50 MEANING AND METHODOLOGY
50CHAPTER 3
GENERALIZATION IN COMPARATIVE AND
HISTORICAL SOCIAL SCIENCE
The Difference That Interpretivism Makes
ROBERT ADCOCK
My initial thought as an undergraduate at the University of Minnesota was that I had gone to “the
U” to train to become a research chemist; but this plan did not survive my introduction to ex-
tended lab research, and by my sophomore year I was hunting for alternative endeavors. Supple-
menting my chemistry major with a second major in political science left my interests something
of a hybrid mess. But I settled on graduate school in political science, declaring in my personal
statement that my “research interests” lay in “questions about the nature of a ‘scientific’ ap-
proach to the ‘political’ realm and the role of ‘theory’ within that approach.” Dual interests in
political theory and methods, and a specific concern with contentions around “scientific” method,
have persisted throughout my graduate training. They have, however, been substantially reshaped
and specified.
Early in graduate school I found myself puzzled by the way that debate around whether politi-
cal science could and should be “scientific” was often conducted among graduate students.
Positions on the issue paralleled disagreements over the merit of statistical and formal tech-
niques, with everyone seeming to agree that such techniques seek to model political science on
natural science. My own experience was, however, that I had not heard of such techniques as
regression when I was a chemistry undergraduate and that my friends who had gone to graduate
school in the natural sciences, although learning complex techniques, were not learning the
ones taught in our methods courses. The confusion that this situation generated for me cleared
up only when I came to recognize that most techniques abused or celebrated in our debates over
“science” had not been imported from natural science, but rather had been developed within the
social and behavioral sciences. Over time I came to believe that not only most techniques, but
also much of the conceptual vocabulary deployed in “methods” conversations—so often seen as
efforts to ape natural science—had a similar line of descent. Thus, in studying “validity” with
David Collier, I found that my understanding of the issues involved was helped most by tracking
the origins of the way social scientists use that concept and by my discovery that these origins
mostly lay in psychometrics.
My realization that techniques and concepts at the heart of much contention over “science” in
the social sciences were often not derived from the natural sciences was complemented by a belief
that I gradually evolved regarding some of the most theoretical stances and arguments deployed
in these contentions. I knew that these often drew on the philosophy of science. At one point, this