86 MEANING AND METHODOLOGY
of inputs into a set of slots or boxes, representing the categories that the individual (researcher or otherwise)
is attuned to at any given moment. The slots change—we drop some, add new ones, alter existing ones—as
understanding changes; and in this and the sorting processes, we actively filter our observations, which in
turn shapes the world of social realities.- Here is where the linguistic argument enters in, that without words, we would be unable to “see”
things. The argument is being advanced in empirical research with color terms (Roberson 2005) and number
terms (Gordon 2004). - This effect on theoretical formulation has been documented in analyses of metaphors of social theory
(e.g., R.H. Brown 1976) and in feminist studies of medical and other natural sciences historically, for ex-
ample in conceptualizations of the fetus, of women’s internal organs, and of women’s health that were
shaped by images or theories of the male bodily norm (see, e.g., Tuana 1989). - As a rule of thumb, researchers plan on twice the amount of time for note taking as they spend
observing or interviewing (i.e., one hour of conversational interviewing requires, on average, two hours for
writing up one’s notes on the interview; transcribing from an audio- or videotape can require much more,
depending on the kind of transcription equipment one has, the level of detail one wants to note, etc.). - For example, by attending to word choice, question order, and training in presentation skills and
processes, or through (e-)mailed questionnaires that eliminate human contact altogether (although they have
a lower rate of return). These techniques and others are detailed in survey design textbooks and sections of
general textbooks. - At the other end of the spectrum from such efforts to “contain” the researcher and delimit interactions
is the argument that the actors in the situation under study are coproducers of evidence along with the
researcher (see, e.g., Behar 1993). - This is the subject area of the philosophy of mind. See, e.g., Sellars (1997) for an overview of the
debates. Latour (1999) has an interesting discussion from a science studies perspective. - These meanings, however, have not been constant over time. An essayist for Webster’s Revised Un-
abridged Dictionary (Trendelenburg, 1998 edition, accessed at http://dictionary.reference.com)) notes: “In
the Middle Ages, subject meant substance, and has this sense in Descartes and Spinoza.... Subjective is
used by William of Occam to denote that which exists independent of mind; objective, what is formed by the
mind.... Kant and Fichte have inverted the meanings. Subject, with them, is the mind which knows; object,
that which is known; subjective, the varying conditions of the knowing mind; objective, that which is in the
constant nature of the thing known.” - This begs the question of who is doing the observing. The medical definition of objective, however,
takes point of view into account: “1. Based on observable phenomena; presented factually. 2. Indicating a
symptom or condition perceived as a sign of disease by someone other than the person affected” (The
American Heritage Stedman’s Medical Dictionary 2002, http://dictionary.reference.com/search?q=objective;
[accessed October 31, 2005 4th ed.]). Subjective, by contrast, means “Of, relating to, or designating a symp-
tom or condition perceived by the patient and not by the examiner” (The American Heritage Dictionary of
the English Language, 4th ed. 2000; emphasis added). Following this definition, to the extent that interpre-
tive analysis is “based on observable phenomena”—even if it is an interpretation of those phenomena, much
as a symptom or condition would be perceived in analysis as a sign of something else—such research is,
definitionally, objective as long as the interpretation is made by someone other than the person experiencing
the condition! This in effect eliminates autoethnography, such as that of Greenhalgh (2001), as well as
interaction between patient and physician in negotiating the diagnosis, which is a central issue in contempo-
rary medical practice. - Herein lie the roots of the difficulties of accommodating a public sociology (e.g., Burawoy 2005) or
an engaged political science within the framework of academic science (the “ivory tower”). - See, e.g., Rein (1976). This line of argument dovetailed with the growing view in both public policy
and public administration, especially in implementation studies, that the politics-administration dichotomy
was an oxymoron: Rather than seeing politics, and, hence, values, as restricted to the pre-administrative
policy-framing and legislative phases, administrative acts came to be seen as themselves fraught with poli-
tics, much as organizational acts were increasingly seen as political (see, e.g., Fox 1990, Nakamura 1990,
Pressman and Wildavsky 1973, and Yanow 1990, in implementation studies; Pfeffer 1981 in organizational
studies). - My argument that these methods do not constitute “sampling” is based on the latter’s close associa-
tion with statistical science, where it is defined and used as “a set of elements drawn from and analyzed to
estimate the characteristics of a population” (American Heritage Dictionary, accessed at http://