“Bright line” is increasingly making its way out of the halls of legal discourse. Its origins appear to be
obscure. In physics and photography usage, it dates to 1890; the earliest legal usage (“bright line rule”)
appears to date from the 1970s. This led Ron Harris to think that the term was not imported directly from
physics into law; the timing, he thought, might suggest a link between law and economics. Thanks to Hila
Keren and to Ron Harris for enthusiastic help trying to source its meaning.
This is what Harding (1993) calls “strong objectivity,” which means taking positionality into account
so that the relationship of knowledge to power can be understood. She argues that the standard, method-
ological positivist assumption that scholarly position can be ignored produces “weak objectivity” in which
power, such as class or gender power, still operates but without accountability.
For example, explaining why he excluded social constructivist approaches to policy from his influential
Theories of the Policy Process, Paul Sabatier writes: “Although it is clear that much of social ‘reality’ is ‘so-
cially constructed,’ these [social constructionist] frameworks: (a) leave ideas unconnected to socioeconomic
conditions or institutions and (b) conceive of ideas as free-floating, that is, unconnected to specific individu-
als and thus largely nonfalsifiable” (1999, 11). Our understanding of interpretive research (including social
constructionist work), described and illustrated throughout the book, is that, on the contrary, it is thoroughly
grounded in specific persons, times, places, conditions, and institutions.
We thank Tim Pachirat for initially drawing our attention to this phenomenon of contending method-
ological camps’ using the same terms but meaning different things by them.
In her study of the graduate curriculum of fifty-seven political science programs, Schwartz-Shea
(2003) found that over half of the programs did not require even minimal exposure to the philosophy of
social science.
For a history of qualitative methods in sociology and anthropology, see Vidich and Lyman (2003).
One of the central, although not always articulated, procedural principles underlying interpretive
methods of accessing and generating data (such as observing, with whatever degree of participation) is the
extent to which researchers form provisional interpretations based on their own personal responses to the
situation(s) under study. The U.S. phenomenologist (albeit better known for his work in developing human-
istic psychotherapy) Carl Rogers described this process as first forming inner hypotheses, in a subjective
mode of knowing within oneself, about what is going on in the event, “making patterned sense out of [one’s]
experiencing” from within one’s “own internal frame of reference” (1964, 112, 110). This interpretation is
then checked in further observation of others’ acts and responses and/or in conversation with those others
about their experiences and responses, and corroborated or refuted.
King, Keohane, and Verba (1994) seem to understand this point. They state: “Although case-study
research rarely uses more than a handful of cases, the total number of observations is generally immense. It
is therefore essential to distinguish between the number of cases and the number of observations” (52). Yet
they go on in chapter 6 to offer advice for increasing the number of observations in small ‘n’ research. To
make sense of this apparent inconsistency requires acceptance of a methodologically positivist framework in
which “observation” means, as they state, “measures of one or more variables” (53; emphasis added). In
other words, “observations” do not “count” unless they are in the form of “variables.” Moreover, “social
inquiry” is restricted to causal inference of a methodologically positivist sort.
See the preceding note. Brady and Collier (2004) address many of the limitations of the King, Keohane,
and Verba (1994) conceptualization of qualitative methods, although this point about observations is not one
they engage.
These interpretive schools and methods have family resemblances, in Wittgenstein’s sense—as Dosse
(1999, xv) noted with respect to contemporary French philosophical schools—but they also have specific
differences. Schwandt (2003) makes a similar point about there not being a direct ideational link between
phenomenology and methods; rather, he argues, one sees similarities of presuppositions, but with different
applications and terminologies in different fields (e.g., the different, and confusing, uses of constructionism
in sociology; constructivism in psychology; and both in international relations, but meaning different things;
see note 13). At the same time, methodologists are beginning to talk about “phenomenological methods” or
“phenomenological approaches” to research and about hermeneutic social science methods (see, e.g., Bentz
and Shapiro 1998, chapter 7; Groenewald 2004).
We are opting in this book not to draw a conceptual distinction between these two terms, for the
reason that the philosophy of science literature that informs our point of view does not draw such a distinc-
tion, nor is it predominant in the policy, public administration, and organizational studies fields that are our
disciplinary homes. We would be remiss, however, not to note that in comparative politics and in interna-
tional relations, this distinction appears to be taking root—or, at least, appears to be contested terrain. Daniel
