100 MEANING AND METHODOLOGY
interpretive study?, reading the criteria literature suggests that there is little consensus. I show,
however, that a different approach reveals considerable commonalities, commonalities indicative
of the interpretive gestalt.AN INDUCTIVE APPROACH TO A SET OF CRITERIAA skeptic might argue that the apparent lack of consensus displayed in Tables 5.1–5.3 does not
much matter to research practice. But many within the interpretive epistemic community may
well concur with Miles and Huberman that “shared standards are worth striving for” (1994, 277)
because shared standards facilitate scholarly communication and judgment, and that means, pre-
sumably, that they should be clear and coherent. Stepping back from this seemingly reasonable
presumption, one might still question whether such “striving” reveals an unstated, positivist faith
in the possibility of a coherent taxonomy of a universalistic, Platonic nature. This is the conclu-
sion of J.K. Smith and Deemer (2003) who, in their analysis of the criteria literature, self-consciously
refuse to provide any recommendations of particular criteria or any taxonomies such as those
presented in the texts discussed above. Instead, they argue that there cannot, and never should, be
a definitive list of evaluative criteria. They reject the possibility of any universal, unchanging
taxonomy as ahistorical and as inconsistent with the epistemological and ontological presupposi-
tions of interpretivism. A definitive list, they argue, is inconsistent with interpretive skepticism
concerning the possibility and usefulness of universal knowledge; research is an historically situ-
ated endeavor, and the criteria literature should reflect this situatedness. They also argue that an
ahistorical approach makes the discussion of research quality too abstract. The meanings of crite-
ria are “ever subject to constant reinterpretation” as they are applied to concrete studies (J.K.
Smith and Deemer 2003, 445).
Smith and Deemer’s fundamental insight is that the criteria literature should not be chasing the
mirage of the perfect, timeless taxonomy. Yet they acknowledge that having some slowly evolv-
ing list of criteria offers a pragmatic place to begin the assessment of research, a point of entry
into the hermeneutic conversation about the quality of any particular interpretive study, with
additions or deletions emerging in response to new thinking, new practices, and changing re-
search priorities. What Smith and Deemer do not do is to take the next step: assess which of the
plethora of criteria and techniques should constitute the first entries in that evolving list. Perhaps
they resist this step because of its political implications of scientific inclusion and exclusion and, as
important, because they would have to find grounds on which to justify their decisions—bringing
back the specter of a timeless, universal taxonomy. One way out of this conundrum is to step back
from a normative, a priori, deductive approach and, instead, to develop a list that draws, inductively,
on what is taught and used most widely, as an indicator of accepted interpretive practices. Although
I agree that epistemic communities develop and change, for any particular period of time it should
be possible to name a core list of criteria and techniques provisionally useful for the pragmatic work
of judging the quality of interpretive research, yet always subject to revision.
In analyzing the cacophony of terms displayed in Tables 5.2 and 5.3, it is possible to ask which
ones are used and taught most often. What is it about a particular term (whether identified as
criterion or as technique) that adds to its “in-use” attractiveness? And, as important, how does that
term reflect and construct interpretive practices? In addition to analyzing the textbooks repre-
sented in those tables for answers to these questions, I examined some twenty-five other social
science methods textbooks, skimming the textbooks’ tables of contents and criteria chapters (where
they existed) and inspecting the indexes for relevant terms, using what might be called an “index
test” for the prevalence of the terms.