Beware of allies! Notes on analytical hygiene in actor-network account-making

Qual Sociol DOI 10.1007/s11133-013-9256-6 Beware of Allies! Notes on Analytical Hygiene in Actor-Network Account-making Jan-Hendrik Passoth & Nicholas J. Rowland # Springer Science+Business Media New York 2013 Abstract In science and technology studies (STS), reflexivity is not the foremost political or ethical concern that it is for some postmodernists, feminists, anthropologists, or those earnest students of Bourdieu. For us, reflexivity is a practical methodological concern. When reflexivity is raised in our scholarly communications it is, without irony, about crafting scientific communications (i.e., scholarly accounts like articles or books) reflexively. This paper therefore is an actor-network account of making reflexive actor-network accounts, specifically, in the process of writing-up qualitative research findings. It is a paper about research. It is a paper about the research process. As our empirical contribution, we report on research we previously conducted and about the subsequent steps we took toward a (publishable) way of reporting it. We are trying to honestly disclose how the process of preparing a reflexive account is more than merely a matter of cleaning- up the messiness of data, but also, and perhaps foremost, a process of finding, aligning, and occasionally distancing our accounts from our allies—in our case, actor-network theory (ANT) and reflexivity. Keywords Actor-network theory . Reflexivity . Qualitative research . Grounded theory Reviewer #1: This is a paper that addresses the issue of reflexivity in science [and technology] studies by a case study of the fate of a paper written by the same author.1 In science and technology studies (STS), reflexivity is not the foremost political or ethical concern that it is for some postmodernists, feminists, anthropologists, or those earnest students of Bourdieu. For us, reflexivity is a practical methodological concern. When 1 Our guest editors suggested that we do this during the final round of reviews. Guest editors: “We encourage you to commence the paper with a punchier opening paragraph - perhaps with a particularly telling comment from a reviewer.” Reviewers themselves appeared cognizant of this possibility too, even in the first round; Reviewer #3 wrote: “Boy I hope I don’t say anything dumb enough to have my comments featured in the next version.” J.<H. Passoth (*) Technische Universität Berlin, Berlin, Germany e-mail: [email protected] N. J. Rowland The Pennsylvania State University, Altoona, USA e-mail: [email protected]  Qual Sociol reflexivity is raised in our scholarly communications it is, without irony, about crafting scientific communications (i.e., scholarly accounts like articles or books) reflexively. Chief concerns in this line of inquiry include the twin issues of “messiness” and “analytical hygiene,” which guide and provide purpose to our discussions. Only occasionally, however, are these discussions anything more than analytical, and so, in this article, our contribution is to provide an empirical case for consideration by our readers a case study about the fate of a paper that we wrote while trying so hard to be reflexive. Hidden within, we contend, are the principal advantages of seeing reflexivity as a practical methodological concern: our re- search, rather than political or ethical in character, can be empirical. To this end, we employ actor-network theory, an approach so close to the heart of STS that they are sometimes taken to be synonymous with one another, to direct our research on reflexivity. Introduction This paper is an actor-network account of making reflexive actor-network accounts, specif- ically, in the process of writing-up qualitative research findings. It is a paper about research. It is a paper about the research process. As our empirical contribution, we report on research we previously conducted and about the subsequent steps we took toward a (publishable) way of reporting it.2 We are trying to honestly disclose how the process of preparing a reflexive account is more than merely a matter of cleaning-up the messiness of data, but also, perhaps foremost, a process of finding, aligning, and occasionally distancing our accounts from our allies–in our case, actor-network theory (ANT) and reflexivity. The paper is not an abstract meta-discussion of ANT, or a commentary on any other post- pluralist solution for finding truth. Instead, we present an empirical case study: an instance of blending a data analysis strategy with an account-making method while also attempting to do ANT research “after method.”3 Coming from the small world of science and technology studies (STS), we conduct research on the infrastructure of the state. Analytically, we build on an actor-network approach to state theory and have tried to capture the multiplicity of dispersed and integrated practices that make, shape, and then re-shape the infrastructure of contemporary states and their related art of governing (Passoth and Rowland 2010, 2013). Recently, we studied the project-work that goes into establishing a new platform for digitally enhanced health technologies–partly because it was one of the arenas in which game- changing infrastructures, techniques, and practices are envisioned and invoked and partly because eHealth was a living example of contemporary stateness, namely, a post- foucauldian, neo-liberal vision of health care (Foucault 1977, 1991; López and Domènech 2009; Mort et al. 2003, 2009; Passoth and Rowland 2009, 2011; Rose 1999a, 1999b; Webster 2002; Winthereik et al. 2007). From that work we build this new paper. In this new paper, we will examine the transformations we made in a previous analysis based on grounded theory. Our previous analysis followed the paper trail left behind by The eHealth Initiative (eHI), an organization in the United States whose mission is to “drive improvement in the quality, safety, and efficiency of health care through [use of interoper- able] information and information technology” (eHealth Initiative 2009). eHI’s aim was to 2 We shift from present tense to talk about the paper you are reading now and past tense to talk about the papers we wrote in the past for sake of clarity. 3 See Law (2004) on the idea of re-inventing methods to deal with the fluidity and multiplicity of the social world. Qual Sociol engage and integrate multiple and diverse stakeholders—including hospitals and other health care organizations, clinician groups, consumer and patient groups, employers and purchasers, health plans, health care information technology organizations, manufacturers, public health agencies, academic and research institutions, and public sector stakeholders (eHealth Initiative 2009). We documented the state-related infrastructural changes eHI (and other enrolled allies) were putting forth to transform the delivery of health care in America. Our previous research project went through three major transformations: we presented it twice, once at the 34th annual meeting of the Society for the Social Studies of Science (4S) in Washington, D.C. (Passoth and Rowland 2009) then again at the 106th annual meeting of the American Sociological Association (ASA) in Las Vegas, Nevada (Passoth and Rowland 2011), in substantially different forms, and now it is under review in a near done draft. Data in this paper, which is about our previous paper(s), will take the form of our former drafts and our efforts to revise and defend them. To be absolutely clear from the beginning: This is not a paper about how hard it is to get a paper through the publishing process (Merton 1968). Nor do we intend it to be a comment on peer-review as an appropriate or inappropriate mechanism for sifting through papers de- serving of publication.4 Also, by presenting our story, we do not intend to say anything about the quality of our previous research or anything about its potential. In contrast, we are trying to present something practically reflexive. We start with an account not unlike Latour and Woolgar’s (1979) ethnographic account of the Salk Laboratories, which was an ethnographic account of putting together ethnographic accounts, or, what Latour (1988) would later call an exercise in “meta-reflexive” account-making. We then attempt to go beyond our previous understanding of reflexivity toward a self-exemplifying “infra-reflexive” account in this paper. One way to envision Latour and Woolgar’s contribution to the social sciences might be to say that they showed how they did what they did without the excessive hygiene all too common in scientific accounts. We aim to get at that “messiness” from another angle: We treat reflexivity as our subject of empirical study rather than merely a methodological imperative. In doing so, we try hard to avoid a predictable but somehow still tempting trap: As we shift away from cleaning-up accounts and toward engaging their messiness, one might get the terrifically calamitous impression that while cleaned-up papers are artificial, reductive, and, thereby, not scientifically honest, it is the messiness that offers a more authentic, comprehensive, and, thereby, scientific account. This is simply wrong. It is also a fatal future direction for qualitative analysis. The twin processes of cleaning-up and leaving some of the messiness behind are both effects of weaving a network around your assumptions, data, and related claims In the most general sense, our analysis (of our previous analyses) is set against the backdrop an old debate in STS that when controversies emerge regarding truth, the truth of the claim under scrutiny cannot be used to justify its “facthood” and, as a result, other techniques must be utilized to reach closure on the controversy (for case studies, see, e.g., Brante and Hallberg 1991; Guice 1998; Klintman 2002; Rappert 2001; Reno 2011; Richards 1991). We make the same argument here for reporting methods and data analysis. Getting methods (and/or data analysis reporting) “right” or “wrong,” “reflexive 4 That is not to say that there is not a literature on this topic worthy of consideration. For example, critic of peer review, medical doctor, and editor of The Lancet, a British medical journal, Richard Horton (2000, 148– 9) writes, “We know that the system of peer review is biased, unjust, unaccountable, incomplete, easily fixed, often insulting, usually ignorant, occasionally foolish, and frequently wrong.” Or, consider Lee and Bero’s (2006) position, which they argue in their Nature article, “Peer review is not currently designed to detect deception, nor does it guarantee the validity of research findings.” Still, these considerations, while paramount to the scientific enterprise, are outside of the scope of this article.  Qual Sociol enough” or “too reflexive,” has a lot to do with other factors that are not at all about being right or wrong, reflexive or un-reflexive. Our analysis in this paper is a story about these “other techniques.” And yet, it is explicitly not a story about rhetoric, persuasion, or even mutual adaptation of the writing and review processes.5 It is a story about enrolling some allies and distancing others. Getting it right, we contend, is a struggle in which just the right distance between allies must be achieved in practice, although, as we shall see, not in theory. We begin by examining a germ of reflexivity fit for actor-network account- making. On Reflexivity6 Reflexivity has, for a long time, been a cornerstone of actor-network account-making. Adopted from the ethnomethodologists and masterfully adapted in Laboratory Life, Latour and Woolgar (1979) invoked reflexivity when they, as (social) scientists, went to study (laboratory) scientists. In that context, they used reflexivity “to refer to the realization that observers of scientific activity are engaged in methods which are essentially similar to those of the practitioners which they study,” but the authors cleverly side-stepped the basic issues of scientists observing other scientists, and instead drew analytical attention to a “latter part of the discussion,” namely “our awareness of certain methodological problems which we face in the constriction and presentation of our discussion” (1986, 30).7 By “discussion,” the authors were referring to those processes by which observations get turned into written “accounts” (30). Latour and Woolgar (1986, 245) also adopted reflexivity to expand the reader’s understanding of how manuscripts are constructed by social scientists; just as the laboratory scientist must construct scientific accounts from their chaotic findings, so must the qualitative researcher “create order” in their otherwise chaotic field site. When scientists of any ilk finally start writing, a lot of that seemingly intransigent messiness gets left out of the final manuscript. But the messiness is important. Some of it should be retained. On point, Jonas Salk, author of the introduction to Laboratory Life, famous immunologist, and director of the institute where Latour conducted his fieldwork, claimed that Latour and Woolgar’s use of reflexivity was their “chief accomplishment” because it allowed them to “reveal how ‘human aspects’” got “excluded from the final stages of ‘fact production’” (13). But Salk’s praise also put these sociologists studying 5 There is a terrific literature about how to write social scientific accounts (see, e.g., Becker 1986; Emerson 2011; Silvia 2002; Weiss 1994), especially those dreaded dissertations (see, e.g., Bolker 1998; Krathwohl and Smith 2005; Single and Reis 2009), as well as a literature in the broader academic enterprise about how to write non-fiction well (Zinsser 2006). These are valuable resources, which we routinely utilized; however, commenting on them or their utility is outside of the scope of this article. 6 We have significantly expanded this section since the first round of reviews. In our letter back to the editor, we wrote: “We initially thought it would be irritating to have a section on the debate on reflexivity in qualitative research, science studies and ethnomethodology […] But after having the opportunity to position the paper in those debates we are convinced that this new section makes the point of the ‘practical problem of reflexivity’ even stronger and it helps pointing out that the paper is not a critical comment on the review process, but in fact an honest appraisal” (Letter to the editor, May 26, 2013). Maybe now the section is a little too long; however, we can let readers decide how much reflexivity is enough. 7 No doubt, the close reader will have just noticed that while only a few lines above we cite Latour and Woolgar’s original 1979 book, published by Sage, we now shift and instead utilize the reprint, published in 1986 by Princeton University Press. We do this purposefully. ANT buffs may already know this story, but the 1979 book was titled Laboratory Life: The Social Construction of Scientific Facts; however, the authors were later so dissatisfied with the subtitle that they renamed the subsequent 1986 reprint Laboratory Life: The Construction of Scientific Facts in defiance of the term “social” in STS, sociology, and conceivably elsewhere. We generally agree with this assessment and protest, and use the 1986 edition for the rest of the paper. Qual Sociol laboratory scientists in a tough spot. Latour and Woolgar could not deliver crisp, orderly accounts of what they observed ethnographically while, at the same time, criticizing lab scientists for masking all the messiness and errors (i.e., those “human aspects”) that go into the production of any scientific account. As Latour and Woolgar (1986, 275–6) put it, revising our: epistemological preconceptions about science raises awkward questions about the nature of its social analysis. Can we [as social scientists] go on being instrumentally realist in our own research practices while proclaiming the need to demystify this tendency among natural scientists? The question was surely as provocative then as it is now. However, as we search for answers, we encounter this fact: As account-makers, we are not entirely free to make accounts in any way we deem fit, and this is true even when we attempt to demystify the accounts of others. Still, on balance, reflexivity also afforded Latour and Woolgar (1986, 283) enough analytical space to recognize that “our writing is conventionally constrained by the use of report-like formats.” They chose to explore alternative forms of reporting as compared to traditional report- like formats. They provided readers with an incomplete or unpolished ethnographic account of what they observed sitting in the corner of Salk’s laboratories. Referred to as an “anthropology of science,” they mindfully offered readers a “preliminary presentation of empirical materials,” empirical material which one might expect from an “intrepid explorer” returning from the field with “a body of observations” about indigenous peoples (27, 28). As such, Laboratory Life bears the marks of its maker; it is unavoidably and insolubly honest and incomplete. Errors are left un-erased. Failed experiments and unsightly observations go nakedly reported. Their challenge to conventional sociological account-making did not go unnoticed by critics. In reviews of the book, the authors’ form of reflexivity was challenged on grounds of relevance and its general unfriendliness to readers. The 1986 paperback edition of Laboratory Life contains a seldom read but intriguing postscript wherein Latour and Woolgar respond to these criticisms directly. Of particular note, Westrum (1982, 438) criticized the format and presenta- tion of the empirical materials stating that reading the book was “rather like taking an extremely bumpy ride over fascinating terrain,” and, thus, “the book’s major defect … [was its] lack of a sense of unity.” Latour and Woolgar (1986, 276) responded, claiming: Westrum speaks of a lack of a sense of unity, the lack of continuous action and the relative incoherence of the narrative. But our aim was precisely to avoid giving the kind of smoothed narrative characteristic of traditional constructions of the “way things are.” Truly, the move toward a reflexive ethnographic account saved them from the two most pressing problems for STS at the time. First, back then, scholars were no longer satisfied with scientists’ and engineers’ descriptions of science and technology “as usual,” which usually amounted to little more than retrospective accounts; therefore, social scientists began collecting first-hand observa- tions of science and engineering in action in order to go beyond traditional accounts scripted by the myths and ethos of the scientific and technological enterprise. Second, once in the field, there was the question of what to take-for-granted. Latour and Woolgar successfully used reflexivity as a way to situate what they observed as foreign-from but next-to themselves, the observers, as if a reminder to see even the most taken-for-granted aspects of scientific practice as permanently exotic and forever new. In overcoming those problems, reflexivity was their ally, and they treated it more like to an attitude toward inquiry and account-making than a strict rule for observation or analysis. However, reflexivity—now a cornerstone of qualitative research (Adler and Adler 1996; Alvesson and Sköldberg 2000; Findlay 2002; Luttrell 2010; Steier 1991; Wichroski 1996), not only in the actor-network theory (ANT) tradition (Latour 1988), but also in feminist studies (Naples 1996), social studies of science (Ashmore 1989; Ashmore et al. 1989; Lynch 2000;  Qual Sociol Woolgar 1988a, b), medical sociology (Findlay and Gough 2003), organizational studies (Chia 1996a, b; Hardy et al. 2001), and elsewhere—is not at all a harmless ally from ethnomethodology or cultural anthropology to be enrolled at will by the analyst, even though Latour and Woolgar (1986, 18, 30, 273–286) seemed to imply as much in Laboratory Life. Of course, reflexivity is the beating heart of any ANT account. But this heart was born of a strange marriage, and, in Reassembling the Social, his retrospective introduction to ANT, even Latour (2005, 54–55) admits it; it is, he writes: accurate to describe ANT as being half Garfinkel and half Greimas: it has simply combined two of the most interesting intellectual movements on both sides of the Atlantic and has found ways to tap the inner reflexivity of both actor’s accounts and of texts. When dealing with reflexivity in practice, very soon, things get messy—swampy, even (Findlay 2002). Tapping the reflexivity of both actor´s accounts and texts seems like a harmless endeavor (Latour and Woolgar 1986); however, now—after reflecting on numerous texts written about how to deal with the problem (Ashmore 1989) and the opportunity (Findlay 2002, 212; Findlay and Gough 2003) of reflexivity years after Latour and Woolgar’s Laboratory Life, especially in the social studies of science (Ashmore et al. 1989; Lynch 2000; Woolgar 1988a, 1988b)—we are not so sure. At least since Clifford and Marcus’ (1986) initiative to strengthen awareness about the need for reflexivity in writing research (and culture), Ashmore’s (1989) playful and insightful dissertation challenging researchers in the social studies of science to live-up to the critical, reflexive standards that they employ in their work on their work, or ethnomethodology’s slow march from the outskirts of participant observation and conversational analysis into the inner sanctum of sociology’s canonical writings (Lynch 2000, 43; see also, Garfinkel 1967; Garfinkel and Sacks 1970), we are increasingly aware of both the need and the impossibility of making reflexive accounts that count as such. We are told that we need to “celebrate the monster” (i.e., reflexivity) by Ashmore (1989, 88), especially if we assume, broadly speaking, that our “patterns of explanation would have to be applicable to sociology itself” (Bloor 1976, 4–5). But this is not so simple because reflexivity is, in theory, a quite chameleonic monster. The work of Ashmore (1989), Mulkay (1985), Woolgar (1988b), and many others (e.g., Latour and Woolgar 1986), is, according to Latour (1988) and, later, Chia (1996a, 1996b), meta-reflexive, best defined as: [a]n exercise in introspection [which] is usually concerned with improving the adequacy of the connection between analysis … and the objects [under analysis] … Far from raising any fundamental problem, this kind of reflexivity sustains and enhances the Scientific axiom of the research effort. (Woolgar 1988b, 22) Meta-reflexivity has been called a “sterile distraction” (Booth 2000, 1) and even “suicidal” (Adkins 2002, 337; Latour 1988, 169). Overt attempts by authors to be outwardly reflexive during account-making results in a number of predictable problems (Lynch 2000), which include, (1) typically unsuccessful rhetorical stabs at gaining “moral rectitude and epistemological panache” (Pels 2000, 2), (2) a clumsy inclination, especially among scholars in the sociology of science, to commit “self-exempting fallacies, which constitutes an embarrassing and public ‘failure of nerve’” (Pels 2000, 2), and (3) the unfortunate inability to distinguish the reflexivity-of- actors from the reflexivity-of-accounts (Czyzewski 1994). Each of these problems is doubly troublesome when (reflexively) giving accounts of the accounts given. In order to give an account of the accounts given, one has to start with the meat-and-potatoes of reflexivity. As our starting point, we agree with Michael Lynch (2000, 26, 47); while reflexivity is an “essential human capacity,” being reflexive in our scientific accounts “in general offers no guarantee of insight or revelation.”8 For the ethnomethodologist (and us, as ethnomethodology-semiotics 8 Emphasis is the author’s. Qual Sociol mongrels), nearly any plea regarding the benefits of reflexivity over an un-reflexive account is taken to be based on a misleading dichotomy and amounts to no more than a pious appeal for scientific credibility. On balance, though, quite often appeals to the value of overt reflexivity have been taken as a necessity for the production of any persuasive account in the social sciences to count as scientific at all (see, for example, Bloor 1976). There is simply no such thing as an un-reflexive (seemingly objective, untainted) account; according to Lynch (2000, 47), “the ‘essential’ reflexivity of accounts … is part of the same infrastructure of objective accounting no less than of self-conscious efforts to be reflexive.” Reflexivity is, on the whole, uninteresting, ubiquitous, and unremarkable, and surely not a “methodological virtue and source of superior insight, perspicacity or awareness,” except when “the reflexive relationship between accounts and accountable states of affairs can become vicious” (26, 34). What needs to be included in an account of an accountable state of affairs is already the result of a deeply reflexive practice. The fact that something is erased from an account—striking- through a reflexive statement about the creation of the text, a failed attempt at making a claim, or an unsuccessful meta-reflexive loop about how the circumstances of making the text influenced the potential of reflexive account-making at all—already refers reflexively to the conditions of making sense of the text. And, it is these conditions, and the reflexive positioning of a text to them, that can become “interesting” (Lynch 2000, 34). Focusing on these practical conditions of sense-making that “undo,” or, at least, manage “the deleterious effects upon its readers of being believed too little or too much” (Latour 1988, 166) is more akin to what Latour called “infra- reflexivity” than to any attempt to merely add (or subtract) reflexivity. Thus, for contrast, Latour juxtaposes—admittedly, only rhetorically (Pels 2003, 161) or semiotically (Lynch 2000, 34)—meta-reflexivity with infra-reflexivity in “The Politics of Explanation.” If, Latour (1988, 170) writes: meta-reflexivity is marked by an inflation of methods, infra-reflexivity is characterized by their deflation. Instead of piling on layer upon layer of self consciousness, why not have just one layer, the story, and obtain the necessary amount of reflexivity from somewhere else? … [and] just offer the lived world and write.9 In this context, reflexivity refers to “any text that takes into account its own production,” but it is utterly unremarkable to be reflexive, and, on this point, Latour (1988, 168) and Lynch (2000, 34) appear to find common ground, but for different reasons. For Lynch (2000, 26), because reflexivity is a ubiquitous characteristic of human (inter)action, any explicit attempt to add reflexivity guarantees its bearer no virtue during account-making. In cataloging extant forms of reflexivity—mechanical, substantive, methodological, meta-theoretical, interpretative, and even ethnomethodological—Lynch shows that every claim purporting epistemic privilege is revealed, in Latour’s terms, to be meta-reflexive, as they participate in the same scientific enterprise they seek to overturn or improve upon. The only exception is when “accounts and accountable states of affairs can become vicious,” but, at this point, “[t]he injunction to be reflexive … does not tell us how much, or how little, to [reflexively] mention” (Lynch 2000, 34, 46). “There is no one way to be, or not to be, reflexive,” according to Lynch (2000, 46), ergo, the answer to “how reflexive?” is to be found in the particulars of each circumstance of use. For Lynch and Latour, all accounts are equally reflexive, and, thus, even accounts of accounts of accounts (e.g., Woolgar 1988b) are no more or less reflexive than another because “the whole vertigo of self-reference stems from the very naive belief that the same actor appears in both the first … and last text” (Latour 1988, 169). At any rate, naïve meta-reflexivity constitutes no viable future direction for the social studies of 9 Emphasis is the author’s. Also, this distinction shares, as Adkins (2002, 336) notes, similarities with May’s (1988) distinction between endogenous and referential reflexivity.  Qual Sociol science anyways. In place of self-referencing forms of reflexivity, we need self-exemplifying methodological restraint and certain stylistic commitments, chief among them, the knowledge that “my own text is in your hands and lives or dies through what you will do with it;” but, more than that, we recognize that any deconstruction using meta-reflexivity to buttress its account must, in principle, fail “to talk about anything” (Latour 1988, 171, 173). So, how does one find a way to “maintain the necessary reflexivity without whirling helplessly in our efforts to outdo and outwit each other in proving that the other is a naive believer?” (Latour 1988, 170) We are left with a solid-state conundrum. No additional loop brings us closer to what we talk about. Even an account that does not refer to the circumstances of its production and the epistemo- logical or moral position of its author carries mundane reflexive marks of the context of its use and circulation. But, there is a way out, and, in writing-up qualitative results, we have all experienced it. As a practical matter, social scientists deliberately, undeniably, or unknowingly try to add reflexivity to their accounts. We know, in theory, they cannot, but, in practice, they do. Bit-by- bit—and, to some scholars, adding tidbits of reflexivity to their accounts is an endlessly honest and noble pursuit—until, suddenly, perhaps surprisingly, you have enough reflexivity, or, worse, too much. Audience members attending our presentations tell us as much during post-presentation discussion. Anonymous peer reviewers of our manuscripts may require as much during post-review revisions. Even extending this section on reflexivity is a result of such a requirement. We are all always obliged to draw the line somewhere. Implicitly or otherwise, we must all ask “how reflexive is reflexive enough?” And it feels as though that question should have a simple, concrete answer. And it has, but it is one that has to be given again and again each time a reflexive account is given. To learn about this seemingly un-learnable situation, one would need an empirical account of a relativist’s account-making: an infra-reflexive account of a meta-reflexivist’s account.10 Analysis of Analyses In 2008, we happened upon the “eHealth Initiative” (eHI), an American nonprofit non- governmental organization set on improving health on a national scale in the U.S. through new technology (eHealth Initiative 2009). According to Lehoux (2006, xvi), eHealth technology rarely refers to “a stand-alone device,” and, instead, is typically “a component of larger health care delivery systems,” which, in the case of ehealth, is constituted by but not limited to virtual medical teams, tele-medicine, citizen-oriented informatics, real-time mobile patient status updates, electronic patient record systems, and electronic prescription systems. We were fascinated by the complexity and openness we found and were intrigued by the paradoxical combination of both vagueness and concreteness of the setup that was envisioned in eHI’s program documents. As a result, our dataset, which we analyzed in the 2009 4S paper, the 2011 ASA paper, and the current manuscript under review, was created from publicly available documentation hosted on eHI’s website.11 From fall 2008 until 10 Why do this at all? If we examine this conundrum, then we might finally make good on Latour’s (1988, 173) lauded and laughed-at claim: Let us go back to the world, still unknown and despised. If you sneer at this claim and say “this is going back to realism,” yes it is. A little relativism takes one away from realism; a lot brings one back. Additionally, this approach is positioned to confirm or challenge Lynch’s (2000, 46) position that “[t]here is no one way to be, or not be, reflexive,” and, perhaps, help determine if Latour’s implicit insight was right: that self-exemplifying accounts must be, in some respects, meta-reflexive before being infra-reflexive stories, which do not add to, but accompany, in irreductionist terms, the account at hand. 11 Available at: http://www.ehealthinitiative.org/ Accessed: November 10, 2012. Qual Sociol summer 2009, we reviewed all of the documentation available at the time, which included 12 programmatic and technical documents ranging from two to 107 pages in length (averaging 35 pages). We generated general themes after reviewing all of the materials, parsed them down into specific codes, which we then used to analyze the four documents in our dataset that we considered the most robust. Data were coded in a variation of the Straussian variant of grounded theory (Strauss 1987). As a research team separated by an ocean, we were not able to do our initial line-by-line open coding in a setting of face-to-face interaction. In 2008, therefore, we conducted the first open codings separately, then used VoIP (Voice over Internet Protocol) to establish a complex set of shared but open codes. Axial coding was then undertaken separately a second time, based on the mash-up of codes established over VoIP. Finally, by early 2009, through selective coding—based on a review of the material a final time together over VoIP—we identified and organized our findings into major themes.12 Paper Presentation at the Society for Social Studies of Science, 2009, Washington, D.C. At that point, our data analysis section, labeled “Empirical Findings,” was organized according to three themes, which included sections devoted to the social, political, and technical facets of eHI’s proposed project. As it happens, that exploratory paper, which we subsequently presented at 4S in 2009, used some (we later found out) fairly unacceptable categories for presenting our data. The simplicity of our early work was not veiled in the slightest during this presentation. For example, our headings in the analysis section read: & “eHealth as a political project” [which we presented] & “eHealth as a technological project” [which we presented] & “eHealth as a social project” [which was incomplete and thus not presented] These inflexible—if not outright naïve—distinctions between “social” and “technological” aspects of a health care project were unacceptable and deemed less than convincing to the STS- steeped audience at 4S in 2009. For example, a fair criticism of our work at the time, from a source that will remain anonymous, could be paraphrased as such: Compared with ANT, this is a rather functionalist view assuming a number of “domains” [political, technological, social] already existing. From an ANT vantage point—more broadly from constructivist thinking—there are no such domains, since the technical, political, etc., are seen as constitutively emergent. Thus, while integra- tion may be an aspiration of the authors of e-Health documentation, it is hard to use as a starting point for a performative analysis aiming to understand how the field is “set.” Our analysis was deemed functionalist and we were really surprised that we seemed to have employed a functionalist approach inadvertently. This early analysis section was short, composed of 2,119 words. Our methods section was also boilerplate, consisting of only 352 words, which were chiefly devoted to describing our coding technique and defending (somewhat poorly) the grounded theory tradition; however, there was a small kernel of proto-reflexivity there, which we would later cultivate. 12 When we reformulated our data analysis for presentation at ASA in 2011, we followed the same process of returning to our data. Open coding was followed by axial coding until we arrived at a basic set of themes. The only difference being that when we initiated analysis the second time, we had in mind that rather than search for recurrent themes, we searched for evidence that supported a more cartographical account of our findings such that we searched for the “traces” left behind by health data from its origins in self-care into the broader socio-technical data scaffolding proposed by the eHealth Initiative.  Qual Sociol Table 1 Word count and change (as a percentage) between drafts Methods (word Change (as Analysis (word Change (as count) percentage) count) percentage) 2009 4S 0,352 – 2,119 – 2011 ASA 2,065 +586.65 % 3,046 +143.75 % 2012 under review 0,726 −380.39 % 2,993 −002.50 % Total Change 0,374 +206.26 % 0,876 +141.25 % In that presentation, we (Passoth and Rowland 2009, 3) stated that: This is one of the rare instances where we, as authors, will not follow Bruno Latour’s (1987) ontological and methodological imperative to “follow scientists and engineers through society.” The problem is that the American health care market is so fragmented (eHealth Initiative 2007). Therefore, following just one actor (i.e., a hospital, a patient, etc.) might be a mistake or potentially misleading. Instead, we elected to follow eHI’s paper trail because this organization is set on integrating the fragmented healthcare marketplace and is simultaneously composed of individuals from all over it meaning that eHI is simultaneously central and peripheral in the American healthcare marketplace. To save us from our previous fate as functionalists, we then reconsidered Latour’s imperative to follow the actors; to this end, we followed the lead of Mol’s (2003) The Body Multiple. Mol followed a disease, atherosclerosis, using ethnography, a method typically used to study humans in group settings. As a medical issue, atherosclerosis, the disease, was at once human and non- human; it was social, biological, and technical simultaneously, hence, Mol’s expert use of the multiplicity concept to sensitize readers to this important ontological facet of life as we know it. This, we honestly believed, would solve the problems we faced in our seemingly functionalist portrayal of the eHealth Initiative and would help us to present and elaborate on the strange combination of vagueness and concreteness that we tried to grasp by telling the story eHI three times—as a political, technological, and social project—in our first attempt at writing the paper. By blending the actor-network attitude of ethnographic “following” with our wish to conduct a content analysis of extant documentation, we ultimately determined that a more cartographic representation of our data would be appropriate, hence, analysis became less about uncovering common themes and instead took the form of map-building to show where health data travel. We also sought to be reflexively honest in our portrayal of this transfor- mation of our analysis section in our analysis section. Paper Presented at the American Sociological Association, 2011, Las Vegas, NV After 4S 2009 and after returning to original actor-network accounts like Laboratory Life, we recognized the central role of reflexivity in actor-network account-making and then reorganized the methods section and our data analysis section to be as reflexive as possible for our paper at ASA 2011.13 In this paper, we expanded the methods section, expanding it nearly six-fold in length to 2,065 words. As an attempt to blend the ethnographic attitude of empirical honesty with 13 An admittedly simple but telling way to illustrate the changes over time is found in word count. Thus, for interested readers, we report word counts as well as percentage change between drafts presented in a table format (Table 1). Qual Sociol content analysis, we reflected plainly and ingenuously on the changes we made from the previous exploratory work. We (Passoth and Rowland 2011, 12) stated: We do not argue for making grounded theory the basic and seminal analytical technique for doing research. On the contrary: in some cases it could be the worst combination to choose. Coding, at least with a focus on constant comparison and on reduction, can as well lead to homogeneity and to a tendency to discard findings that do not fit into neat coding schemes [which we learned from experience]. Post-ANT analytics encourage researchers to do the opposite: the method has to be used as a heterogeneity enhancing procedure. Despite our best efforts, in our first coding attempt, we failed; we failed to keep the coding scheme open as long as possible resulting in a reduction rather than expansion of codes, themes, and categories. As a way of introducing an element of self-reflexiveness to content analysis, we learned from the mistakes discovered in our first attempt to understand and interpret the documents and have re-arranged our codes and analysis, loosening some codes and dropping ready-made categories. Originally, we cataloged the exigencies of eHealth and eHI as concurrent political, technological, and social projects—and we thought that what we were observing was a truly heterogeneous “lashing together” of human and nonhuman barriers, mediators, and drivers (Latour 1991). However, upon completion, we realized that we had made the exact mistakes we tried to avoid; we generated a monolithic and clear cut story too clean and bereft of multiplicity, and adopted, uncritically, the ready-made categories to interpret our data. In retrospect, the lashing-together was not at all the classic ANT story of a network growing longer and stronger. In our methods section, we also discussed two issues we consider paramount to any use of an actor-network approach for conducting research, whether ethnography is adopted or some other method. First, we asked whether or not “certain methods are more appropriate” for actor-network account-making, and tentatively concluded that: ANT is not a method; however, a core directive to “follow the actors” is easily misinterpreted as a methodological imperative (Latour 1987). It is the assumption that while in-the-field a social scientist “is likely to be wrong about his or her assumptions when encountering ontological work in practice,” which originally made ANT a distinctive scientific enterprise (Gad and Jensen 2010, 73). Ergo, ANT is anti- method. It is the antidote to adopting slews of assumptions with which scholars see those things that they study. And although ethnography, especially multi-site ethnog- raphy (Marcus 1995, 1998), seems to be a good candidate at first glance (Hine 2007), it is not the method for doing ANT. It is exactly because ANT cannot be yoked with a particular method that this must be done nearly anew every time. (2011, 10) Second, we asked, if we are to follow the actors, then “[w]hich actors should we follow to avoid false interpretations and arbitrary conclusions?” and concluded that: Emphasis on multiplicity rather than plurality also provides a possible answer to this question. How to decide which data to gather, which actors to follow, and, in doing so, how to avoid writing not only risky (Latour 2005) but arbitrary accounts? Thinking of the answers to questions like that not as objects [in themselves] but as “boundary objects” (Star and Griesemer 1989), and the work involved in enacting them as “boundary work” (Gieryn 1983), a multiplicity of things (and people) to a multiplicity of people (and things) keeps the ANT researcher from assuming that there is any way to get the network absolutely right. As long as we do not assume that a better picture of the network can be gained by  Qual Sociol adding more and more perspectives, the problem of where to begin following actors is not so big because following different actors will not bring us closer to some indelible truth. By studying efforts to shape the multiplicity of networks by problematizing and punctualizing or enrolling and dissenting (Callon 1986), one can observe how networks become multiple or balance fragmentation and how others are made to stabilize amid flux or become taken- for-granted despite being multiple. (2011, 11) Next, we presented our reformulated data analysis, which was transformed, now explic- itly not about those three factors. Instead our analysis took on the likeness of a cartographical account that traced “health data” from its diffuse network origins in self-care all the way to the enactment of a state-like entity trying to make the nation healthier with better data. Our headings in the analysis section now read: & “Lashing together (in the name of self-care),” which was about the path that eHI saw for health data that originates in diverse forms of self-care in diverse non-local environs. & “Integrating (and distributing),” which traces health data from its diffuse origins through electronic patient records systems into hospitals, which are organizational hubs for health data in eHI’s vision. & “Interoperable (data is a form of public health),” which shows how data, from their diverse sources, enter into interoperable systems that homogenize those data in order facilitate their use outside the confines of the hospital or even immediate care for use among public health officials and government bureaucrats, all of which begins to reveal “the state” (or traces of it) as it uses health data “to improve health and health care for Americans” (eHealth Initiative 2007, 12). While the analysis section was met with some praise by audience members, the reflex- ivity in the methods section attracted some criticism, namely, that there was simply too much reflexivity. To summarize this concern, members of the audience and those that reviewed the paper outside of this context repeatedly voiced two criticisms: 1. Knowing about the mistakes that you made does us, as an audience, no good; keep the errors to yourself because all they manage to do is “muddy” the writing, distract the reader, and ruin your credibility in the process of making your empirical contribution. In effect, lie to us. Erase the past and just let those data speak for themselves. 2. Seeing the errors made in previous drafts is too personal to have in this sort of scientific analysis; it humanizes the account and distracts from the empirical and theoretical aims of the paper. In effect, get out of your own account and just let those data speak for themselves. It is worth mentioning that these comments come from individuals deeply indebted to the actor-network tradition, and we were surprised that even they suggested that this piece was too reflexive for its own good. Still, we submitted the paper to a peer-reviewed academic journal after only minor editing from the 2011 ASA paper, as we were confident that the reflexivity would find more favorable company in blind review … Paper Currently Under Review, Received “Revise and Resubmit,” Awaiting Final Verdict … but that did not happen. During the peer review process, the reflexive discussion that we added to the methods section for ASA 2011 was criticized again. Concern over the honesty Qual Sociol resulting in a loss of credibility was not voiced. In contrast, for sake of saving space and enhancing lucidity, the reflexivity discussion, we were told, must be drastically reduced. For example, to paraphrase one reviewer, the paper in its current form is dense and contains a number of unnecessary detours, especially related to methods, which ultimately distract from the main thrust of the paper. A related concern, but from another reviewer, suggests that the paper was too long, and could be truncated provided we (the authors) were willing to drop the methodological detour to justify content analysis, given that ANT is neither a qualitative method in itself nor merely some alternative theory about the nature of technological devices. To this end, we reduced much but not all of the reflexivity in our methods section, shrinking it to 726 words. Our analysis section, which was not criticized, was shortened modestly to 2,993 words, and remained essentially intact after being reformulated for ASA 2011. And so, the paper stabilized, and we await the final verdict to see if this new arrangement of data and method was reflexive enough. Reflecting on Reflexivity As we transformed a presentation from the microcosm of professional conferences into a working manuscript for academic, peer-reviewed publishing, we encountered remarks about how reflexive we needed to be during our account-making, in particular, in our methods section. After delving into the reflexivity literature, we concluded that no “amount” of reflexivity could have made our account more reflexive because, in addition to reflexivity being part of the intransigent character of all forms of account-making, overt pleas for the epistemic virtue of adding or subtracting any form of reflexivity is an immediate dead-end for the analysts and a long-term dead-end for whatever (inter)disciplinary homes they inhabit (Ashmore 1989; Latour 1988; Lynch 2000). Our empirical analysis confirmed each of these insights. Still, the question, “how much reflexivity was enough?” seemed all too real as we accepted critiques of our presentation(s) and received reviews of our paper. For us, the practical problem was, how do we settle this obviously irresolvable suggestion- turned-requirement? Latour (1988, 175) suggests that the solution ought to be found in the “[h]ybridization [of genres] instead of [within academic] disciplinary bound- aries;” however, this insight appeared to fail us precisely when we needed him to be right (i.e., in order to publish professional work in ways that register on promotion and tenure documentation at the universities we inhabit). Interestingly enough, with this paper about the other presentations and papers, we have encountered a similar bind regarding the right amount of reflexivity to have appear: during reviews of the paper you are holding, one reviewer insinuated that our take on reflexivity was, ironically, not reflexive enough. Another solution may be to go back to the beginning. If we were to follow the early advice of Latour and Woolgar (1986, 28, 27, 283) from Laboratory Life, which appeals to us as post-ANT scholars (on post-ANT, see Gad and Jensen 2010; Law and Hassard 1999), then we would have simply provided “a body of observations” comprised primarily of “preliminary … empirical materials” which would not be “conventionally constrained” by the book format. The promise being that a kernel of honesty could be harvested from leaving all that mess in the text, that the reflexivity of the account could be preserved so long as we did not subtract too much from the manuscript at the outset. In a similar vein, Woolgar, in the foreword to  Qual Sociol Ashmore’s (1989, xvii-xx, xix) The Reflexivity Thesis, appears14 to beg Doug Mitchell, then Senior Editor of the University of Chicago Press, to publish Ashmore’s doctoral thesis ““as is”” rather than see it “‘turned into a book’” for fear that something would be lost during the subtraction; that its “‘thesis-ness [would] be exorcised.’” The point was to preserve the reflexive play between “‘thesis-as-argument’” and “‘thesis-as-an-occasioned-academic-product.’”15 Again, by not subtracting, authors (and foreword writers) appear to—by some feat of reverse alchemy—add to the reflexive potency of the account. It has been argued that Woolgar and Co. had more latitude to publish highly-reflexive, unorthodox accounts—for instance, more leeway than scholars attempting to publish in more conventional peer-reviewed journals—explicitly because of the book format they published in; however, as these two above examples illustrate, the book format did not appear to facilitate the presentation of their accounts as they claimed from within their accounts. What have we learned pursuing our apparently naïve approach to reflexivity? Rather than unfinished or exorcised manuscripts, we found ours well-worn. An old insight in STS informs our answer. Controversies regarding truth are rarely settled by appealing to the inherent truthiness of the claim under scrutiny. As it happens, other techniques must be utilized to reach closure, and in our case, the level of reflexivity in our manuscript was governed by concerns regarding length, which we met, and on the grounds of clarity, implying that these reflexive detours detracted from the main thrust of the argument. We did not consider these reflexive considerations to be orthogonal to the argument or our analysis. In contrast, we envisioned them as enlivening the manuscript. And yet, apparently we breached a threshold perhaps only tacitly understood by our reviewers; we were, it seems, either too present or present too long in our own scientific account. Something had to be erased. But not entirely. Like a line erased in a drawing, the image still indelibly bears something of that line, if only evident as a residual mark. We contend that reflexivity in actor-network account-making is something of a similar process. Reflexivity gets added and erased, underscored and covered-up, and this back-and-forth process of adding and then subtracting reflexivity, in the end, results in what an artist might call a “well- worked” canvas. It shows signs that it was made, but not its maker per se. No doubt, whatever we have learned from this reflexive account of reflexive account-making is surely contingent, and probably only reflects the tacit understandings of certain scholars in a particular subfield at a particular moment in history. Still, we consider this notion of the well-worked paper, which is not necessarily the polished piece, but one that has borne numerous transformations not yet completely erased, to be a tenable way to balance the desire for methodological reflexivity on behalf of the analyst with the practical realities of expectations among readers and reviewers regarding how much analytical hygiene is necessary in a scientific paper. The well-worked canvas metaphor might just as well be thought of as a “tuned-in” manuscript. After getting feedback that the manuscript was not reflexive enough, making respective changes, followed by feedback that it was now too reflexive, and then making those changes, the result is a “tuned” paper; however, it is important to go beyond the meta- reflexive understanding of account-making, even if there is a practical significance for scholars actively publishing. Authors can add overt traces of reflexivity to a manuscript just as easily as they can erase any lingering traces of its leftovers from previous drafts. It is, at this point, important to recognize that in doing so the manuscript does not become special; 14 Of course, we are left only with appearances, for we know not what really happened; we do not know if the foreword was a joke, the truth wrapped in a joke, or the truth – not that it matters for the present argument. 15 Woolgar’s penchant for scare quotes here speaks to how special this move was to be. To stress this, we decided to include his quotation marks within our quotation marks so that readers can see firsthand his instructive scare quotes. Qual Sociol you just get a tuned-in paper that passes through the publication process. It is not more reflexive than previous drafts; it is not less reflexive than previous drafts. Additionally, we also showed how, once stabilized, even intentionally reflexive papers leave few obvious traces of their revision over time–especially revisions enhancing or depleting a paper of its reflexivity. We observe that even attempts to avoid the trappings of excessive analytical hygiene in the write-up of scientific accounts, a position which Latour and Woolgar (1986) do so much to cultivate, seem doomed to inevitably result in what they seek to avoid. Reflexive accounts get revised to be reflexive in just the right ways. Unwelcome reflexivity gets erased from manuscripts, and the papers themselves become silent partners in scientific tradition of their production. Ironically, even scientific accounts that aim to be reflexive fail to show readers the practical, everyday doing of science that lead to reflexive account-making. Thus, you cannot learn to be reflexive by reading reflexive accounts per se, just as one cannot learn to do science by reading scientific accounts of experiments. Bears As we reflect on the role of reflexivity in actor-network account-making, it is instructive to return to the British children’s story “Goldilocks and the Three Bears.”16 One version goes: Three good- natured anthropomorphic bears—papa bear, mama bear, and baby bear—reside in a house in the woods. Each bear owns a chair and bed custom-designed for their particular size. One day, after preparing a meal, the bears decide to take a walk as their porridge cools, and this is when Goldilocks, a lovely, but ultimately criminal little girl, happens upon the house and decides to break in. She eats one bowl of porridge, but considers it too hot, then, not satisfied, she eats one of the other bowls, finding it too cold, but the last bowl is “just right,” and this satisfies her so much so that she needs to take a break. She sits in each of the bear’s chairs, selecting one to sit in, which she subsequently breaks. The destruction of property makes Goldilocks tired, so she goes upstairs for a nap, testing each bed before settling into the bear’s bunk that was “just right.” When the bears return from their walk, Goldilocks, upon being discovered, flees the scene. These days, the Goldilocks story is typically framed like this to children, foremost a story about the progressive discovery of what is “just right.” One way to conceptualize our case study in reflexivity would be to see it like this old children’s story. While 4S 2009 was not reflexive enough and ASA 2011 was too reflexive, the manuscript under review got reflexivity “just right.” And that might very well be Latour’s position on the question of “how reflexive is reflexive enough?” to which we anticipate the Janus-faced response to be “it is ‘just right’ when others are convinced that it is.” While provocative, and potentially fair, neither Latour’s answer nor ours is good enough. An additional layer of this explanation remains, which is also nestled in the Goldilocks story, provided we recognize that the original nursery is not at all like the contemporary children’s tale; the original nursery describes an impudent intruder incapable of controlling her impulses when she encounters the possessions of others (Tatar 2002). The original telling was, thus, also a tale about discovery, but a cautionary one; children were to be imparted with the wisdom that when exploring something new, we expose ourselves to certain hazards, and, occasionally, we must 16 In 1837, Robert Southey, an English poet, published the original story, “The Story of the Three Bears,” which contained three bears, which were friendly, who lived in the woods. One day, an old, nasty woman happens upon their home while the bears are away. They were away while their hot porridge cooled, at which point Goldilocks breaks into their home and proceeds to eat the smallest bear’s porridge, and then after breaking the bear’s chair, settles into his bed for a nap. When the bears return from their walk, they discover Goldilocks who, upon being found, dashes out the window. In 1849, the story is transformed into the story that is commonly associated with the tale.  Qual Sociol steal away into the woods if we are detected. This is why we warn readers to “beware of allies,” especially allies like reflexivity, because it encourages you to enter into your own scientific accounts (and taste the porridge). If you do this, and then you linger too long, you’ll be detected and may be forced to flee from your manuscript pages. We did … they are bears, after all. No Bears, but a Tailor And that is where a meta-reflexive account would end, and should end: dead with Goldilocks out in the woods. In all fairness, Latour (1988, 169) did squarely warn readers that meta-reflexivity was potentially “suicidal.” In contrast, our current adventures in infra- reflexivity result in different stories with better, more ambiguous endings. Our experience is more informed by a German fairy tale: “The Valiant Tailor” from Brothers Grimm by Joseph and Wilhelm Grimm. As the story goes, a mild-mannered tailor spreads some newly begotten jam on a thick slice of bread, but, being a diligent worker, completes the coat he is constructing before breaking from work for sustenance. Upon completing his labors, he observed that a swarm of uninvited flies have settled on his snack, and, with a mighty blow, killed no fewer than seven. Impressed by his bravery, he fashioned a belt (or “banner”), which announced seven with one blow. Leaving the confines of his shop, the tailor went out into the world to make a name for himself. He met a giant who, upon seeing the belt, assumed seven with one blow referred to men. After outsmarting the giant and his gigantic peers—in their very own cave, no less—the tailor set out to a local kingdom where his identity would be mistaken once again. After some successful misadventures, he was now supposedly responsible for slaying two giants, nabbing a unicorn, and capturing a wild boar, all single-handedly. Upon discovering that this hero—and, now, son-in-law—was but a lowly tailor, the king gave orders to have him captured in the night. But, as it happens, a friendly squire shared the king’s plot with the tailor. Then, when the soldiers stood outside his bedroom door to seize the tailor, The little tailor, who was only pretending to be asleep, began crying out with a clear voice, “Boy, make the jacket for me, and patch the trousers, or I will hit you across your ears with a yardstick! I have struck down seven with one blow, killed two giants, led away a unicorn, and captured a wild boar, and I am supposed to be afraid of those who are standing just outside the bedroom!”17 Shortly thereafter, the tailor becomes king. In the end, like any experience, our experience is about as self-exemplifying as the tailor’s exploits. By what measures have we succeeded or failed? We do not know; we cannot know; it is out of our hands, and has been for some time. Surely, we have learned, in true meta- reflexive form, to “celebrate the monster of reflexivity,” as Ashmore (1989, 88) suggested. On balance, however, like Goldilocks, we also learned to beware of bears, that reflexivity is every bit as much a monster as an ally, and that fleeing the scene, when you are caught lingering in your own accounts, seems like a viable alternative to celebrating the hungry bears. The practical solution to the circumstances we encountered is just that, a practical solution, and it provides no more insight about getting it right beyond the immediate exigencies of our predicament. If reviewers say it is “right,” then it is, and if our peers would have been different, then the paper would have gotten it right differently (there, or in 17 This passage is borrowed from an on-line source graciously hosted by the University of Pittsburgh. Available at: http://www.pitt.edu/~dash/grimm020.html Accessed 05 May 2013. Qual Sociol some other journal). After all, Latour (1988, 171) writes, “my own text is in your hands and lives or dies through what you will do with it.” A truly relativist and pragmatist approach to reflexivity in academic account-making would conclude that reflexivity is a moving target and whatever will satisfy reviewers is the reflexivity that will be offered by and be available in published work. This paper is self-exemplifying in that respect. Ultimately, the account we provide here is no more or less reflexive than the fairy tales we conclude with. Thus, in true infra-reflexive form, we have learned that no number of additions, subtractions, self-referential looping, or any other reflexive gymnastics will make a paper more, less, definitively, or woefully reflexive. However, we did detect a wrinkle in Latour’s insistence that accounts be self-exemplary. What if our accounts are like the banner around the tailor’s belt-line? Have we killed giants or flies? Like all who meet the tailor, you will decide for us.18 References Adkins, Lisa. 2002. Reflexivity and the politics of qualitative research. In Qualitative research in action, ed. Tim May, 332–348. London: Sage Publications. Adler, Patricia A., and Peter Adler. 1996. Parent-as-researcher: The politics of researching in the personal life. Qualitative Sociology 19: 35–58. Alvesson, Mats, and Kaj Sköldberg. 2000. Reflexive methodology: New vistas for qualitative research. London: Sage Publications. Ashmore, Malcolm. 1989. The reflexive thesis. Chicago: University of Chicago Press. Ashmore, Malcolm, Michael Mulkay, and Trevor Pinch. 1989. Health and efficiency: A sociology of health economics. Milton Keynes: Open University Press. Becker, Howard S. 1986. Writing for social scientists: How to start and finish your thesis, book, or article (Chicago Guides to Writing, Editing, and Publishing). Chicago: University of Chicago Press. Bloor, David. 1976. Knowledge and social imagery. London: Routledge & Kegan Paul. Booth, Charles. 2000. The problems and possibilities of reflexivity in strategy. Electronic Journal of Radical Organization Theory 4: 1–16. Bolker, Joan. 1998. Writing your dissertation in fifteen minutes a day: A guide to starting, revising, and finishing your doctoral thesis. New York: Owl Books. Brante, Thomas, and Margareta Hallberg. 1991. Brain or heart? The controversy over the concept of death. Social Studies of Science 21: 389–413. Callon, Michel. 1986. Some elements of a sociology of translation: Domestication of the scallops and the fishermen of St Brieuc Bay. In Power, action and belief: A new sociology of knowledge, ed. John Law, 196–233. London: Routledge & Kegan Paul. Chia, Robert. 1996a. The problem of reflexivity in organizational research: Towards a postmodern science of organization. Organization 3: 31–59. Chia, Robert. 1996b. Organizational analysis as deconstructive practice. Berlin: Walter de Gruyter. Clifford, James, and George Marcus. 1986. Writing culture. Berkeley: University of California Press. Czyzewski, Marek. 1994. Reflexivity of actors versus reflexivity of accounts. Theory, Culture & Society 11: 161–168. eHealth Initiative. 2007. eHealth Initiative Blueprint: Building consensus for common action. Accessed at: http:// www.ehealthinitiative.org/sites/default/files/file/eHiBlueprint-BuildingConsensusForCommonAction.pdf No longer available. Previously accessed 8 August 2009. Original available from authors. eHealth Initiative. 2009. About the eHealth Initiative. Accessed at: http://www.ehealthinitiative.org/about.html No longer available. Previously accessed 8 August 2009. Original available from authors. Currently available at: http://www.ehealthinitiative.org/about-us.html Accessed 10 November 2012. Emerson, Robert M. 2011. Writing ethnographic fieldnotes (Chicago Guides to Writing, Editing, and Publishing). Chicago: University of Chicago Press. Findlay, Linda. 2002. Negotiating the swamp: The opportunity and challenge of reflexivity in research practice. Qualitative Research 2: 209–230. 18 This reminds us of Dave Wallace’s (1996, 389) excellent advice: “The truth will set you free. But not until it is finished with you.”  Qual Sociol Findlay, Linda, and Brendan Gough. 2003. Reflexivity: A practical guide for researchers in health and social sciences. Oxford: Blackwell. Foucault, Michel. 1977. Discipline and punish: The birth of the prison. London: Allen Lane. Foucault, Michel. 1991. Governmentality. In The Foucault effect: Studies in governmentality, ed. Graham Burchell, Colin Gordon, and Peter Miller, 87–104. Hemel Hempstead: Harvester Wheatsheaf. Gad, Cristopher, and Casper Bruun Jensen. 2010. On the consequences of Post-ANT. Science, Technology, & Human Values 35: 55–80. Garfinkel, Harold. 1967. Studies in ethnomethodology. Englewood Cliffs: Prentice-Hall. Garfinkel, Harold, and Harvey Sacks. 1970. On formal structures of practical actions. In Theoretical sociology: Perspectives and developments, ed. J.C. McKinney and E.A. Tiryakian, 337–66. New York: Appleton. Gieryn, Thomas F. 1983. Boundary-work and the demarcation of science from non-science: Strains and interests in professional ideologies of scientists. American Sociological Review 48: 781–795. Guice, Jon. 1998. Controversy and the state: Lord ARPA and intelligent computing. Social Studies of Science 39: 103–38. Hardy, C., N. Phillips, and S. Clegg. 2001. Reflexivity in organization and management theory: A study of the production of the research ‘subject’. Human Relations 54: 531–560. Hine, Christine. 2007. Multi-Sited ethnography as a middle range methodology for contemporary STS. Science, Technology & Human Values 32: 652–671. Horton, Richard. 2000. Genetically modified food: Consternation, confusion, and crack-up. Medical Journal of Australia 172: 148–149. Klintman, Mikael. 2002. The Genetically Modified (GM) food labeling controversy: Ideological and Episte- mic Crossovers. Social Studies of Science 32: 71–91. Krathwohl, David R., and Nick L. Smith. 2005. How to prepare a dissertation proposal: Suggestions for students in education and the social and behavioral sciences. Syracuse: Syracuse University Press. Latour, Bruno. 1987. Science in action: How to follow scientists and engineers through society. Cambridge, MA: Harvard University Press. Latour, Bruno. 1988. The politics of explanation: an alternative. In Knowledge and reflexivity, new frontiers in the sociology of knowledge, ed. Steve Woolgar, 155–176. London: Sage Publications. Latour, Bruno. 1991. Technology is society made durable. In A sociology of monsters, ed. John Law, 103–31. London: Routledge. Latour, Bruno. 2005. Reassembling the social. New York: Oxford University Press. Latour, Bruno, and Steve Woolgar. 1979. Laboratory life: The social construction of scientific facts. Beverly Hills: Sage. Latour, Bruno, and Steve Woolgar. 1986. Laboratory life: The construction of scientific facts. Princeton: Princeton University Press. Law, John, and John Hassard (eds.). 1999. Actor network theory and after (Sociological Review Monographs). Oxford: Blackwell Publishers/The Sociological Review. Law, John. 2004. After method. London: Routledge. Lee, Kirby, and Lisa Bero. 2006. What authors, editors and reviewers should do to improve peer review. Nature. Available at: http://www.nature.com/nature/peerreview/debate/nature05007.html Accessed 9 Aug 2013. doi:10.1038/nature05007. Lehoux, Pascale. 2006. The problem of health technology: Policy implications for modern health care systems. New York, NY: Routledge. López, Daniel, and Miquel Domènech. 2009. Embodying autonomy in a home telecare service. The Socio- logical Review 56(2): 181–195. Luttrell, Wendy (ed.). 2010. Qualitative educational research: Readings in reflexive methodology and transformative practice. New York: Routledge. Lynch, Michael. 2000. Against reflexivity as an academic virtue and source of privileged knowledge. Theory, Culture, & Society 17: 26–54. Marcus, George E. 1995. Ethnography in/of the world system: The emergence of multi-sited ethnography. Annual Review of Anthropology 24: 95–117. Marcus, George E. 1998. Ethnography through thick and thin. Princeton: Princeton University Press. May, Tim. 1988. Reflexivity in an age of reconstructive social science. International Journal of Social Research Methodology 1: 7–24. Merton, Robert K. 1968. The matthew effect in science. Science 159: 56–63. Mol, Annemarie. 2003. The body multiple. Raleigh: Duke University Press. Mort, Maggie, Tracy Finch, and Carl May. 2009. Making and unmaking telepatients: Identity and governance in new health technologies. Science Technology Human Values 34: 9–33. Mort, Maggie, Carl May, and Tracy Williams. 2003. Remote doctors and absent patients: Acting at a distance in telemedicine? Science Technology Human Values 28: 274–295. Mulkay, Michael. 1985. The word and the world: Explorations in the form of sociological analysis. London: Allen & Unwin. Qual Sociol Naples, Nancy A. 1996. A feminist revisiting of the insider/outsider debate: The “outsider phenomenon” in rural Iowa. Qualitative Sociology 19: 83–106. Passoth, Jan-Hendrik, and Nicholas J. Rowland. 2009. eHealth: Rethinking state-of-the-art technology for state theory. Washington: Presentation at the Society for the Social Studies of Science Annual Meeting. Passoth, Jan-Hendrik, and Nicholas J. Rowland. 2010. Actor-network state: Integrating actor-network theory and state theory. International Journal of Sociology 25: 818–841. Passoth, Jan-Hendrik, and Nicholas J. Rowland. 2011. Seeing the state in self-care. Las Vegas: Presentation at the American Sociological Society Annual Meeting. Passoth, Jan-Hendrik, and Nicholas J. Rowland. 2013. The state multiple. San Diego: Presentation at the Society for the Social Studies of Science Annual Meeting. Pels, Dick. 2000. Reflexivity: One step up. Theory, Culture & Society 17: 1–25. Pels, Dick. 2003. Unhastening science: Autonomy and reflexivity in the social theory of knowledge. Liverpool: Liverpool University Press. Rappert, Brian. 2001. The distribution and resolution of the ambiguities of technology, or why Bobby can’t spray. Social Studies of Science 31: 557–91. Reno, Joshua. 2011. Managing the experience of evidence: England’s experimental waste technologies and their immodest witnesses. Science, Technology, & Human Values 36: 842–863. Richards, Evelleen. 1991. Vitamin C and cancer: Medicine or Politics? London: Macmillan. Rose, Nicholas. 1999a. Powers of freedom: Reframing political thought. Cambridge: Cambridge University Press. Rose, Nicholas. 1999b. Governing the soul: The shaping of the private self. London: Free Association Books. Silvia, Paul J. 2002. How to write a lot: A practical guide to productive academic writing. Washington: American Psychological Association. Single, Peg B., and Richard M. Reis. 2009. Demystifying dissertation writing: A streamlined process from choice of topic to final text. Sterling: Stylus Publishing. Star, Susan Leigh, and James R. Griesemer. 1989. Institutional ecology, ‘translations’ and boundary objects: Amateurs and professionals in Berkeley’s Museum of Vertebrate Zoology, 1907–39. Social Studies of Science 19: 387–420. Steier, Frederick. 1991. Research and reflexivity. London: Sage Publications. Strauss, Anselm. 1987. Qualitative analysis for social scientists. Cambridge: Cambridge University Press. Tatar, Maria. 2002. The annotated classic fairy tales. New York: W.W. Norton & Company. Wallace, David Foster. 1996. Infinite jest. New York: Little, Brown and Company. Webster, Andrew. 2002. Innovative health technologies and the social: Redefining health, medicine and the body. Current Sociology 50: 443–457. Weiss, Robert S. 1994. Learning from strangers: The art and method of qualitative interview studies. New York: Free Press. Westrum, Ron R. 1982. Book Reviews: “The human nature of science: Researchers at work in psychiatry” Stewart E. Perry (New York: Macmillan, 1966); “Laboratory life: The social construction of scientific facts” Bruno Latour and Steve Woolgar (Beverly Hills, CA: Sage Publications, 1979), Science Commu- nication 3: 437–440. Wichroski, Mary Anne. 1996. Breaking silence: Some fieldwork strategies in cloistered and non-cloistered communities. Qualitative Sociology 19: 153–170. Winthereik, Brit Ross, Irma van der Ploeg, and Marcus Berg. 2007. The electronic patient record as a meaningful audit tool: Accountability and autonomy in general practitioner work. Science Technology Human Values 32: 6–25. Woolgar, Steve (ed.). 1988a. Knowledge and reflexivity. London: Sage Publications. Woolgar, Steve. 1988b. Reflexivity is the ethnographer of the text. In Knowledge and reflexivity, ed. Steve Woolgar, 14–36. London: Sage Publications. Zinsser, William. 2006. On writing well, 30th anniversary edition: The classic guide to writing nonfiction. New York: Harper Perennial. Jan-Hendrik Passoth is a PostDoc at Technische Universität Berlin. He connects sociological theory and science and technology studies by working on problems of social structure and infrastructures, human and nonhuman agency, and discourse and material culture. He writes for the blog “Installing (Social) Order.” Nicholas J. Rowland is an Associate Professor of Sociology and Science and Technology Studies at Pennsylvania State University. He also connects sociological theory and science and technology studies by working on problems of social structure and infrastructures, human and non-human agency, and discourse and material culture. He writes for the blog “Installing (Social) Order” too.