The ethics of collecting data in public space

Editor’s note: We find ourselves in a time of deep global crisis when reflections on research ethics take on new urgency. Language on the Move is delighted to bring to you a series of texts that aim to rethink research ethics in Applied Linguistics. The texts in this series have been authored by members of the Research Collegium of Language in Changing Society (RECLAS) at the University of Jyväskylä in Finland. Their frustrations with a narrow legalistic understanding of ethics brought them together in a series of meetings and long debates in unconventional contexts, where they explored an understanding of ethics as foundational to and intertwined with all aspects of doing research. The result of these meetings and conversations is a series of “rants”, which they share here. In this rant, Sigurd D’hondt examines the legal and ethical implications of the EU’s General Data Protection Regulations for ethnographic research. In doing so, he offers a quick introduction to Goffman, Habermas, and the public sphere.

To view the other RECLAS ethics rants, click here.

***

Let me start with a brief warning: most of this rant is not about ethics but about legal compliance, and deals with the impact of the EU’s newly adopted General Data Protection Regulation (GDPR) (EU) 2016/679. This may seem a bit odd at first, because law and ethics represent two fundamentally different orders of normativity, one a formal system of bureaucratically enforced rules, the other grounded in the self-regulation of conduct. Yet, engaging with this new bureaucratic-legal framework does require social scientists to take an ethical stance of some sort. This is particularly the case for those of us who do research in/on the public realm.

GDPR became operative on May 25, 2018. It sets a new standard for the technical and organizational measures that public authorities, business, and non-profit organizations must take to prevent data security breaches and to give individuals control over their own personal data. It states, for example, that personal data cannot be exported outside the EU, and that processing is only allowed if one can prove that the data were collected in accordance with one of the lawful grounds recognized by the regulation. When you are processing someone’s personal data, you should also let them know that you are doing so, by providing them a privacy notice.

GDPR caused quite a stir among anthropologists and social scientists, with certain voices bordering on the apocalyptic. A workshop at SOAS, for instance, had the title “Is anthropology still legal?” (Humphris 2018). Others were more optimistic and argued that the new regulation might help social science researchers to cast off the straitjacket of biomedicine (Sleeboom-Faulkner and McMurray 2018), or pointed out “regulatory provisos […] that accommodate for the specific requirements of ethnographic research” (Corsín Jiménez 2018: 1), creating room for ethnography’s exploratory character and recognizing the dialogical nature of its data-constitution practices.

This rant picks up one specific issue: the way GDPR engages with the ‘public sphere.’ Contrary to what many of us (including myself) initially naively assumed, the new directive does not provide a free ticket for processing personal data that are already publicly available. A crude lexicostatistic analysis of the regulation reveals that the adjective public systematically collocates with authorities, interest, and security. Public availability and public accessibility are occasionally mentioned, but as a basis for positing additional restrictions rather than as a license for processing. In short, GDPR is concerned with personal data, regardless of whether they were collected in a ‘public space’ or in the ‘public domain’. The GDPR’s restrictions on their use apply regardless. Yet, interpreting and applying GDPR does force us to critically interrogate what we precisely mean when we talk about public space. This is what this text purports to do. I will try to recover some of the polysemy hidden underneath the surface of this presumably monolithic category, and to this end, I will be tossing Goffman against Habermas (without claiming any credits as to the originality of such a move, see, for example, Sarangi 2011). Note that I am not interested, at least not initially, in distinguishing between ‘public place,’ ‘domain’, or ‘sphere.’ The underlying spatial metaphor that is present in all three of them refers “not [to] a thing but rather [to] a set of relations between things” (Lefebvre 1991: 83), and that is what matters at this point.

The intimacy of the public sphere

One anchoring point for theorizing the public sphere is Goffman’s well-known distinction, first advanced in The presentation of self in everyday life (1959), between front– and backstage. Frontstage is the arena where social actors act out the roles that are societally and institutionally expected from them, carefully controlling their performance in the knowledge that it will be evaluated by an audience of others. The backstage is the region where actors prepare for this role performance, where they can deviate from the script and engage in the performance of a more authentic self, unexposed to the public gaze. The notion of frontstage, then, is the one that is readily equated with the ‘public’ sphere (see, for example, Sarangi 2009). However, one should not forget that frontstage also inevitably refers to a constellation of people entangled with one another through co-presence (Goffman 1963). Co-presence occurs whenever social actors are aware that they are within each other’s perceptual range. It manifests itself in relentless mutual monitoring and in the reflexive self-regulation of one’s own conduct, in an attempt to regulate whatever information about ‘self’ and ‘other’ might be leaking through.

If we look at it from this perspective, a large part of what we routinely gloss as ‘public’ suddenly acquires an unexpected intimate quality. Public conduct is not just a scripted role. It is a scripted role that we inhabit, and therefore it will always contains glitches, moments of uncertainty, and failures to meet expectations. These are highly intimate moments, and GDPR enforces respect for this intimacy. Nobody likes being caught on camera picking their nose, or arguing with their partner on the way to the supermarket. As a rule, GDPR prevents the processing of personal data collected during such ‘public’ performances of intimate conduct. The same rules apply as to other information which allows individual data subjects to be identified: You can only process personal data collected in a public place (for example, footage containing faces and/or voices) if you can demonstrate that processing is done based on one of the lawful grounds specified in Art. 6 (consent of the data subject, processing in the public interest, etc.). In a way, this makes perfect sense. The fact that someone engages in a certain kind of behavior in a publicly accessible space should not be taken to mean that it is up for grabs as data.

This new GDPR framework does not entirely prohibit Goffman-styled inquiries into how people navigate public spaces. Although it forbids processing personal data (photos, video footage, audio recordings) collected without permission from the data-subject, carefully anonymized field notes (which do not provide any clues as to the identity of the data subject) are still perfectly lawful. A good example is my earlier work on the various forms of practical geographical knowledge that Dar es Salaam commuters need for traveling from one part of the city to another (D’hondt 2009). As a rare instance of conversation-analytic research that does not use recordings of conduct, it combined a systematic interrogation of my own practical knowledge, as a user of informal minibus transport, with vignettes illustrating “noteworthy patterns of behaviour” (Iphofen 2015: 47) that I occasionally jotted down while on the move. As such, it represents a form of observational practice that, although not covert, elides the formal solicitation of consent (Iphofen 2015: ibid.). This practice can be questioned from an ethical angle, but under GDPR it does not pose a problem of legal compliance since no personal data are being collected. Dar es Salaam is a megalopolis of over four million inhabitants. Each day, literally hundreds of thousands of city residents rely on minibus transport for getting to work, school, etc. There is no way in which my fellow commuters whose conduct accidentally drew my attention could ever be identified from the short written notes that I entered in my notebook, let alone from the schematic descriptions of courses of action that ended up in the resulting publication.

Public, as in res publica

In The structural transformation of the public sphere (1962[1989]), Habermas describes how the aggregate of print media, coffee houses, lodges and reading clubs that spread through 18^th century Europe resulted in the formation of a new critical space, separate from the state, where individuals could interact and exchange ideas on public matters in a way that had never been possible before. Habermas’ account of this budding ‘public sphere,’ which mediates between the private life-world and the state apparatus, has subsequently been criticized from various angles. Later on, Habermas (1981) himself added that in modern mass-societies, the available space for rational-critical debate is gradually contracting, as clear-cut demarcations between the public and private, and between state and society, are eroded by bureaucratization and consumerism. Feminist scholars pointed out the policed nature of the public-private distinction and the pervasiveness of exclusionary mechanisms, while poststructuralists and postcolonialists called into question its monolithic character and insist on a plurality of public spaces (for a useful overview, see Koller and Wodak 2008). Still, the idea of a universally accessible common deliberative space has a strong normative presence in contemporary society. It is enshrined in Art. 10 of the European Convention on Human Rights, on the freedom of expression and information. Social scientists are an integral part of this deliberative space, and instances of critical interventions that have transformed ongoing societal debates are manifold. Blommaert and Verschueren’s (1998) critical inquiry into the rhetoric of tolerance underpinning the 1990s debate on immigration in Belgium is a classic example of a study that fundamentally altered the structure of the debate it commented upon, by pointing out uncanny similarities between the discourse of far-right parties and of the self-declared tolerant majority. Arguing pro domo, I would say that my current ethnographic work on trial performance at the International Criminal Court (ICC; see, for example, D’hondt 2019) falls under the same category. Transnational legal orders, like the one epitomized by the ICC, exert a growing impact on national legal systems. However, the way in which the public exercise of authority by this transnational institution is negotiated in the court’s daily operation has never before been subject to critical scrutiny.

International forums like the ICC (but also the UN’s Universal Periodic Review or the UN Human Rights Committee, see Cowan and Billaud 2017, Halme-Tuomisaari 2018) are undeniably ‘public’ in the sense of Goffman, as the actors involved will always tailor their frontstage conduct to that of the other participants who are co-present (one might also add virtual audiences here, as many of these events are live-streamed). Still, one can sensibly argue that many of the restrictions on data processing mentioned in the previous section do not apply here, precisely because what transpires on these international forums is intrinsically connected to the deliberative space Habermas identified. Importantly, GDPR offers us a good starting point for developing such an argument.

GDPR formulates a set of general principles for personal data processing that apply universally. However, in addition it also specifies a number of ‘specific processing situations’ in which the application of these rules is legitimately curtailed (‘derogations’). It is up to the respective EU member states to specify the precise scope and content of these exemptions. In the national data acts through which the different member states implement GDPR, they must itemize the articles of the original GDPR that do not apply in these specific processing situations. Hence, GDPR Art. 89(2) stipulates that national legislations may provide exemptions for data processing “for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes.” Significantly, this article does not contain any provisions as to where the research in question is supposed to take place, and Recital 159 explicitly includes “privately funded research.” In fact, the only place in GDPR that explicitly mentions academia is Art. 85, on the freedom of expression. This article gives EU member states the responsibility to specify exemptions in order to “reconcile the right to the protection of personal data pursuant to this Regulation with the right to freedom of expression and information, including processing for journalistic purposes and the purposes of academic, artistic or literary expression.” Of course, the nature of these exemptions will slightly vary form one member state to another, but the overall effect of Art. 85 is a “redefinition of the social sciences on a par with journalism, characterizing their activities as in the public interest and their pursuits as in themselves valuable as academic knowledge” (Sleeboom-Faulkner and McMurray 2018: 23). Not surprisingly, the British Academy and the Economic and Social Research Council in 2017 advised that British universities should maximally exploit the derogations this article provides. Similarly, the European Association of Social Anthropologists explicitly insists that “ethnographic research [should be placed] within the special clause on ‘academic expression’ included in Article 85(2) of the GDPR [, … which] has been designed to guarantee the critical social value of humanities and social sciences research” (EASA 2018: 1).

It seems, then, that there are at least two sets of derogations that social scientists can invoke to legitimize their data processing: the research exemption (Art. 89) and the one that guarantees the freedom of academic expression (Art. 85). For those of us working in critical research traditions, this second exemption is particularly relevant and liberating. Although GDPR makes no mention of the public sphere as such, it nevertheless provides the necessary exemption social scientists require for participating in the shared deliberative space outlined by Habermas, guaranteeing their right to critically engage with ongoing societal debates and allowing data processing on the condition that it is necessary to execute that right.

Ethics, after all

As I anticipated at the start, this excursion into legal compliance finally takes us back to ethics. For the time being, GDPR is still in the process of interpretation, and we are still waiting for case law in order to find out where lines will be draw and how GDPR will eventually be interpreted. This is a moment of uncertainty, and there is a tendency among universities and funders to err on the side of caution.

The chilling effect of these new regulations is real. It should be clear, however, that GDPR, in fact, offers quite some maneuvering space to those of us who are working with public data. At this stage, it would be perilous for applied linguists and ethnographers to assume a quietist stance.

The most dangerous kind of censorship is self-censorship.

In the current circumstances this aphorism is more relevant than ever. As scholars, we should be prepared to defend the available space for critical intervention. Universities should be prepared to back its scholars in this, as this debate revolves around basic democratic principles. In the end, even going to court should not necessarily frighten us. After all, it is an opportunity to stand up for the values we believe in.

References

Blommaert, Jan, and Jef Verschueren. 2002. Debating Diversity. London: Routledge.
British Academy & ESRC. 2017. A British Academy and Economic & Social Research Council submission to the Department of Culture, Media and Sport (DCMS) call for views on the General Data Protection Regulation derogations.
Corsín Jiménez, Alberto. 2018. A Data Governance Framework for Ethnography. Madrid, CSIC.
Cowan, Jane K. and Julie Billaud, 2017. The ‘public’ character of the Universal Periodic Review: Contested concept and methodological challenge. In: Niezen, R. and M. Sapignoli (Eds.), Palaces of Hope. Cambridge, Cambridge University Press, pp.106-126.
D’hondt, Sigurd. 2009. Calling the stops in a Dar-es-Salaam minibus: Embodied understandings of place in a drop-off routine. Journal of Pragmatics 41, no. 10 (2009): 1962-1976.
D’hondt, Sigurd. 2019. Humanity and its beneficiaries: Footing and stance-taking in an international criminal trial. Signs and Society 7 (3), 427-453.
European Association of Social Anthropologists. 2018. EASA’s Statement on Data Governance in Ethnographic Projects.
Goffman, Erving. 1959. The Presentation of Self in Everyday Life. Garden City, NY, Doubleday.
Goffman, Erving. 1963. Behavior in Public Places. New York: Free Press.
Habermas, Jürgen. 1962 (1989). The Structural Transformation of the Public Sphere. Cambridge, Polity.
Halme-Tuomisaari, Miia, 2018. Methodologically blonde at the UN in a tactical quest for inclusion. Social Anthropology 26 (4), 456-470.
Humphris, Rachel. 2018. Is anthropology still legal? Notes on the impact of GDPR. Birmingham, IRiS.
Iphofen, Ron. 2015. Research Ethics in Ethnography/Anthropology. European Commission, DG Research and Innovation.
Koller, Veronika, and Ruth Wodak. 2008. Introduction: Shifting boundaries and emergent publics. In: R. Wodak and V. Koller (Eds.), Handbook of Communication in the Public Sphere. Berlin: Mouton de Gruyter, pp. 1-17.
Lefebvre, Henri. 1991. The Production of Space. Oxford: Wiley-Blackwell.
Sarangi, Srikant. 2011. Public discourse. In: Jan Zienkowski, Jan-Ola Östman, and Jef Verschueren (Eds.), Discursive Pragmatics. Amsterdam: Benjamins, pp. 248-265.
Sleeboom-Faulkner, Margaret, and James McMurray. 2018. The impact of the new EU GDPR on ethics governance and social anthropology. Anthropology Today 34 (5), 22-23.