…Believing, with Max Weber, that man is an animal suspended in webs of significance he himself has spun, I take . . . the analysis of [those webs] to be therefore not an experimental science in search of law but an interpretive one in search of meaning. Clifford Geertz (1973, 5)…
This oft-quoted line from anthropologist Clifford Geertz’s The Interpretation of Cultures has constituted, for many, the summons to interpretive social science. But that call does not instruct readers directly and explicitly how to analyze those webs of meaning, and the text makes only passing mention (through no fault of the author, who had other purposes in mind) of the vast early- to mid-twentieth-century literature dealing with the tremendously significant components of that sentence: phenomenological analyses exploring the ways in which humans weave not only the social world in which we live but the very identities we construct for ourselves as we live in those worlds; hermeneutic treatises developing ways of interpreting the various sorts of webs woven out of human experience; other philosophical works explaining the relationship between meaning and law, meaning and experiment, meaning and interpretation; and, most crucially, what it might portend in the context of the human sciences to interpret the social world in order to understand not only what it means to those for whom specific webs have meaning, but also how it means, both to situational actors and to researchers studying those situations.
Reflecting on the problematic of “how” things mean, in the context of doing social science in the spirit of nineteenth- and twentieth-century science, leads rather quickly to the understanding that this is not a matter merely of what “tools” to pick up to strike the nail or to plug the hole in the dike. Rather, we are concerned here, first, with epistemological matters: questions regarding the “knowability” of the subject of study, the capacity of human animals to “generate” or “discover” or “find” or “construct” knowledge about the social webs under their analytic microscopes, and, hence, the character of those claims to knowledge. Moreover, knowing which tool is best suited to the nail or the hole depends on the character of the subject of study, and so epistemological presuppositions themselves rest on the presupposed reality status of that subject. Ontological matters are, then, also a concern: whether the subject of study is considered objectively real in the world, in which case it is believed to be capable of being “captured” or collected, discovered or found, and “mirrored” in theoretical writings, or is considered as socially constructed (in the phrase made widely known through P. Berger and Luckmann ), in which case its character may be apprehended only through interpretation.
Methodological justification, then, cannot be made in the void of ontological and epistemological entailments. A researcher who presupposes that the social world is ontologically constructivist and epistemologically interpretive is more likely to articulate research “questions” that call for constructivist-interpretive methods.3 These turn a reflexive eye not only on the topic of study but also on the scientist generating or constructing (rather than “discovering” or “find- ing”) that knowledge and on the language she uses in that “worldmaking” (in Nelson Goodman’s  term). Such reflexivity might ask such questions as, for example, what the Middle Ages or the Middle East are in the “middle” of and who “put” them there, and why. It is an approach that sees concepts and categories as embodying and reflecting the point of view of their creators (as Ido Oren, Robert Adcock, and Dean McHenry do in their chapters with respect to political science theorists and Pamela Brandwein, Patrick Jackson, and Ronald Schmidt do in theirs with respect to jurisprudists, political parties, and legislators). In this view, the subject of social scientific study “is not an inert fact of nature . . . merely there” (Said 1978, 4) and social scientific theories are not “mirror[s] of nature” (Rorty 1979). All ideas including those of the natural and physical sciences have “a history and a tradition of thought, imagery, and vocabulary that have given [them] reality and presence” (to borrow Edward Said’s words [1978, 4–5] from the context of Orient and Occident). In this sense, then, social scientific texts do not merely present their subjects through the lenses of their data, but represent and represent constitute, construct them.
The philosophies that developed these concerns emerged in engagement with the critical positivism of the later nineteenth century and the logical positivism of the early twentieth century, through both critique of their perceived shortcomings and positing of what proponents felt was a more logically compelling delineation of the entailments of human social life in constituting Self, Other, and the broader social group. “Interpretive” methodology has since become an umbrella term subsuming several different schools of thought, including those drawing, explicitly or im- plicitly, on phenomenology, hermeneutics, or (some) Frankfurt School critical theory, along with symbolic interaction and ethnomethodology, among others. Many of these ideas dovetail with late-nineteenth to early twentieth century pragmatism4 and later-twentieth-century feminist epistemology and research methods (e.g., Falco 1987; Harding 1989, 1990; Hartsock 1987; Hawkesworth 1989; Heldke 1989; Miller 1986; Modleski 1986) and science studies (e.g., Harding 1991, Latour 1987, Longino 1990, Traweek 1988).
This chapter treats, with a rather broad brush, those philosophical presuppositions held in common by several of these schools of thought, which provide conceptual grounding for inter- pretive methods. Not only do these logically prior, or underlying, notions of social realities and their “knowability” distinguish them from the positivist presuppositions with which they took issue (discussed by Mary Hawkesworth in chapter 2). These philosophical argumentations also provide the methodological principles shared by and manifested in the various interpretive methods of accessing, generating, and analyzing data, for all their other differences. The philosophical terrain encompassed here is vast; discussing it in any detail would require a book-length manuscript itself; and, indeed, many books have been written, both primary sources and secondary analyses, on these questions. If space allowed, other schools of thought could also be more exten- sively discussed in terms of their influence on and/or manifestations in interpretive research practices, specifically, American pragmatism, toward the more philosophical end of the spectrum of influence (see note 3); ethnomethodology (developed by Harold Garfinkel), also of U.S. origins, toward the more explicitly methodological end; and critical theory (both Frankfurt School and wider ranging), symbolic interactionism (developed by George Herbert Mead and Erving Goffman), and feminist theories in the middle of the continuum (that is, combining both conceptual frame- works and “tools”). In my reading of this history of ideas, the foundational concepts, relation- ships, and understandings enacted in interpretive research methods are well established in the various philosophical arguments of phenomenology and hermeneutics; the other schools of thought articulate similar ideas, based on similar presuppositions, albeit with different orientations or even, as with Frankfurt School critical theory, in critique. Here, I sketch out the parameters of the arguments as engaged in phenomenology and hermeneutics. Although I am eliding differences that from other perspectives may be crucial, for what I wish to argue those distinctions are less central. My purpose is to show that interpretive methods do not just spring ab origine, at whim, but rather have considerable philosophical grounding. I hope to scatter sufficient bibliographical breadcrumbs that a reader curious for greater depth and detail will be able to follow the trail of argument further than space here allows.
The ideas developed in this chapter build on earlier work, including a chapter focused on organizational studies as an interpretive science (Hatch and Yanow 2003). My thanks to Mary Hawkesworth and to Peri Schwartz-Shea for their close critical readings of earlier drafts. All errors of interpretation remain my own.
1. Geertz imagines an exchange in which the muscles of the tradesman’s eyelid contract. How should this contraction be understood, he asks: was it an involuntary twitch, or was the tradesman signaling some meaning through an intentional wink? He uses this to illustrate the point that we need much more information— about the character of the events, the persons, the times, and so on—in order to be able to interpret which it was or to consider that it might have had other meanings altogether. This “richer,” fuller description is what he characterizes as “thick” description.
2. For these reasons, I avoid talking about the “nature” of the subject of study: From an interpretive ontological perspective there is nothing “natural” about our topics of investigation. Whereas this term (the “nature” of something) may seem like a dead metaphor and my point, a minor linguistic quibble, I believe that source meanings do tend to ride in on the backs of words, even when that knowledge is tacit. I thank Davydd Greenwood for drawing my attention to the metaphoric character of this term.
3. This is itself a delimiting statement: It posits that some subjects of study are more usefully ap- proached through interpretive methodologies and their attendant tools of analysis than others. The dividing line rests with the presuppositions brought to a “research question” by the scientist: Questions about the same research subject may be formulated differently, depending on the epistemological and ontological orientations of the researcher (and, of course, on her education and training—connections that may well be interrelated; see the discussion of presuppositions in the book’s introduction). I put “research question” in quotation marks because it has become clear to me that different methodological camps understand this phrase differently. Whereas projects influenced more by methodological positivism treat this phrase literally —for them, a research question is a full-blown statement, a hypothesis—for interpretive researchers, the “question” is more commonly a topic, a puzzle, or a tension that draws their attention, often because of some prior, possibly experiential knowledge that informs their curiosity and suggests that this is an area worthy of research attention. This difference in approach is manifested also in research designs: Interpretive research designs more commonly begin with what might best be called hunches, rather than with hypotheses.