The public sector introduces digital technologies (DT) to deliver public services and to support the political discourse based on ideas of innovation, cost efficiency, and effectiveness. Current literature reflects on what is needed to succeed in so doing where failure is generally explained as the lack of achieving expected benefits. We here argue for the need to discuss digital technology's potentially harmful effects on citizenship, not only as the absence of achieving what is considered good but as direct negative consequences for citizens. Based on the dimensions of citizenship and the concept of administrative evil, we suggest an analytical approach to identify the harmful consequences of introducing digital technologies in governments. We present an exemplary case from Norway to demonstrate the value of our analytical approach. We argue for the need for more research focusing on exploring the causes and causality of administrative evil to increase awareness of the negative implications of digital public services provision on the dimensions of citizenship.
The e-government literature champions using technologies to make bureaucratic systems
more efficient [
We challenge such logic by arguing for the need to explore the negative consequences of digitalization for citizenship. We do so by discussing how DT may harm citizenship in digitalized societies where the adoption of digital services has become an integral part of a global political culture [3]. Citizen-centricity is important since DT should be designed to address citizens' interests [4], raising the need for more awareness of the potential adverse outcomes that may be ingrained in them.
The terminology of citizenship add value to capture the dynamics between citizens and the state. Citizenship usually refers to the connection between a nation or territory and their individual members [5, 6], entitling civil, political, and social rights and obligations. Contemporary digital societies encompass individualized processes, governance structures, and services that promote self-directed citizenship in a new technology-enabled democracy [7]. Through DT, resources can be relocated, and work can be transferred to citizens who engage in digital service procedures, with different outcomes depending on individual preconditions [8].
Previous literature acknowledge that DT can negatively impact the most vulnerable citizens [9, 10], raising the need to critically reflect on the potential negative outcomes of DT. In this paper, we explore the impact of DT on citizenship by adopting the administrative evil theory, as exploring administrative evil dynamics will allow to explore conditions and causes for the negative impacts of DT on citizenship. The concept of administrative evil can be understood as a masked behaviour performed by bureaucratic apparatuses that inflicts pain and causes harm to innocent people, possibility inadvertently, without perpetrators – the bureaucrats - being aware they are doing something wrong. Quite the contrary, they may believe they are complying with their roles as expected by pairs and superiors, following the best technical rationality [11].
To analyse how DT adoption can carry traces of evil, seen in the negative outcomes for citizens that harm citizens' relations with the public sector and jeopardise citizens' abilities to take part in society, we combine administrative evil and citizenship theories. The value of our approach is explored by an exemplary case study of the Norwegian Labour and Welfare Administration (NAV).
The paper is structured as follows. First, we present our conceptual approach, combining citizenship and administrative evil theories in the context of digital societies. Second, we present our research design and the NAV case description. Preliminary findings and analyses are then presented, and the discussion and conclusion follow. A future version of this paper will include more illustrative cases and further methodological steps.
This section brings together our conceptual approach. In 2.1, we discuss citizenship in a digitalized society, adopting Stokke's framework [5]. We then discuss how the public sector bureaucratic apparatus fosters citizenship through DT (section 2.2). Finally, we mobilise administrative evil theory in relation to DT in the public sector (section 2.3).
Although the concept of citizenship has been extended through the years because of
migratory movements, globalization, and political struggles regarding identity [5],
citizenship usually refers to the link between a particular nation or official territory and the
members of that community. This relation recognizes and assures individuals as detainers
of an identity, rights, and obligations capable of participating and acting in a specific
geographical space [6]. More than that, citizenship constitutes the nuclear idea of welfare,
aimed at assuring some degree of equality and well-being to all and combating excessive
social stratification [
Numerous conceptual models on citizenship delineate the complex relationship between
citizens and the state as citizenship reflects historically and geographically situated political
relations, ideological orientation, comprehension of democracy, and expectations of
citizens' roles [5, 6], although the focus on rights and participation is central in most models
[
In several countries, DT are the primary way for citizens to engage with the government, demonstrating an ideological pouch for digital society and what has been called "digital citizenship" [6]. Such citizenship is formed in relation to DT and is characterised by online encounters between citizens and the state participating out of a digital-first ideal [6]. Ideally, DT can create ownership and empowerment for citizens and bring efficiency gains for governments [8] with the "self-serving citizen" profile borne out of such ideals [12]. When work tasks are transferred to citizens in the digitalized society, fewer resources will supposedly be required from administration and public officials [8].
DT are usually designed with citizens' interests and needs in mind, easing citizens' workload in service procedures [4]. In this sense, DT can reduce burdens for the public sector and citizens [13]. However, DT rarely benefit every citizen [14]. In fact, DT harm some citizens more than others by imposing burdens that make citizens struggle to access services [15].
While citizenship figures in a digitalized society, it is, in reality, rooted in offline lives. This makes the "digital citizen" a state that can be achieved rather than a fixed construct [19]. However, implying the existence of a "digital citizen" or a "digital citizenship" leads to assuming the existence of a second-class non-digital citizen who would be automatically excluded in the digitalized society. The use of DT by governments can be viewed in this sense as both a means to achieve and fulfill citizenship dimensions or imposing obstacles that prevent citizenship actualization in practice. Therefore, a broader understanding of citizenship in a digitalized society is needed, emphasizing society's responsibility to put preconditions in place for all citizens to participate.
Actualizing citizenship is fundamental for individuals' inclusion in society, life, and wellbeing, making the citizenship dimensions essential to understand concerning DT. Citizenship is not fixed; it is formed from political decisions, making its preconditions depend on the political agenda that sets the DT targets expressed in government DT practices [6]. To study the impacts on citizenship resulting from DT use in the public sector, we adopt Stokke's [5] integral framework composed of four essential and interrelated dimensions: 1. Membership - distinguishes insiders and outsiders in a community from ethnocultural and juridical-political perspectives. 2. Legal status implies a contractual relationship between citizens and state, with rights and responsibilities. 3. Rights – relates to Marshalling [17] threefold typology of civil, political, and social rights associated with membership and legal status. 4. Participation - involves active engagement in governance, public affairs, and community responsibilities.
The possibility of citizens exerting their rights is directly affected by the services delivered by public sector organisations. In democratic countries, citizens have the right to elect representatives who, in turn, appoint bureaucrats to manage the public sector bureaucratic apparatus that will deliver the public services [20]. DT are one of the cornerstones through which contemporary public bureaucratic apparatuses work. Hence, DT influence how citizenship is shaped, promoted, and developed in a digitalized society. To understand the implications of DT for administering citizenship, we must point out the drivers of public sector organisations and their potential implications.
The bureaucratic apparatus is an instrument through which public governance promotes citizenship. Such apparatus is guided by ethical principles that can be inspired by democratic or bureaucratic values [20]. When public bureaucracies act on democratic values, they foster public interest, citizen empowerment, social equity, justice, and benevolence. When acting on bureaucratic values, they promote efficiency, effectiveness, loyalty, and accountability.
From a public governance perspective, policymakers, in their activities, act teleologically, promoting the greatest good for the greatest number of people, and see the bureaucratic apparatus as a means to achieve this goal [21]. Public bureaucrats, on the other hand, act deontologically, seeing the bureaucratic structures as the primary goal in reaching the objectives set by the policymakers. So, concerning our context, politicians would issue policies to use DT to create better conditions for most citizens following a teleological approach, while bureaucrats would focus on pushing DT use by citizens to materialise public policies. Such perspective difference can be the source of a significant drift in public sector action: administrative evil.
The concept of administrative evil refers to harmful consequences materialised by administrative actions of bureaucratic systems that, at the same time, correspond to a rational action of bureaucracies while also producing a negative outcome [11]. When bureaucratic structures act this way, they may harm and engage in acts with various detrimental outcomes, such as white lies, damage to organisational reputation, psychological harm, ethical harm, or even physical harm [11, 21, 22]. Such harm is often not recognised as a negative outcome; quite the contrary, bureaucrats frequently believe that they are doing their duties and that their actions are necessary to follow their deontological stance [23]. Hence, administrative evil must be unmasked, unveiling factors and causal mechanisms that produce the negative outcome to make it visible to everyone [11]. In this work, we posit that when DT promoted by public policies affect the dimensions of citizenship, they can produce negative effects that harm citizens, materialising into elements of administrative evil.
Exploring administrative evil requires investigating the bureaucratic context in which the phenomena occur and observing how DT are integrated. Administrative evil is unveiled when the negative outcome can be connected to characteristics of the organisational action that make the reproduction of the harmful outcome systematic. The literature refers to several dimensions of administrative evil, such as bureaucratic impersonality and professionalism, technical rationality, market logic, decentralisation or compartmentalisation, loss of ethical or moral standards, or technological amplification [11, 21, 22, 24]. However, these dimensions are not exclusive, and future research may explore new dimensions or mechanisms of evil.
Administrative evil has already been associated with the use of DT in organisations. Still, current literature limits the conceptual exploration of evil on emerging technologies such as artificial intelligence and blockchain [25–27]. No research has explored administrative evil in using DT regarding citizenship with an empirical approach, and the literature provides no specific frameworks for the analysis of administrative evil. In this work, we will explore administrative evil by combining three elements: 1. Contextual factors – such as bureaucratic impersonality, professionalism, technical or technological rationality – associated with the negative outcome (i.e. the digital platform is the more convenient and cost-effective way to deliver services to citizens). 2. A negative outcome – such as psychological, reputational, or organisational harm related to citizenship – that is considered normal by bureaucrats (i.e. it is normal that a citizen cannot exert one of his rights if he does not use the online digital platforms). 3. A bureaucratic logic - connecting contextual factors with negative outcomes (i.e. street-level bureaucrats pushing citizens to go online when they indicate difficulties using online systems because no services can be provided if not online).
Our work is motivated by the following RQ: How can DT implementation in governments harm citizenship in a digitalized society? To answer this we explore the prospect of DT use in the public sector aligned with the concepts of citizenship [5] and administrative evil [24], applied to one illustrative case from Norway. The illustrative case is described in section 3.1 and is well documented in literature from a citizen perspective [9, 15, 29].
Regarding citizenship, we build on Stokke's [5] framework by analysing the impacts of each of the dimensions of citizenship proposed by the author. Concerning administrative evil, we highlight how the application of DT harms dimensions of citizenship, searching for traces of evil in the negative dissonance between objectives (contextual factors) and negative outcomes of the DT use that together compose a bureaucratic logic. The context is DT in self-service practice, which we analyse out of the dimensions of citizenship and detect traces of administrative evil.
Following a qualitative approach, a future version of this ongoing research will explore the factors and causal mechanisms behind the bureaucratic context associated with DT, in which the phenomena occur. Data for this phase of the study will be collected through interviews with bureaucrats involved in DT design and implementation on the cases selected.
The Norwegian Labor and Welfare Administration (NAV) has been a key player in realizing the digital transition of the public sector in Norway with key milestones in 2005, 2006, and 2010, developing a web portal nav.no and the online service platform myNAV [29]. The COVID-19 pandemic accelerated NAV's digitalization, replacing physical services with remote options on nav.no and self-service stations at local offices. However, research in NAV has shown that the digital-first ideal sometimes collides with citizens' needs and interests [9]. Despite the success of digital service development, the digitalization of welfare services poses challenges for citizens who struggle to access services online [15, 29].
A research project that focuses on how citizens in need of welfare benefits experience the digital services in NAV, demonstrate how many citizens who visit NAV offices struggle to use digital services. Self-service stations provided at the NAV offices offer citizens to access the platform nav.no and allow citizens to log in to their sight of myNAV. Some citizens managed their self-services at the stations, while others struggled with access or needed assistance from frontline workers [9].
During observations we found that most citizens who asked for personal assistance to apply for financial benefits were directed by frontline workers to use digital self-service stations and log on to nav.no. Citizens who asked for paper applications did so as they struggled to manage their errands at the digital service stations. They were often denied paper applications as the office protocol was to increase the number of digital applications and to promote self-services as the first choice. Many citizens could not engage as active service co-producers as they struggled to navigate the platform and submit personal information. The fear of doing something wrong, lack of digital knowledge, language barriers, and lack of access to digital services needed to identify them restricted them from participating in digital self-service procedures. As a result, several gave up when they could not manage the digital services at nav.no [9].
To actualize dimensions of citizenship became challenging for citizens who struggled to use digital self-services. Some of the citizens who required NAV services were often vulnerable in several aspects relating to physical or mental health issues, poverty, and language barriers. For many citizens that came to the office, the digital stations and nav.no were more of an obstacle than an access point to NAV.
This section analyses how DT in NAV produced negative outcomes for citizens seeking services, highlighting the traces of administrative evil and its impact on citizenship dimensions [5]. Findings are summarised in Table 1, showing traces of evil in harmful outcomes that challenge citizenship, such as exclusion, dropout, and rights difficulties.
Concerning citizenship's membership dimension, the digital community that myNav represents was supposed to include citizens. However, this was only a reality for citizens who knew NAV services and could understand myNav to practice such membership. Many citizens struggled to access the full-service span at myNAV and to find the services they needed online. Without this, the membership was of little value. Digital accounts and accessible service stations did not secure membership and belonging for all citizens. In fact, this shows that digital service stations and platforms do not foster community inclusion and membership only by functionality that supports such membership.
It shows traces of evil rising from the contextual ideal of digital efficiency and bureaucratic logic that most citizens shall use DT according to office protocol. Citizens with preconditions to take part could access membership, while the exclusion from membership rises when citizens are directed to a digital community without the ability to be full members.
We also found negative outcomes concerning the dimension of legal status. In NAV, the system provides different authorization options related to the "legal status" dimension, which refers to identity [5]. While in theory, digital identification supports legal status, our observations demonstrate that some citizens are denied digital ID due to addiction or health issues or failure to submit their personal data. Consequently, DT prohibits them from actualizing their legal status. Citizens who managed to actualize their legal status by logging in to nav.no also faced barriers when they struggled to navigate the platform and understand the services and the NAV system. In these cases, the actualization of the legal status that granted them access to rightful services was of little use [19].
When the frontline workers ensured that citizens had a BankId, they often left citizens on their own to engage in the self-service, resulting in several citizens failing to log in. The organisational logic that DT save resources from administration gave public officials less room to assist citizens. Some citizens gave their IDs to other citizens to help them. The bureaucratic logic of DT use and contextual efficiency of self-service direction to exchange human assistance took focus away from ensuring legal status realisation.
Finally, negative outcomes were also found in relation to the citizenship dimension of rights. In NAV, these rights are seen in welfare services. When citizens apply for financial welfare services in NAV, they must match the eligibility criteria to receive benefits. Still, eligible citizens directed to use digital self-service often were afraid to make mistakes and face the consequences of potential digital wrongdoing. This prevented some citizens from following through. Language barriers made citizens miss out on services as they could not read about their rights at nav.no. Many needed personal guidance and information about their rights and could make choices based on a novel understanding of such rights.
In this case, the logic that DT can compensate for human interaction ended up excluding citizens from their social rights. The context was that digital self-services provided efficiency from an organizational goal and motive but restricted how citizens exercise their rights. Citizens' struggle at service stations or, worse, dropping out of the welfare system was the outcome of DT, and as such, preconditions challenge equality concerning rights.
We could not identify negative outcomes in the participation dimension of citizenship in the NAV case, as it is often related to political participation and community engagement. Future case studies will cover the participation dimension.
Our findings indicate that DT, even if implemented out of good intentions, may result in negative outcomes for some citizens who encounter digital government. This means that digital policies that promote digital government actions can be argued to account for administrative evil in practice [11]. The excessive focus on the bureaucratic logic behind DT in the public sector, permeated by values such as efficiency, impersonality, professionalism, and technical rationality, may be a source of a moral inversion, where public and democratic values inherent to citizenship are overlapped by the precedency of system functionality, masked under the requirements of the job and good technique. While promoted for effectiveness, universality, and innovation in public governance, DT create new barriers for many citizens, hindering them from exercising their full citizenship.
DT require some preconditions from the citizens – such as literacy, assistance from intermediaries, system understanding, and access to digital ID. These are preconditions that need to be in place for citizens to interact online and benefit from DT. However, such preconditions are not granted for all or immediately. As bureaucrats force citizens to go online, this leaves no room to deal with non-digital stances.
In NAV, frontline workers with limited time ended up directing citizens to digital selfservices, without considering their competences and skill to use such services. Many citizens experienced stress, anxiety, frustration, and dismay when they were not able to take part or were not able to apply for services to which they were entitled. This made some citizens face challenges in actualizing their citizenship in the DT encounter. The lack of preconditions and resources to address exceptions to digital interactions manifests the evil dimensions in the DT adopted by NAV. The first factor (lack of preconditions) creates exclusion, and the second (not addressing exceptions) makes the excluded invisible or forgotten to the public apparatus.
We refrain from generalising such mechanisms to all DT used in the public sector to
foster citizenship. As amply reported by the literature [
We do not claim that the exclusion of citizens or damage to citizenship was intentional. Indeed, administrative evil is masked and often not visible to the perpetrators. It may be unintentional, but we posit it is purposeful, resulting in a causal mechanism activated by the two factors identified. We argue that critically discussing potential harmful traits of DT in administration is important, and we suggest further exploring how to unmask evil. We argue that such outcomes and traces of evil are important to unmask to safeguard citizenship rights and that our theoretical concepts introduced in this paper will provide value in doing so.
Our results account for some implications for practice and research, as well as we have limitations to acknowledge.
As a practical implication, we want to emphasise the importance of critically reflecting on the path that the digitalization of government embarks on. The logic that focuses on limited resources will inevitably lead to actions driven by the best value for money that may exclude the interest of citizens. When the public sector does not account for all citizens' interests and needs, it does not fulfil the government's responsibility. In public sector digitalization, it is important to put emphasis on the more marginalized citizens and develop strategies for their inclusion, as well as without DT. Digital government should support diversity rather than serve citizens as a homogenous group, be accountable to all citizens, and safeguard citizenship across different dimensions [4].
On the research side, our conceptual approach and illustrative case demonstrate the need to foster a digital government that mirrors citizens' needs in a digital analogic continuity to promote a government system that safeguards citizenship. Based on our findings, we argue for the relevance of discussing citizenship in a digitalised society instead of digital citizenship to avoid reproducing the duality of digital and analog citizens, which leads us to the inclusion-exclusion dichotomy. Furthermore, we argue for the importance of further exploring the needs and preconditions of all citizens to avoid consequences related to administrative evil. Our findings show that digital systems will reinforce administrative evil if this is not critically reflected.
Secondly, we identify a connection between DT use and negative outcomes for citizenship from an administrative evil perspective. We do not argue whether such a connection is transient, i.e., can be removed by DT or process redesign, or permanent, i.e., changing DT affordances will continuously recreate it. Future research should explore the role and characteristics of DT better and investigate how their action possibilities may avoid or reproduce conditions for administrative evil.
Finally, in this paper we advocate for causality in DT use and administrative evil outcomes towards digital citizenship. However, we limit our exploration to identifying a connection between two factors that manifest evil, exemplified by one case study. Furthermore, the case analysed does not allowed us to observe the participation dimension of citizenship. We will address this limitation in future research, in which we will progress our investigation from several case studies. We plan to do it by extending this study integrating the cases of M5S movement in Italy and the Desenrola Brazil case. In the M5S case a political party used DT to build a digital space for including and fostering citizen debate. In this case, the focus on political rights diverges from the NAV case as the M5S understand DT in relation to a democratic process with the aim of improving transparency and democracy. The Desenrola Brazil case is interesting as it provides a rational, simple, and agile ambience where selected groups of citizens could engage and negotiate their debts. By connecting with the private sector and government companies efficiently, they could benefit from huge discounts and be able to "clean their names" through accounts on the Gov.br platform. As we are familiar with these cases, we know that both cases exemplify different angles on how citizenship can be harmed and denied when DT is adopted.
This study was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior – Brasil (CAPES) – Finance Code 001. 17. 18.