Abstract
Social scientists – and especially those who subscribe to an interpretive stance, as most ethnographers do – have found themselves many times having to defend their craft from competing paradigms and (external) critiques (Pereira, 2012). Steven Lubet’s book sets itself more particularly against the acclaimed ethnography conducted by Alice Goffman, ‘On the Run’ published in 2014. But we should not be mistaken: it is a much broader assault on ethnography couched as a friendly review.
Lubet claims to depart from a sympathetic position towards ethnography and the work of urban researchers. Nevertheless, he proceeds to systematically poke holes in urban ethnographies and figures from the overall literature. Such a wolf-dressed-as-sheep stance is visible in the very title of the book, Interrogating Ethnography: Why Evidence Matters. While one would expect from such headline an inquiry on what ethnography entails and the extent evidence is brought to bear into theoretical explanations, the book reads like an absurd manual on ‘What Lawyers Can Teach Ethnographers’.
The author promotes the idea of ‘evidence-based ethnography’ and offers to measure ethnographic accounts against courtroom routines (and sometimes journalistic ones – apparently even journalists do it better than ethnographers). He comes to list ways through which ethnographic accounts could become more evidence-based, with ethnographers compelled to fact check each of their stories and the reports of their informants. To explain how this process can translate into more robust ethnography, Lubet uses the image of an ethnography ‘trial’ (borrowed from Mitchell Duneier) as a mechanism for ethnographers to assess the empirical ‘validity’ of their work – would it stand in front of a trial? This metaphor of ethnographers defending their work in a courtroom relies, however, on strong and erroneous assumptions with regard to the objectives and the output of ethnography. Because of those fallacious assumptions, the ‘ethnographic trial’ ends up being more Kafkaesque than anything else.
Misunderstanding ethnography does not make for a good interrogation
First, Lubet expects ethnography to produce ‘generalizable’ accounts and a ‘reasonably reliable rendering of the social world’ (p. 2). It is clear from the onset that the author misunderstands the very nature of interpretivism as evidenced in his call for ‘testable reproducibility’, ‘independently evaluated’, ‘replication’, ‘reliability’ and ‘falsification’ of ethnographic accounts. Ethnography is an interpretive work, and its goal is to offer a theoretical narrative of social life (Schwartz-Shea and Yanow, 2013).
While urban ethnography may arguably follow a more ‘reportage’ style, either due to the nature of the themes (e.g. violence, poverty) or the traditions within such a sub-field, ethnography is an approach to social science research (Nader, 2011). That is, a way of making sense and reporting social phenomena in an empirical-lead, bottom-up manner that uses naturally occurring phenomena as the building blocks of scholarly understanding. Such attention for the ‘ordinary’ in some cases take an ethical stance – as evidence in the work of Alice Goffman – as a way to give voice to emarginated groups.
Ethnographers have no ambition to formulate iron-clad causal claims; rather they aim to chronicle social life, recognizing their partisanship, positioning and plurality. Also, they are not afraid of ‘be[ing] influenced by the identification with [their] subjects’ (one of the many critiques Lubet makes about Goffman, p. xii) they are interested in presenting the raw facts as they see them, while acknowledging their lens and biases (Anteby, 2013).
We may agree that law and courtroom procedures offer the potential to enhance the way ethnographers tell their stories and juridical metaphors might take ethnographers to use a broader range of sources beyond their informants and observation. But by contrast with police or legal reports, ethnography does not aim to condemn or vindicate individuals, thus it should not be held ‘accountable’ by such standards. Thus, Lubet fundamentally misunderstands what ethnography relies on: the idea of empirical relativism. Claims and empirical accounts are always relative and subjective to the person making them, and that is why ethnography does not present itself as a ‘replicable’ empirical approach.
Ignoring the methodological debates around ethnography
Second, Lubet ignores the methodological and theoretical discussions in the social sciences literature and reviews/reads texts without understanding their scholarly context. He discusses the theme of trusting informants; how to differentiate between one own’s analytical perception and the account of informants; or the extent ‘facts’ matter for the overall ethnographic enterprise. These, however, are not newly discovered themes as they have been a source of concern for ethnographers from various disciplines for many decades (Bourdieu, 2003; Clifford and Marcus, 1986; Irwin, 2006).
All the discussions on reflexivity and the limits of textuality, authorship and representation have been reminders for ethnographers to the importance of differentiating what Lubet calls first-hand testimony and hearsay and producing accounts that are attentive to specify on what basis ethnographers advance their claims and how it relates to emic ideas (Brewis, 2014; Campos, 2012; Pereira, 2013). Similarly, the issue of rumors and folklore and, more broadly, the extent the accounts of informants are factual is a part of the past and present history of anthropology, which is well exemplified in the classic polemic on Margaret Mead’s fieldwork in Samoa (Freeman, 2014; Orans, 1996; Shankman, 2009).
While in most parts, Lubet’s apparent ignorance towards the history of ethnography only leads to re-purpose as new what is already known (as if he was selling ice to polar bears), in at least a couple of moments it becomes epistemologically dangerous – something which is ironic in light of his own admonishments against ethnographers committing crimes and his critiques of Alice Goffman. More specifically, Lubet advances a disconcerting argument towards dropping anonymity in ethnography. While this has been the center of current discussions (Jerolmack and Murphy, 2019), what is concerning here (and which showcases the author’s sheer ignorance for ethnography’s history and methods) is the reasons provided by Lubet. He insists that anonymity is not justifiable due to scientific reasons (of a positivistic science). Yet, in the process, he overlooks the extent anonymity is first and foremost neither a device to secure access nor to ‘generalize’ stories or not even a simple ticking-box exercise for institutional review boards. It is based on an ethical stance and a concern for informants. The question is not whether informants might be willing to be named, but that ethnographers and the epistemic community to which they belong have a duty towards preserving them. And after very brutal origins, enmeshed with colonialism, it is probably best to err on the side of caution and maintain anonymity to prevent harm.
By the same token, Lubet regularly insists on the importance of not trusting informants’ account at face value and interviewing other people, consulting documents and so on. Much discussion has filled pages of ethnography journals on the possible types of violence – symbolic or otherwise – and exploitation ethnographers might subject informants. He condemns some anecdotes reported by Alice Goffman on the basis that some of her informants have transformed reality. But in no way does Goffman present those accounts as the unique truth; she always acknowledges the prism through which events are described to her via her informants. The risk of questioning all informants may lead researchers to lose sight of respect for informants (as happened in Laud Humphreys’ work, exposing the sexuality of his informants).
The trial on trial
In the trial of ethnography put forward by Lubet, he seems to take the role of prosecution, defence and judge. Yet if ethnography is a scholarly enterprise, the ultimate arbiter is scholarship in itself – or perhaps the informants themselves. By relying on experts and official documents to challenge existing ethnographic accounts, Lubet does not get us any closer to a final ‘evidence’ but provides only an alternative position from which to argue. In fact, many of the sources Lubet points out (from documentation to alternative experts), and then uses to invalidate the work of Goffman, are no more legitimate and only deliver another perspective that is not necessarily closer to the ‘truth’.
What is more, legal procedures, and the trial format in particular, is not without its problems. There are conflicting perspectives and truth claims within law itself (e.g. which precedent to follow) and these are resolved through particular interpretations (Goodwin, 1994). Moreover, trials are not impartial arbitrators – just consider the experience of minorities facing the legal system (Bastian, 1993).
In sum, Steven Lubet seems to ask ethnography to drop its objectives, ideals and standards in an irrelevant quest for ‘generalizability’; to address issues which have already been discussed and debated; and to make him the judge of it all. This makes Lubet’s book more a work of fiction best read in the Kafka tradition.
