The sorts of questions that Frank Kermode asks in History and Value have broadly to do with why literary works consciously bonded to the historical age in which they were produced are often denied value. His discussion is situated in the context of British writers of the 1930s, whose present neglect, he believes, is caused by their espousal of political positions that were later proved to be untenable. The main thrust of his argument is clearly toward demonstrating the hollowness [End Page 377] of criteria of value based on notions of continuity between past and present. Neil McEwan's Perspective in British Historical Fiction Today, on the other hand, advances precisely the reverse argument, and it does so in a way that unwittingly fortifies, by contrast, Kermode's argument. McEwan's contention that the most enduring historical fiction tends to establish a sense of continuity between the past and the present disturbingly suggests that there is an "essential nature to be discovered within the features of any particular culture" for such determinations of value to be made in the first place.
Reaction against essentialism of any kind may well be what distinguishes Kermode's work from McEwan's. Indeed, historical fiction is a good example of the point that Kermode wants to make in his book: that value originates in the compulsion to create an imaginary whole that we then call history. What McEwan accepts unquestionably and perhaps a little too complacently, Kermode sees as the starting-point for constructed value-systems and manufactured histories that then function to advance spurious truth-claims. It is in part to deconstruct those systems that Kermode deliberately expresses less interest in examining why some books have come to be part of the canon and therefore "valuable"—a project that must invariably leave the truth-claims of those constructed histories intact—than in explaining why some books have fallen into neglect and disrepair. It is the breaks and discontinuities that interest Kermode, the seeming isolation of literary works to the age that produced them.
Kermode's chief problem in History and Value is to define value in positive terms without surrendering the Marxist base of his criticism. Although he is sympathetic to the Marxist project of uncovering the historical conditioning suppressed by a work's aesthetics, he is nonetheless cognizant of the essentially negative terms in which the project is seen by Marxist critics. For in repudiating transcendental notions of value in art, the archeological activity endorsed by Marxist literary theory implicitly locates value in the hidden or suppressed quality of literature's material relation to historical process. Kermode hastens to say that there is nothing wrong in considering value in terms that render art a document in the class war. But why, he asks, do even the most uncompromising of Marxist critics like Terry Eagleton find it so difficult to get away from the idea that there is something in art that makes it art and not a mere document in the class war?
Kermode is excellent in demonstrating how, in this respect, Marxist and liberal humanist theories of art converge. Marxist literary theory in his view is no less essentialist, no less able to escape history's mediation of value in its assumption that there is a "real history" that is covered over by false ideological versions of it. By insisting that value lies in the historical conditioning that is concealed by aesthetic expression, Marxist theories of art paradoxically save the value of past art. For art produced even under "old abhorrent dispensations" now superseded can legitimately claim value because of its usefulness in uncovering a suppressed history. Even Marx and Engels were susceptible to this current of thought, as the great esteem in which they held Balzac proves.
Kermode's book could equally be titled "Time and the Canon" in its suggestive analysis of the tensions between the specificity of history, its subjection to spatial and temporal boundaries, and the transcendent tendencies of literature, whose works are conventionally linked in...