Lynn Thorndike, that "human nature tends to remain much the same at all times";4 still less that