History as Science

Reading the important introduction to Hayden White's Metahistory: The Historical Imagination in Nineteenth-Century Europe (1973), I am struck by White's lucidity. I do not recall a clearer explanation of why history is not a science.

[H]istory differs from the sciences precisely because historians disagree, not only over what are the laws of social causation that they might invoke to explain a given sequence of events, but also over the question of the form that a "scientific" explanation ought to take. There is a long history of dispute over whether natural scientific and historical explanations must have the same formal characteristics. This dispute turns on the problem of whether the kinds of laws that might be invoked in scientific explanations have their counterparts in the realm of the so-called human or spiritual sciences, such as sociology and history. The physical sciences appear to progress by virtue of the agreements, reached from time to time among members of the established communities of scientists, regarding what will count as a scientific problem, the form that a scientific explanation must take, and the kinds of data that will be permitted to count as evidence in a properly scientific account of reality. Among historians no such agreement exists, or ever has existed. ... [H]istorical explanations are bound to be based on different metahistorical presuppositions about the nature of the historical field, presuppositions that generate different conceptions of the kind of explanations that can be used in historiographical analysis.
White, Metahistory, 12-13. 

The assertion, "[t]here is a long history of dispute...", catches my eye, however. White offers here a proof that settles the question of the unsettled matter of historical method and narrative. How is it possible to make such an assertion without taking sides in the dispute under investigation?