58 MEANING AND METHODOLOGY
tentions against which to explicate the agenda of comparative historical analysis and to explore
the contrasting, interpretive approach to comparative and historical social science.Modernist Epistemology and Comparative Historical AnalysisIn setting out to explicate comparative historical analysis,^12 we should be wary of claims about its
qualities that, while they might do important rhetorical work for proponents of the approach, do
not capture relevant distinctions. Proponents recur, for example, to the claim that this ap-
proach focuses on “big questions” of major substantive and normative importance. This claim
is, of course, not novel. Indeed, along with the perhaps not unrelated claim of building on a
grand European tradition, it parallels the identity that Parsons crafted for comparative inquiry
in the post–World War II era. Some interpretive social scientists have, moreover, also found
appeal in this self-image of asking big questions and perpetuating the heritage of Weber or
other classic European figures. Such parallels may testify to the continuing contemporary
vitality of certain strategies of legitimation among social scientists, but they help us little with
the task at hand.
A more fruitful starting point is to spell out further the modernist epistemology that compara-
tive historical analysis inherits. Earlier I spotlighted some moments in the genealogy of modern-
ism in American social science: emphasizing the rise of new correlation-based statistical tools,
the adoption of ideal reconstructions of experimental inquiry as a model for all scientific inquiry,
and the later crafting of a variant of modernism open to the merits of qualitative inquiry. Modern-
ist epistemology grew out of, and remains rooted in, the first two of these developments. Their
interplay supported a new interpretation of statistical and experimental inquiry as sharing a single
underlying logic: a logic that in turn came to be widely treated as the logic of all scientific inquiry.
This logic is embodied in the now-familiar methodological language in which talk of “controlled
experiments” and “control” groups passes over into talk of “statistical controls” and “controlling
for confounding variables.” As this logic/language was extended to analyze practices beyond
those with reference to which it originated—as, for example, when qualitative cross-societal studies
came to be interpreted as approximations to a “crucial experiment” or as setting up “controlled
comparisons”—it carried with it the presupposition that such studies, if they are “scientific,” must
by definition seek to construct knowledge with a modernist form.
What are the characteristic traits of this modernist form of knowledge? First, it is built upon a
certain way of interpreting the world: Reality as experienced is mentally carved up into concep-
tually isolated factors, which are, in turn, conceived of as potentially standing in various possible
relationships with one another.^13 Second, knowledge construction is taken to center around the
formulation and evaluation of propositions characterizing these relationships (their existence,
type, or size, and often also their “causal” direction). Third, generalization—that is, the pursuit of
knowledge specifically “general” in character—is taken to denote the subset of this endeavor in
which those propositions characterize relationships as recurring across some portion of time and
space. Among comparative social scientists working from a modernist standpoint, such generali-
zation has often been equated specifically with formulating and evaluating claims about relation-
ships that are characterized as recurring across more than one macro-societal unit.
Within comparative inquiry, the endeavor to produce knowledge claims with a modernist form
has proven compatible with a range of views on further issues. Most important for this chapter, it
has not entailed the choice of any specific approach—formal-theoretical, statistical, or qualitative
—with which to pursue the endeavor. The agenda of comparative historical analysis arises out of
its proponents’ efforts, beginning in the 1970s, to articulate and defend a particular stance on this