Saturday, August 1, 2009

Turner on Archival Research as Empirical (or why statisticians should "cozy up" to historians...)

Archival Research as Empirical has just been posted by William B. Turner, Visiting Fellow at the James Weldon Johnson Institute, Emory University. The abstract is very short, so here's an excerpt from the beginning of the paper:

Within the past few months, Harvard Law School has posted two jobs ads that use the term “empirical” in defining the position. A quick review of the job responsibilities makes clear that, at Harvard Law, “empirical” equals statistical data. This definition would seem to exclude categorically any form of archival research such as that which historians characteristically conduct.

But why is archival research not “empirical”?

In this paper, I want to examine more carefully what we mean by the term, "empirical” to demonstrate that archival historical research that makes no use whatsoever of statistical data is just as “empirical” as research that depends on statistical methods. Empiricism is, at base, a theory of knowledge, but a theory that has triumphed for its ability to enable manipulation of the physical world toward technological and engineering feats....

In its origins, empirical research assumed direct sensory experience of the phenomena under study. On careful examination, it becomes clear that there is nothing distinctively “empirical” about research that relies on statistical methods, certainly not that would distinguish it on those grounds from archival research. Both statistical and archival methods operate at a significant remove from the object of study, albeit in different ways.
Image credit.