In its original form as the 1973 T. S. Eliot Memorial Lectures at the University of Kent, this must have been exhausting. One suspects that it's a great deal more manageable as a book. The subject may be inadequately described as the attitudes various literary epochs have held toward their formative pasts. Until rather recently the classics of other ages constituted a stable set of data in educated minds, although not everyone went as far as Eliot in proclaiming Europe still fundamentally a Roman province to be judged according to Roman-Christian canons. Beginning with Eliot's "imperialist" model, Kermode shows how it gradually strains at the seams when required to accommodate not only Virgil, Dante, and the Elizabethans but the more recalcitrant Milton, ambiguous Marvell, and urbanely unheroic or a-heroic English Augustans. With a provocative break in method, Kermode then abandons the Eliot framework to analyze the roles of past and present in the novels of Hawthorne. In Hawthorne's world, meanings do not stay where an observer has put them: in time ambiguity becomes instability; species are confusingly represented by anomalous individuals; progress and degeneration compete as historical processes. Such is the vision Kermode finds behind Hawthorne's deliberately perplexing narratives, and it points the way to an epoch in which criticism cannot rely on objectively "real" meanings to evaluate literary classics. Kermode concludes--by way of some rather annoying meanderings on Wuthering Heights--that in our day the genuine "classic" literary value must be a loosely structured, multi-significant inclusiveness, an ability to encompass a broad variety of interpretations. One is grateful for dozens of individual literary insights (the Hawthorne chapter alone is worth the price of the book), but Kermode employs a lot of grandiose machinery to formulate a conclusion which most students of literature have heard expressed more simply.