By Vita Peacock
Directions May 2024
Introduction
The practice of giving our interlocutors ersatz names, to protect their identities for various reasons, has a substantial history in anthropology. Pseudonymization is equally a technique formalized, and sometimes demanded, by ethics assessors both in and outside the profession. Erica Weiss and Carole McGranahan recently reopened the question of pseudonyms (2021), arguing that they instantiate widely cherished anthropological values in a way that has become self-evident. Is anonymizing through pseudonyms always the best response to complex encounters in the field? Some of the interventions they assemble explore contexts where it is, in fact, naming that may present the most ethical choice (cf. Cubero, 2021; Turin, 2021). But going beyond pseudonyms asks something further. It invites a more fundamental appraisal of anthropological representation in general. What if the options available to us as ethnographers did not need to shuttle between naming and not naming, and the logics of revelation and concealment they stand in for. What other strategies may there be?
This question imposed itself on me—first as an inkling and latterly with greater force—when I began working with privacy and data protection advocates in Germany in 2019. [1] With a direct history that stretches to the 1970s, but that has acquired transformed and renewed vigor under the impact of digitalization, Germany has long been home to civic opposition to forms of information collection. Among both influential and lesser-known participants of Netzpolitik, as this community is currently known, one of the main concerns has been around the construction of “personality profiles”—the automated analysis of personal information to make decisions and predict behavior, particularly when combined with biometric data. [2] Simultaneously, as the recipient of a grant from the European Research Council (ERC), in which I am legally bound to uphold European General Data Protection Regulation (GDPR), like other anthropologists I have been compelled to think carefully about it (Cant, 2020; Humphris, 2018; Salinas, 2023; Yuill, 2018). What makes this question unusually acute in this context is the cultural overlap between these two phenomena. As a community of advocates for privacy and data protection, Netzpolitik played a not insignificant role in the political momentum that yielded GDPR, which is in part a Europeanization of German data protection norms.
A response arrives by shifting the axes of representation. Here, I pursue some of the claims made by Mirco Göpfert (2020), that find earlier precedents in Michael Taussig (1999) among others. Göpfert argues that there has been a historical assumption in anthropology that locates the core of culture “backstage” (2020, 493), which it becomes the ethnographer’s duty, if not to reveal, then at the very least to know. Indeed, this assumption has colored much of my previous fieldwork. An alternative to this traditional opposition between revelation and concealment—that pivots around a monist notion of empirical truth—is to resituate representation as between opacity and legibility—that pivots instead around its own point of intervention. Ethnographic writing then becomes a matter of leaving some parts of the field dark while rendering others readable (cf. Buitron & Steinmüller, 2023). Drawing on three ethnographies that illustrate in different ways how such an approach might be followed through (Puccio-Den, 2022; Simpson, 2014; Yonucu, 2022), here I sketch the outlines of a way of writing opacity that I tentatively name spirit portraiture.
Spirit portraiture is a mode of textual characterization that omits any information through which that person could be later reidentified, either with or without the use of computing. As a rhetorical technique it responds to anthropology’s demand for legibility, characterizing interlocutors in a textured way through gestures, comportment, and speech not digitally recorded. It equally responds to privacy advocates’ demands for opacity, by detaching any information that would be defined as personal under GDPR. We could conceive of the spirit portrait as the antithesis of the biometric profile. Where the latter combines unique biological characteristics with digitally documented individual behaviors, the spirit portrait represents the person through embodiment and social and environmental interchange, only including material that appeared first in analogue form. By sundering the link between biological characteristics and digital data—an intersection around which rising regimes of biometric surveillance are based—it also invites us to consider the ways in which cultural subjects diverge from these increasingly prevalent socio-technical assumptions.
Spirit Portraits
Collective opposition to profiling is a key concern and point of action in Netzpolitik. Profiling is usually taken to mean efforts by private companies to reify internet users into a coherent collection of behaviors, which can then be used or sold, particularly for advertising purposes. As one privacy advocate puts it: “All your patterns go into data factories, where they are sold and resold to create a profile. They can show—what are your fears and your joys? What is your body weight? What is your intelligence? What is your income? Everything is collected in this black box to give you an advertising identifier, for example.” [3] Profiling can also involve folding in the unique biological characteristics of natural persons, such as the pattern of their fingerprint, face, or iris, or a variety of movements, such as gait or typing style, by state or nonstate actors. Their opposition to the latter has so far materialized in campaigns against the compulsory inclusion of fingerprints on German identity cards (the #PersoOhneFinger, or person without a finger campaign), as well as for an EU-wide ban on facial recognition (the #ReclaimYourFace campaign) among others.
One of the intentions of GDPR (which, while not without qualification, Netzpolitik views in positive terms) was to contain the increase in profiling that has been enabled by proliferating computing. The legislation is thus oriented around a concept of “personal data,” which it defines in Article 4(1) as, “any information relating to an identified or identifiable natural person.” Like other ERC grantees, my research in Germany is legally bound to GDPR through the signature of a tripartite contract that is appended to a Data Management Plan (DMP). The DMP describes the various techniques adopted by myself and the other members of the project to uphold a number of data protection imperatives. One of these is the principle of “data minimization,” which undertakes to limit the amount of personal data that is collected during the research. [4] This essay should not be read as a defense of this principle as bluntly applied to anthropology, which would make most fieldwork as historically constituted impossible, as several social science associations have stipulated (ASA, BSA, ESA, 2018). Nonetheless, its emic connection to this particular field site invites me to incorporate it as a culturally specific representational technique. Here, I take data minimization at face value as an experiment in writing opacity.
Spirit portraiture applies this definition as proscription to ethnographic writing. It characterizes a natural person, while excising any information that may de-anonymize them. This includes not only physical characteristics, such as hair color, height, or a noteworthy tattoo, but also forms of data that could be subject to computational search or computational analysis. This entails direct quotation of digital data of any kind, as well as the inclusion of any multimodal digital material. As my interlocutors occasionally told me, due to advances in voice recognition, even an audio file could now be biometrically analyzed to reidentify the speaker, rendering the very concept of an anonymous interview—if digitally recorded—an oxymoron. It is worth making clear at this point that spirit portraiture is distinct from another opacity technique used by anthropologists, which draws on Max Weber’s notion of an ideal type (Weber, 1949). While the latter is a fictional character constructed out of culturally “typical” characteristics, a spirit portrait is a portrait of a natural person, though without any information that could reidentify them as such. The question then becomes—what remains when these details are stripped away?
Consider the following:
They approach in a particular manner. Shoulders back, chest perceptibly thrust forward in an embodiment of self-command. There seems to be a sense of anticipation to the interview, like a challenge they are deciding to meet, tentatively, but with willing. As we walk I discuss the interview’s mechanics, and they say they do not want to be digitally recorded. Without pressing the matter I explain that this will mean I will be writing intensively throughout, and may not be able to catch every word. Almost instantly they switch positions, saying that minimizing data about themselves can be a kind of “vanity,” and that they trust me.
We sit down in a vegan restaurant at their behest, on a mezzanine overlooking an internal courtyard. With the recorder now on, I spend much of the interview wondering whether I should have accepted their initial request without remark, as they enter into a noticeably performative relationship to the device lying between us, as if there were third party there to whom they were really speaking, rather than me. When there is a clamor in the courtyard below, as workmen unload trolleys of drinks, they pause in anticipation that this may prejudice its later transcription. So absorbed by their utterances I have barely noticed.
Their self-command comes through in speech as well as gesture. They weigh their words with tangible care, expending a visible effort to respond to my questions as thoughtfully as they can, even looping back to earlier topics we had discussed to create their own narrative arc across the interview. At the very end however there is a moment of vulnerability—the muscles of their face slacken and a look of fear travels through their eyes—as they describe politically formative childhood experiences. Once the recorder is off they visibly relax, and express the hope that I have found something “useful.” After what has been an intense, even latterly emotional encounter, we shake hands slightly awkwardly, and they pass me leaflets and postcards of events and associations they are involved in. As elsewhere, while digital mediation invites thirdness into the encounter, paper transports a more direct kind of intimacy.
The description omits any information that could be used, with or without computation, to reidentify the subject. Granular details are substituted for encompassing terms like “vegan restaurant” or “childhood experiences.” In the former, although I loosely describe the setting of the restaurant, that this could be anywhere in four cities across Germany makes it almost impossible to identify, even using Open-Source Intelligence (Ghioni et al., 2023). Where direct quotations are included, these are from exchanges that take place either before or after the recording. No physical characteristics are described, which in this case includes sex and gender. I will return to this below.
What remains is the sense of a conversation, enveloped in the manifold ways in which the spirit animates the body, including in relation to the ethnographer. Gestures and comportment (“shoulders back”) carry attitudes and ideas within them, while the appearance of emotion (“the muscles of their face. . . slacken”) suggests the eruption of memory that is subjectively constituting (Beatty, 2014). Because I, as interviewer, am already de-anonymized, I can include whatever relevant details about myself. In this case this includes the making of a methodological decision, and subsequent doubts about it, as well as the traffic of gestures (the shaking of hands, the receipt of leaflets), that indicate the peculiar combination of proximity and distance that the interview achieves. What does this description tell us? Alone perhaps not very much. But in the context of this particular research, it illustrates something of analytical import—namely, attitudes and behaviors around recording. The signpost “as elsewhere” signals continuities with other ethnographic phenomena that can then be expanded. How else might spirit portraiture be written?
Everything is said with emphasis, moving between tonal and gestural extremes. They have a way of talking that can move swiftly from slow and quiet, to fast and loud to make a conceptual point. Gestures often accompany and visualize these thoughts. In an instant they unfold and retract their middle finger to convey their feelings about power, at another they raise their arm at right angles to spatialize how far their capacity for trust extends. Fortunately, I am just about within reach. Indeed their approach to me is imbued with an understated but marked generosity and care. They invite me to their home, and open the door wearing a medical mask and reassuring me they have tested negative for COVID (even though the latter is no longer required). As I go in it transpires that they have prepared lunch for us to eat together, that waits patiently on the stove; and afterward when pouring me tea they do so in a mug brought back from Scotland, as I had mentioned it. This combination of harshness and softness, of hard lines with care, materializes in their attitude to those who take the fatalistic and disempowered attitude to campaigns for privacy. These character types need a jolt, they say, “Well, nothing will change if you don’t do anything!”
They are effortlessly egalitarian. Although they move between emotional poles, there is an ideological thrum beneath it that remains constant, that they are willing to theorize only as being people-oriented. When we are working on a campaign together, and I express concern at my own inexperience, they calmly reply that what they contribute, and what I contribute, will be different, neither better, nor worse. The egalitarianism has its own kind of hardness—a binary of in and out rather than the acceptance of gradation within. When I suggest taking a more muted role in proceedings, they respond that it is entirely out of the question; stepping down in any way must entail stepping out. For them these kinds of granular details about process are constitutively important. At one point they are concerned that a turn of phrase we are using in the campaign would not be credible (konsensfähig) among those whom the campaign is seeking to mobilze.
For them the technical world is alive. They own an old-style laptop possessing profoundly human characteristics: with not only a name but a biography of its own, telling me with something approaching parental pride how many Congresses it has been to. [5] Like most other privacy advocates they also own a smartphone, but it is the kind that can be de- and reassembled at will. When they want to be completely alone they modify their physical environment by shutting the door, and taking the battery out of the phone completely. Sometimes when they travel they swap phone networks or financial details with other advocates—simply for the solitary reward of confusing the surveilling forces moving through the technical world that they perceive.
Here the portrait once again leans heavily on the body. Tone, tempo, gesture, comportment—all convey something about how the interlocutor is positioned and positions themselves in the world. It also weaves in material effects: a mug from Scotland, food waiting on the stove, the ideological remnants of being people-oriented without naming political persuasions that would count as personal. Again, there are several arrows pointing outward that make this description connectable with other material, through which it is able to make a broader point. In this case, it signals the relation between egalitarianism and exclusion, and the intense involvement and inalienable relationship to computational technologies. Other impersonal details such as mask-wearing and COVID-testing instantiate wider cherished, and to some degree definitional, attitudes to public health in Netzpolitik, that diverge from privacy advocates in other parts of the world. It remains now to discuss the wider scholarly context within which such a way of writing could be applied.
Opacity and Legibility
Obviating the binary choice between naming and not naming, is to pry open the horizon of representation in anthropology. Here, I would summarily draw attention to three ethnographies that already go well beyond pseudonyms in making sociocultural formations readable, while also staking limits on readability in meaningful ways. Deborah Puccio-Den’s study of the Sicilian Mafia (2022), Audra Simpson’s study of Mohawk life across the border of the United States and Canada (2014), and Deniz Yonucu’s study of the policing of Alevi Communities in Turkey (2022)—though differently articulated—share a number of commonalities that reorient representation around an alternative axis of opacity and legibility. This is connected, in each, to an unabashed historicism, as well as a more or less open challenge to epistemology in anthropology, as it has been historically defined.
In each work, opacity—understood here as leaving some information in the dark—is a central feature of the writing that is impelled by the ethnography itself. For Puccio-Den this is the “silence” (omertà) that obtains around the concept of the mafia (2021, 1), removing it from a space of discourse. For Simpson it is her “refusal” to articulate Mohawk life within timeless modalities such as kinship or ceremony, that have characterized 19th- and 20th-century ethnographies (2014, 7). Meanwhile for Yonucu, who draws on Simpson’s concept of refusal, opacity arises by carefully obscuring identities in her photographs, as well as other contextual information about the neighborhood she worked in, that could be used to serve the ends of policing. At this point, it might be observed that in all three works, the opacity impelled by the field is or has been a response to demands for legibility coming from a nation-state—whether Italy, the United States/Canada, or Turkey. The historicism that runs through these works, therefore, provides essential context to how the emergence of nation-states in different parts of the world has served over time to dispossess and/or criminalize different peoples and modes of life, with effects that are ongoing.
What is particularly intriguing—and should offer hope for the perpetual reinvention of ethnography—is the way in which rendering some phenomena opaque opens others up for legibility and understanding. Faced with a field of silence, Puccio-Den articulates this world of the unsaid through the bodies and embodiment of antimafia activists of various hues, as well as through material traces left by the mafia themselves. This leads her to question the reliance on discourse that can sometimes overburden contemporary anthropology. Likewise, Simpson and Yonucu combine leaving some information in the dark with displaying other kinds of information that have historically been more hidden. This is particularly the case in Yonucu’s deliberate combination of ethnography with archival research, to assemble a jigsaw puzzle of counterinsurgency techniques which, while a partner to the state, have tended to position themselves out of view. One conclusion arising from these different works, which could be elaborated at far greater length, is that writing opacity need not be mourned as a loss of intelligibility. It can instead initiate a metachallenge to anthropology and its own techniques of representation, while simultaneously continuing to advance it.
As political form, the state likewise plays a central role in the emergence of Netzpolitik. The first associations in Germany to share civil concerns around privacy and data protection were formed in and around attempts by the former West German Federal Republic to reintroduce a national census in the 1970s and 1980s. [6] While the attempt to reintroduce the census failed at the time, it succeeded in seeding a census boycott movement (Hannah, 2017), the repercussions of which continue to reverberate. Natalia Buitron and Hans Steinmüller argue that state demands for legibility cohere around a politics of centralization (2023). Those who do not wish to be governed in this way may therefore respond with decentralization and opacity, which is arguably the case here. However, major social and technical changes since the 1980s also reconfigure this dynamic. Netzpolitik straddles a significant historical turning point in where demands for legibility are coming from. Once the monopoly of nation-states (Breckenridge, 2014), biometric surveillance in particular is now increasingly dispersed and privatized through the capacities of computing to map biological properties and behaviors, which can be crosslinked to other forms of digital data. Having in previous iterations been oriented around the state, Netzpolitik now fights on several fronts at once in its own struggle for rights to opacity (cf. Birchall, 2021).
In this context, in which it is the response to demands for legibility that constitutes the very form of cultural participation my interlocutors are taking, I offer spirit portraiture as a way of writing that respects these imperatives. But what of anthropology? Personal identifiers such as class, ethnicity, gender, or religion have been important not only for a “militant anthropology” that seeks to make the politics of the field explicit (Juris & Khasnabish, 2013) but for many kinds of anthropology in general. Like those ethnographies above, this way of writing opacity operates in conversation with other forms of legibility. These again take emic form. While proscribing information categorized as personal, collective cultural legibility can then arise through information categorized as public. Indeed, while the personal may be an area of activity hedged in with taboos, public life in Netzpolitik can be a surprisingly intimate and free-flowing space of exchange—problematizing the very categories of private and public itself.
Conclusion
This essay has journeyed beyond pseudonyms, to experiment with a form of ethnographic writing it names spirit portraiture. While obscuring information that falls under the legal category of personal data, it hints toward protagonists’ cultural connections and histories through their bodies, and the traces of its interactions with the environment. The spirit portrait holds on to the value of empiricism, the information the ethnographer is able to gather through their senses. It also holds on to the value of thick description, in this case characterization, as a rhetorical method that offers readers access to our field sites, and vital ground for the anthropological imagination. By detaching the personal from the person, it incorporates ethical imperatives coming from within and beyond the field, as part of a wider anthropological response to new landscapes of hypervisibility, as well as the legal efforts to contain them.
Rather than revelation or concealment, existing ethnographic works demonstrate how representation can be reoriented around another axis of opacity and legibility. They also raise a question for anthropology in the main that is potentially profound. How far has the state’s demand for legibility, which in its period of gestation were de facto colonial states, overdetermined the very forms of legibility that anthropology itself has sought? Fortunately for the discipline as a whole, it contains within its own practice the permanent potential for reflexivity and reinvention. As demands for legibility continue to deepen and to transmogrify, so too can anthropology’s capacity to redraw the lines of sight.
Acknowledgements
I would like to sincerely thank Deniz Yonucu and Caroline Parker for the invitation to contribute to this section, and for their perspicacious and productive comments on an earlier draft. I would also particularly like to thank Kim Hopper for his lyrical and inspiring peer review. This article was produced as part of the ERC SAMCOM project funded by the European Research Council, grant no. 947867.
Notes
[1] From 2019 to 2023, I undertook approximately 14 months of fieldwork in the cities of Munich, Berlin, Hamburg, and Bielefeld, in several long and short stays that excluded the period of COVID-19 lockdowns. This included formal and informal interviewing alongside participation in meetings and events and was triangulated with digital material produced by a number of associations, as well as archival research in local district courts (Amtsgerichte).
[2] European GDPR defines profiling as “any form of automated processing of personal data consisting of the use of personal data to evaluate certain personal aspects relating to a natural person, in particular to analyze or predict aspects concerning that natural person’s performance at work, economic situation, health, personal preferences, interests, reliability, behavior, location or movements.” GDPR article 4(4); https://gdpr-info.eu/art-4-gdpr/.
[3] Anonymous informant in discussion with the author, December 2021.
[4] GDPR article 5(1)(c) of GDPR states: “Personal data shall be adequate, relevant and limited to what is necessary in relation to the purposes for which they are processed (data minimisation).” The question for anthropology is what “adequate, relevant, and limited” means in the context of ethnography, which often defines the frames of relevance after the fact. https://gdpr-info.eu/art-5-gdpr/.
[5] This is a reference to the Chaos Communication Congress, which is the annual meeting of the Chaos Communication Club e.V, that always takes place between Christmas and New Year.
[6] The Gesellschaft für Datenschutz und Datensicherheit e.V. was founded in 1977; the Chaos Computer Club e.V. in 1981; and FoeBud e.V. (now Digital Courage e.V.) in 1987.
Works Cited
ASA, BSA, and ESA (Association of Social Anthropologists of the United Kingdom and Commonwealth, British Sociological Association, and European Sociological Association). 2018. Anthropology, Sociology, Human Geography and the Implementation of New Data Protection Legislation in UK Academic Institutions. Joint statement by the ASA, BSA, and ESA. https://www.theasa.org/downloads/Joint%20ASA%20BSA%20ESA%20RGS%20statement%200418.pdf
Beatty, Andrew. 2014. “Anthropology and Emotion.” Journal of the Royal Anthropological Institute 20(3): 545–563. https://doi.org/10.1111/1467-9655.12114.
Birchall, Clare. 2021. Radical Secrecy: The Ends of Transparency in Datafied America. Electronic Mediations Series. Minneapolis, MN: University of Minnesota Press.
Breckenridge, Keith. 2014. Biometric State: The Global Politics of Identification and Surveillance in South Africa, 1850 to the Present. Cambridge: Cambridge University Press.
Buitron, Natalia, and Hans Steinmüller. 2023. “Governing Opacity: Regimes of Intention Management and Tools of Legibility.” Ethnos 88(4): 677–701. https://doi.org/10.1080/00141844.2021.2007154.
Cant, Alanna. 2020. “Participatory Research in Mesoamerica and Data Protection in Europe (and Elsewhere).” Annals of Anthropological Practice 44(2): 152–156. https://doi.org/10.1111/napa.12144.
Cubero, Carlo. 2021. “ What Does Anonymity Mean in Anthropological Filmmaking?” In Rethinking Pseudonyms in Ethnography, edited by Carole McGranahan and Erica Weiss, American Ethnologist website, December 13, 2021. https://americanethnologist.org/online-content/collections/rethinking-pseudonyms-in-ethnography/what-does-anonymity-mean-in-anthropological-filmmaking/.
Ghioni, Riccardo, Mariarosaria Taddeo, and Luciano Floridi. 2023. “Open Source Intelligence and AI: A Systematic Review of the GELSI Literature.” AI & SOCIETY. https://doi.org/10.1007/s00146-023-01628-x.
Göpfert, Mirco. 2020. “Epistemophilic Obsessions: Espionage, Secrets, and the Ethnographer’s Will to Know.” HAU: Journal of Ethnographic Theory 10(2): 487–498. https://doi.org/10.1086/709483.
Hannah, Matthew G. 2017. Dark Territory in the Information Age: Learning from the West German Census Controversies of the 1980s. London: Routledge.
Humphris, Rachel. 2018. “Is Anthropology Still Legal? Notes on the Impact of GDPR.” Age of Superdiversity: Rethinking a Rapidly Changing Society (blog), May 16, 2018. https://superdiversity.net/2018/05/16/is-anthropology-legal/
Jeffrey S. Juris, and Alex Khasnabish, eds. 2013. Insurgent Encounters: Transnational Activism, Ethnography, and the Political. Durham, NC: Duke University Press.
Puccio-Den, Deborah. 2022. Mafiacraft: An Ethnography of Deadly Silence. Chicago, IL: HAU Books.
Salinas, Cecilia G. 2023. “Anonymizing in Digitalized Fieldwork: An Art-Based Blurring Approach.” Anthropology Today 39(6): 15–17. https://doi.org/10.1111/1467-8322.12848.
Simpson, Audra. 2014. Mohawk Interruptus: Political Life across the Borders of Settler States. Durham, NC: Duke University Press.
Taussig, Michael T. 1999. Defacement: Public Secrecy and the Labor of the Negative. Stanford, CA: Stanford University Press.
Turin, Mark. 2021. “ Recognizing Authority and Respecting Expertise in Language Work.” In Rethinking Pseudonyms in Ethnography, edited by Carole McGranahan and Erica Weiss, American Ethnologist website, December 13, 2021. https://americanethnologist.org/online-content/collections/rethinking-pseudonyms-in-ethnography/recognizing-authority-and-respecting-expertise-in-language-work/.
Weber, Max. 1949. Max Weber on the Methodology of the Social Sciences. Translated by Edward Shils and Henry A. Finch. New York: Free Pres
Weiss, Erica, and Carole McGranahan. 2021. “ Rethinking Pseudonyms in Ethnography: Anan Introduction.” In Rethinking Pseudonyms in Ethnography, edited by Carole McGranahan and Erica Weiss, American Ethnologist website, December 13, 2021. https://americanethnologist.org/online-content/collections/rethinking-pseudonyms-in-ethnography/rethinking-pseudonyms-in-ethnography-an-introduction/
Yonucu, Deniz. 2022. Police, Provocation, Politics: Counterinsurgency in Istanbul. Police/Worlds: Studies in Security, Crime, and Governance. Ithaca, NY: Cornell University Press.
Yuill, Cassandra. 2018. “‘Is Anthropology Legal?’: Anthropology and the EU General Data Protection Regulation.” Anthropology in Action 25(2): 36–41. https://doi.org/10.3167/aia.2018.250205.