'Where Technology Leads, the Problems Follow. Technosolutionism and the Dutch Contact Tracing App' by Lotje E. Siffels and Tamar Sharon in (2024) 37 Philosophy & Technology 125 comments
In April 2020, in the midst of its first pandemic lockdown, the Dutch government announced plans to develop a contact tracing app to help contain the spread of the coronavirus – the Coronamelder. Originally intended to address the problem of the overburdening of manual contract tracers, by the time the app was released six months later, the problem it sought to solve had drastically changed, without the solution undergoing any modification, making it a prime example of technosolutionism. While numerous critics have mobilised the concept of technosolutionism, the questions of how technosolutionism works in practice and which specific harms it can provoke have been understudied. In this paper we advance a thick conception of technosolutionism which, drawing on Evgeny Morozov, distinguishes it from the notion of technological fix, and, drawing on constructivism, emphasizes its constructivist dimension. Using this concept, we closely follow the problem that the Coronamelder aimed to solve and how it shifted over time to fit the Coronamelder solution, rather than the other way around. We argue that, although problems are always constructed, technosolutionist problems are badly constructed, insofar as the careful and cautious deliberation which should accompany problem construction in public policy is absent in the case of technosolutionism. This can lead to three harms: a subversion of democratic decision-making; the presence of powerful new actors in the public policy context – here Big Tech; and the creation of “orphan problems”, whereby the initial problems that triggered the need to develop a (techno)solution are left behind. We question whether the most popular form of technology ethics today, which focuses predominantly on the design of technology, is well-equipped to address these technosolutionist harms, insofar as such a focus may preclude critical thinking about whether or not technology should be the solution in the first place.
At the outbreak of the COVID-19 pandemic in early 2020, governments and health authorities around the world turned to digital technology for support in addressing the rapid spread of the virus. Particularly promising was the idea of auto-mating contact tracing; a public health strategy for containing infectious disease outbreaks that involves identifying infected individuals and informing the people they have been in contact with that they are at risk (EC, 2020; WHO, 2020). Digitizing (part of) the contact tracing work would enable the employment of this tried-and-tested method on the large scale of infections that COVID-19 brought with it (de Jonge, 2020; Ferretti et al., 2020; Wacksman, 2021). At the same time, digital contact tracing raised substantial privacy and surveillance risks (Ada Lovelace Institute, 2020; McGee et al., 2020), had a high probability of false positives (Lee, 2020; Vaughan, 2020), and its overall effectiveness was (and remains) questionable (Chiusi, 2021; ter Haar et al., 2023). Many countries ended up developing smartphone- based apps that automatically trace points of contact between individuals, and alert users when they have been in contact with someone carrying the virus. The design of the apps and its consequences for citizens’ privacy was fiercely debated and differed per country, with the Bluetooth-based and decentralized design being the most widespread (Shahroz et al., 2021; Lanzing et al., 2022). At the time of writing, most of the apps are no longer in use (EC, n.d.).
On April 7 2020, during the first lockdown, the Dutch government also announced plans to develop a contact tracing app, the Coronamelder. The announcement was a response to the overburdening of manual contact tracers, who could not keep up with the growing need for contact tracing. Half a year later, in the fall of 2020, the Coronamelder was released for public use. But where initially the app was supposed to address the overburdening of manual contact tracers by providing support, the app now no longer seemed to have anything to do with this overburdening. The aim of the app had changed drastically. In this paper we contend that throughout the process of developing and rolling out the Coronamelder, the problem the app was initially meant to solve changed over time without the solution – a smartphone-based contact tracing app – undergoing any modification. We argue that this makes it a prime example of technosolutionism.
Numerous critics have argued that the pandemic has shown how, in times of crisis, there is a tendency to turn to digital solutions to address public policy problems (Klein, 2020a, 2020b; Morozov, 2020). Many references to technosolutionism in relation to contact tracing apps were made, both in general (Harjric, 2020; Michael et al., 2020; Mann et al., 2022), and in the Dutch context (Appelman et al., 2020).
However, with some exceptions (Campbell-Verduyn & Gstrein, 2024; Maschewski & Nosthoff, 2022), there has been little effort to describe the political development and rollout of contract tracing apps and the specific role technosolutionism played in it. Moreover, most discussions of contact tracing apps and technosolutionism make use of what we call a thin conception of technosolutionism. This overlooks its constructivist dimension and equates it with the simpler concept of “technological fix”, thereby failing to recognize the potency – and dangers – of technosolutionist thinking in public policy contexts, and how it might be prevented in the future. Through this article, we hope to contribute to the literature on technosolutionism in several ways: by elaborating a thick conception of technosolutionism that embraces a constructivist approach, by showing what technosolutionism in the context of public policy making does, and by showing which harms a technosolutionist problem construction can provoke. We do this by following the problem that the Coronamelder was supposed to solve, and how it shifted to fit the Coronamelder solution, rather than the other way around, in the period between April and December 2020. Our analysis focusses specifically on the way government officials who were involved in the development of the app conceptualized the problem it sought to solve. We reviewed policy documents during this time that highlight the purpose and development of the app, including advice from committees, press conferences, parliamentary debates and letters from the Minister of Health to inform Parliament. We specifically sought out policy documents where the aim or purpose of the app are described, explicitly and implicitly, as well as the app’s effectiveness. Subsequently, to clarify our findings we had several conversations with persons involved in the development of the app, amongst whom government officials.
The paper proceeds as follows. In section 2, we distinguish between a thick and thin conception of solutionism, where the thick conception follows Morozov’s (2013) definition of technosolutionism, which involves a constructivist and temporal element of looking back from the solution and constructing a problem that justifies its implementation. Taking inspiration from Latour (2003), furthermore, we argue that it is not the fact that problems are constructed per se that is problematic with technosolutionism. Rather, that technosolutionist problems are not constructed well. In the context of public policy making, we contend, good problem construction implies careful deliberation and discussion of alternative solutions. Something we find to be painfully lacking in technosolutionism, we argue in Section 3, as we show how the problem definition follows the solution and is twisted to fit the solution, rather than the other way around. In Section 4 we zoom in on how technosolutionist problems are constructed. We show several moments of contestation, negotiation and stabilization in the construction of the problem that the Coronamelder sought to solve, which help us identify technosolutionism as a mechanism. In Section 5, drawing on our analysis, we describe some of the general harms of technosolutionism in the public policy context. These include a subversion of democratic decision making, the presence of new and powerful actors in the form of Big Tech companies, and the creation of what we call “orphan problems”: problems that initially triggered the development of a technosolution, but were abandoned when they no longer fit the technosolution. We believe that the ethical reflection which had been applied to the Coronamelder at the time was ill-equipped to address these harms that technosolutionism brings forth, since its focus was exclusively on the design of the technology. This focus precludes critical thinking about whether or not the technology should be developed in the first place. We contend that this type of ethical reflection is prevalent in ethics of technology and argue for a broader ethical reflection, one that includes consideration of democratic values and the question of how technological (vs. non-technological solutions) may impact them.