- Purchase/rental options available:
American Literature 74.4 (2002) 911-931
[Access article in PDF]
Robert Lowell, Roman History, Vietnam War
Wai Chee Dimock
What is non-Newtonian time? No ready answer comes to mind. Its opposite—Newtonian time—hardly fares better. Neither term is idiomatic or even vaguely recognizable because time is rarely qualified by these adjectives, or qualified at all. In more than just a grammatical sense, time seems to come all in one piece, in one flavor. It is an ontological given, a cosmic metric that dictates a fixed sequence of events against a fixed sequence of intervals. It is present everywhere, the same everywhere, independent of anything we do. It carries no descriptive label and has no need to advertise or to repudiate that label.
When seen as this uniform background, time is quantifiable. Its measurable segments are exactly the same length, one segment coming after another in a single direction. This unidirectionality means that there is only one way to line up two events, one way to measure the distance between them. Apparently, we need to imagine time in this concrete form—as a sort of measuring rod—to convince ourselves of its absolute existence. One year, one month, one minute—these unit lengths have to be "real" unit lengths, objectively measurable. And as proof of that objective measurement, they have to come already stamped with a serial number. And so we speak of one particular minute as, say, 10:10, followed by the next minute, 10:11, just as we speak of one year as, say, 1965, followed by the next year, 1966. This serial designation puts time completely under the jurisdiction of number.
Most of us take this step quite innocently. Without much thought, we refer to a particular year as 1965, because a numerical bias is so [End Page 911] deeply entrenched in our thinking that it works as a kind of mental reflex. Dates thus acquire a summary authority, with great descriptive and explanatory power. We don't doubt that a number, 1965, can be assigned to one particular slice of time and that this meaningful numerical designation exercises a binding power over all the events that happen within its duration. Number, in this way, works as a kind of automatic unifier: it imposes an identity across-the-board. Because this numerical chronology standardizes time into a sequence of equal units, the location of any event and its proximity to any other is fixed by this sequence. 1965 is separated by only 20 years from 1945, and so it has got to be closer to this year than to the year 65, from which it is separated by 1900 years.
This numerical bias is the unspoken norm for humanists no less than for scientists. Under new historicism, this norm has sometimes turned into a methodological claim, producing a spate of scholarship whose very ground of analysis is numerical time. It is routine for us to seize upon one particular number—the date of a text's composition—and use it to set the limits of an analytic domain, mapping the scrutinized object onto a time frame more or less standardized. One year, five years, ten years: these are the numerical units we use, as a matter of course, when we try to contextualize. Defined in this way, contextualization is based almost exclusively on synchrony. Events are deemed pertinent to one another only if they fall within the same slice of time, which is to say, if they are bound by two serial numbers so close to being consecutive that the distance between them is measurable by single digits. This short duration is supposed to be adequate, to capture both cause and consequence: the web of relations leading to the making of the text and the web of relations flowing from its presence in the world.
Elsewhere I have argued against this synchronic model, offering instead a conception of literary history based on extended duration, what I call "deep time." 1 Here I want to take issue, more specifically, with the numerical bias at work in a...