Reflections on DA-RT (three points)
The first has to do with the kinds of statements interpretive social science tends to wrestle with. Most of these statements are not true or false, but meaningful in a different register, a point ordinary language theorists since Wittgenstein have underscored. Recall from Wittgenstein’s Philosophical Investigations the picture puzzle that can be seen either as the head of a rabbit or as the head of a duck, or for that matter as a picture puzzle. The duck-rabbit, as Wittgenstein calls it, is a simple ambiguous line drawing. If a subject viewing it fails to see the ambiguity in the picture, she will see a duck or a rabbit. She will not be seeing the picture as a duck or as a rabbit, but simply seeing one or the other, the picture that is a duck or the picture that is a rabbit. When she does at some point see the ambiguity, she begins seeing as. She might say something like: “Now I see it as a duck” or “Now I see it as a rabbit” or “I saw it as a duck before, but now I see it can be a rabbit as well.”
The political theorist Hanna Pitkin (1993, pp. 100–101) uses Wittgenstein’s duck-rabbit to explore what we mean when we talk about what might actually be the objective facts of the world, as distinct from what any given individual would or could say about them on any particular occasion. When a naïve subject is shown the drawing and reports seeing, say, a rabbit, it appears at first to be an error. “The real truth,” Pitkin writes, “which even the subject would acknowledge if he [sic] were well informed, is that he is seeing a duck-rabbit as a rabbit. But one might, alternatively, want to argue that truth is a matter of interpersonal, intersubjective agreement, what both men could agree on. In that case the subject’s view is ‘truer.’”
The subject’s view was always true in the mode: “subject reports seeing a rabbit.” Whether the subject really saw a rabbit (or a duck) or a duck-rabbit, the truth will remain that she reported seeing a rabbit (or a duck). Equally true, however, is the view that what the subject reports seeing as a rabbit is (also to be seen as) a duck-rabbit or a duck. The “real truth,” we might say, is that the drawing can be (reported to be) seen in at least four ways, including as the fourth, ink lines on paper. All of these are true facts, with no one of them any truer than the others, not even the one that says that the “real real truth” is that it’s just material ink and paper. It seems harmless to concede some priority to materiality as such in this way, but the point Pitkin makes is that no such concession favoring any of the identifiable truths can be justified.
Pitkin further explores whether when people disagree about the facts—favoring one or another true version, they may be able to find shared vocabulary or some common ground on which they do agree (Pitkin 1993, pp. 101–102). But even this agreement does not necessarily mean that the truth upon which subjects can agree is the neutral objective truth, any more than the picture-puzzle view invalidates the truth of the naïve subject’s perception of the duck-rabbit as a duck (or a rabbit). The upshot, as J. L. Austin (1962, p. 101; cited in Pitkin 1993, p. 102) puts it, is that in some cases—indeed, precisely the ones that tend to be of greatest interest to us qualitative social scientists—there may be “no one right way of saying what is seen,” because “there may be no one right way of seeing it.” In this vein, what we analysts call objective truth always depends on larger contexts—what objectivity means in a certain era or place (including our own), what the scholar is trying to figure out, in what language games her inquiry is nestled.
The second point follows from the first and deals with the issue of replication: Because meaning-making is a social activity, its practices are replicable in the sense that political scientists care about replication—when they do. Subsequent researchers can go to the field and even if they do not talk to the same people or have different experiences talking to the same people, they can nevertheless be attentive to the range of possible interpretations relevant to a particular phenomenon under study. One’s own individual experience of marriage to man X may not be replicable, for example, but the social convention of marriage is. Such scholarly analyses, which are always less (as well as more) than the experience of the subjects themselves, also make sense only within socially accessible (and therefore contestable) standards, including standards for what counts as a fact and what does not. “Facts,” as the literary theorist Stanley Fish puts it, “emerge only in the context of some point of view” (1980, 338). Even something as seemingly straightforward as George Washington having died on December 14, 1799 presumes a specific world of language in which the Gregorian calendar has authority and death is understood to mean the cessation of life on earth. Or, to return to Hanna Pitkin’s words (, 1993, p. 178), “empirical investigation presupposes conceptual definition,” and conceptual definition requires what Wittgenstein calls a “life world.”
Let me combine these two points and suggest the sorts of confusions that can arise in conversations about transparency in the discipline: Proponents of what can be called science in political science often ask interpretive social scientists how they can trust themselves to know that what they propose as truths are true. One of the reasons that it is so difficult to respond to such a question is that it is hard to know what aspect of the claim-making the skeptics are referring to. Is it the conceptual claim? The causal argument? The observations? The presupposition of objectivity—which it has been the point of this memo to undermine—underlying such questions papers over not only the philosophical contestation about what objectivity means (see Daston and Galison 2007), but also the ways in which claim-making works, the pertinence of the particular life world in which such claims come to have authority, and correspondingly, the ways in which there are many different kinds of facts.
I’ve written about these matters elsewhere, so let me end with the third and likely most important point, which is not about political scientists talking past one another. This is the matter of ethnographic commitment and the vulnerability of human subjects. In my work it has been important to write about the people with whom I have interacted in the context of my ethnographic-oriented fieldwork in ways that far exceed what would objectify them as native informants. As the anthropologist Kaushik Sunder Rajan writes, such a categorization would constitute a “betrayal” of the solidarity I feel with them—including the solidarity to disagree, to judge, to be surprised, angry, even repelled. Writing from a situated perspective means not succumbing to titillating curiosity about subjects who find themselves violated and exploited by the conditions in which they are living. I do engage people as sources of ethnographic knowledge, but I want to do so, as Sunder Rajan (following Spivak who follows Derrida) enjoins us to, by answering the “ethical call to vigilance about how one objectifies the other whom one seeks to understand.” This orientation calls for cultivating curiosity, reflexivity, and enduring commitment (all from his forthcoming Pharmocracy, p. 59).
In regard to DA-RT, and in following IRB protocols, not objectifying people means protecting them—to the best of my ability—from harm. Turning over field notes is a red line I will not cross. Nor, returning to points one and two, would turning over field notes give my scientifically minded colleagues the epistemic security they seek. What does seem doable—even salutary—is to provide links or access to primary source material when it is public so that others can help judge whether my, or any other scholar’s interpretation, is cogent. To take an easy example, my analysis of Nathaniel Hawthorne’s The Scarlet Letter is enhanced, not simply by brilliant interpretations produced by Lauren Berlant (among others), but also by reading The Scarlet Letter. To the extent that Andrew Moravcik’s “active citation” means simply linking citations in the text of one’s own work to original material, this process allows readers who do not have access to the original data—and in the contexts I know best, this data includes publicly available writings and various popular culture artifacts from the global south—to join me in the process of theorizing about this material. I take this to be part of what collegial solidarity entails, placing various public “archives” in conversation with one another, allowing scholars to forge connections to and make political judgments about a range of social scientific phenomena. Even this valuable exercise should be an invitation rather than a requirement, in my view, a mode not of overburdening the researcher but inviting her to broaden her communities of argument.