PhD defence by Trine Rask Nielsen
Title
Abstract
The process of granting asylum in Denmark relies on a concrete and individual risk assessment by asylum authorities. Moreover, it depends on the individual’s ability to present a valid asylum motive, supported by documents and objects deemed “credible” by the authorities and aligned with the legal categorization of a “refugee.” As part of this process, there is an increasing expectation that data and large-scale datasets can be transformed into knowledge instrumental for researchers, asylum authorities, and NGOs, either as part of their practice in determining asylum eligibility or for gaining deeper insights into and enhancing asylum decision-making processes. However, data and datasets are simplifications of the world. Therefore, understanding the role of data in asylum decision-making necessitates in-depth, situated investigations into the technologies, individuals, practices, settings, regulations, and politics involved in the construction and interpretation of data representing individuals applying for asylum.
This dissertation is grounded in the research fields Computer-Supported Cooperative Work (CSCW), Critical Data Studies, and draws on principles from Data Feminism. By adopting a socio-technical lens, I qualitatively explore the theoretical concept of “data work” within an asylum casework context. I draw inspiration from a multi-sited approach to ethnography with the aim of understanding the reality that caseworkers model when constructing data informing asylum cases. I shed light on relational aspects of data production, such as care, from a caseworker’s perspective, spanning NGO and authority levels. This includes Red Cross caseworkers at a departure center, the Danish Refugee Council’s legal team, and Immigration Service caseworkers. The main theoretical contribution of this dissertation lies in expanding the concept of data work to include aspects of care.
First, I find that caseworkers’ data work, defined by the production, contextualization, interpretation, and leveraging of data, plays a pivotal role in shaping how the “asylum seeker” is construed in and through data. Streams of data, which come to represent the individual applying for asylum, are produced, interpreted, and categorized by caseworkers and influenced through various steps of translation work and discretionary practices across NGO and authority levels. In this context, I find that existing categories cannot always capture the intricate and complex social realities of individuals applying for asylum or those who have been rejected. Therefore, it is crucial to challenge and explore the relational aspects of casework that underlie how the “asylum seeker” is construed in and through data. Furthermore, the significance lies in the fact that these data turn into large-scale datasets that come to mirror and perpetuate a specific reality or a “ground truth” when data science techniques are applied. This happens as researchers, asylum authorities, and NGOs increasingly utilize and analyze datasets, aiming to gain deeper insights and enhance asylum decision-making processes.
Second, I find that the Danish asylum system consists of intricate data infrastructures maintained by the social practices of work, as well as the physical and material systems and databases established through relational processes and political negotiations at both authority and NGO levels. Caseworkers, spanning these levels, exist within different work settings and situations, each with distinct responsibilities and perspectives. However, they share a goal to better understand and support individuals applying for asylum while ensuring legal compliance. The empirical data presented in this dissertation show how data about asylum seekers produced by both authorities and NGOs flow into and intersect multiple casework systems.
Third, I find that relational aspects of care in casework, encompassing moments of ambivalence, translation work, and attentiveness to “new substantial information” in some cases become indispensable in mitigating the limitations or potential shortcomings present in the formal procedural data-driven approaches. I find that individuals applying for asylum need caseworkers’ translation to understand the process and draw up a map of the asylum system’s many opaque and complex bureaucratic rules and data-driven procedures. My findings exemplify how it is imperative to examine these less apparent “situated actions” before introducing data-science methodologies into asylum casework. This is important because overlooking these informal discretionary work practices, potentially influential in data construction, can impact data quality and consequently result in an asylum case not being sufficiently documented.
Finally, I find that it adds greater complexity to the asylum process when authorities incorporate new data infrastructures, such as extracting data from asylum seekers’ mobile phones, into their daily decision-making practices in an opaque and unsystematic manner. This, in turn, exacerbates challenges related to the agency and accountability of individuals seeking asylum. Altogether, I find that there is a greater need for transparency and collaboration across authority and NGO levels. This is important to ensure that individuals applying for asylum receive fundamental legal support throughout and after the introduction of new data infrastructures. In this context, I argue for a systematic approach to data literacy. However, I find lingering ambiguity among authorities and NGOs regarding the responsibility for supporting data infrastructure literacy initiative to empower asylum seekers in exercising their legal and human rights.
In Part I, I first introduce the four overarching research questions that guide this dissertation. Second, I provide a brief presentation of the asylum procedure, the legal basis for granted asylum, and the various categories of residence permits in a Danish context, followed by a succinct overview of the asylum authorities and procedures in Sweden and Norway. Third, I introduce the theoretical positioning within this dissertation, followed by a description of the methods I used. This includes presenting the main research sites of the project and outlining the methods employed to gather empirical data for analysis and achieve empirical saturation. Lastly, I present the overall contributions and findings of this dissertation. Part II comprises four research papers (two published, and two submitted), each delving into a distinct topic related to the overarching research questions of this dissertation.
Supervisors
Principal supervisor:
Associate Professor Naja Holten Møller, University of Copenhagen, Denmark
Co-supervisors:
Professor Thomas Troels Hildebrandt, University of Copenhagen, Denmark
Professor Thomas Gammeltoft-Hansen, University of Copenhagen, Denmark
Assessment Committee
Professor Irina Shklovski, University of Copenhagen, Denmark
Professor Myriam Lewkowicz, Troyes University of Technology, France
Associate Professor Neha Kumar, Georgia Tech, USA
Leader of defense
Associate Professor Tariq Osman Andersen, University of Copenhagen, Denmark
For an electronic copy of the thesis, please visit the PhD Programme page.