Architexts of Memory: Literature, Science, and Autobiography (review)

By force of habit professionals are prone to mistake the landscape for the window view opened by their specialization, a view whose relevance and explanatory power they are given to overestimate. Economists see maximization of wealth and pleasure behind every endeavor; scientists boil the world down to mathematical forms; sociologists believe group behavior exhausts our understanding of man; and literary scholars see the world as – what else? – a book, language, or to follow the current phraseology, text.

Under this particular strain of the reductive malady, the critic holds that all facts and objects are really the product of ways of speaking and telling stories. The world is overlaid by a thick net of words which, under the best circumstances, scarcely lets us behold things such as they are, in the naked fact. Evelyne Ender's Architexts of Memory is no exception to this reflex of putting the professional medium (here, language) ahead of the stuff it means to represent. The author proposes to illustrate how memory – our accounts of the past – is embryonically structured by narrative language, so tightly indeed that a memory is really indistinguishable from its telling.

Her study, written in a sensitive, melodic style, is effectively a variation on two basic premises: A, that memory is constitutive of selfhood; and B, that memory is textual, in essence. Which, if A and B follow, means that the self, too, is a textual phenomenon, and that you and I are no less woven through with words than the books we read. To anyone what has hung around the literature classroom in the last forty years, this thesis is not likely to deal a systemic shock. But the value of Architexts of Memory lies less in conceptual innovation than in the details: in the fond, sympathetic attention it turns to the writings of Proust, Woolf, George Eliot, Nerval, and other talkative rememberers. The book's metabolic make-up (short on conceptual insight, long on the finely wrought nuance) actually reflects Ender's idea of memory, which holds that the power and beauty of it is never in the generic, but in the poignantly singular. "Personal memories are a highly subjective phenomenon" (12). Indeed. And if Ender is not afraid of sonorous truisms, it is because her fragrantly Proustian sensibility does not believe in generality, and trusts that the human truth, the meaning of life, dwells not in the big picture but the poetic petite madeleine.

Fair enough. Yet, one feels a scholarly volume is answerable for its conceptual underpinnings even if, as in the case here, they are art of the academic stock-in-trade. Thus, for instance, the tenet that autobiographical memory is constitutive of subjectivity or that, as Hume argues, it gives us an identity. On the face of it, the idea seems undeniable; there is no identity that doesn't extend across time; the sense of who I am rests on recollecting who I have been. Nevertheless (Kant's checking move on Hume), there still remains an irreducible "I" that is doing the remembering. Should the self be consubstantial with its memories, there would be no agency external enough to collect and identify with these memories. The self would simply be a swarm of narratives without a subjective center. There therefore has to be a self distinct enough from the stream of memories who sieves, selects, and assigns meaning. This necessary self sets a limit to the notion, shared by Ender, that subjectivity is textual, that is, a byproduct of the language.

Likewise the idea that memory is a textual phenomenon; that narrative is the memory, not an impediment to it. Ender has no less than Proust himself to underwrite this claim (as Proust writes, the remembering person ". . . is face to face with something which does not yet exist, to which he alone can give reality and substance"). This is not false; nor, again, is it entirely true. Witness Proust's own colossal labor to wring out the most...