CRITICAL INTERPRETATION AND INTERWAR PEACE MOVEMENTS 297the interwar period existed for any single scholar to read in a lifetime, let alone for one book.
What was I to do about this problem? Here, I think my own naivete played into my determination
to look at events from a variety of perspectives. Not being a historian, I simply did not realize the
vast number of sources available in any given archive, even after doing initial research in archival
reference sources (e.g., Diplomatic Records 1986). Yet I found myself interacting quite a bit with
historians during my archival research and eventually devised a system whereby I focused on the
contours of particular, more or less temporally bounded events, while cross-checking any inter-
esting assertions in the records with other documents and keeping the option open to investigate
additional evidence (both supporting and contradictory) as I found it. Now, whenever I read posi-
tivist treatises on the “rigor” of the scientific method of “testing,” I know that scholars who em-
ploy archival methods engage in extremely thorough and systematic analysis of evidence and that
the criticism that qualitative and interpretive methods lack such characteristics is simply wrong
(on the term “rigor,” see C. Lynch 2005; and Yanow, chapter 4, this volume). Because of this, I
was extremely pleased after the publication of my book that it received praise from historians.
UNDERSTAND THE PHILOSOPHICAL PROBLEMS OF
EVIDENCE ACCUMULATION
In IR we often talk about a theory’s or an event’s being “overdetermined” or “underdetermined”
by the evidence. World War I is a major example of overdetermination according to conventional
academic wisdom: There were so many significant causal factors that converged to make it hap-
pen that it could not have been avoided. However, it is important to remember that theories about
events (even World War I) are actually underdetermined by the available evidence; that is, schol-
ars can never have enough “facts” to prevent value judgments from entering into our explanations
for events, and the facts themselves are objects of interpretation. As philosophers of science such
as Mary Hesse have pointed out, “Theories are logically constrained by facts,” but “they can be
neither conclusively refuted nor uniquely derived from statements of fact alone” (1978, 1). Her
point is that statements of fact can always be repackaged in different ways, resulting in different
theories (for example, about why an event occurred), and scholars must look beyond the realm of
“fact” to that of value to understand why some explanations are preferred to others.
So how do researchers in the social sciences decide which theory is the most persuasive? The
major criterion for the success of a theory is the “pragmatic criterion,” which means that it ap-
pears to work better than others to explain more aspects of the (political) environment. But the
problem with the pragmatic criterion is that it sometimes runs up against an obstacle that makes
crucial aspects of the theory fall apart. So it was with the pre-Galilean assumption about the
earth’s shape; so it is, I assert, with the realpolitik assessment of the interwar period.
RESEARCH THE PRACTICAL PROBLEMS OF EVIDENCE
ACCUMULATION
Using archives requires an understanding of classification systems. These can differ enormously
from country to country and among nongovernmental groups, as can the rules for declassification
of official archives. For example, in Britain, the PRO declassifies some government documents
after thirty years, others after fifty years, and others after an even longer period of time. I con-
ducted dissertation research at the PRO in 1988–89, which meant that a number of documents
relating to peace groups from the Czech crisis in 1938 had just been declassified. I eagerly or-
dered these documents, and though I did find a great deal of useful and interesting information in