Conceptual underpinnings of translation in bureaucratic encounters

In effect, questions about bureaucratic information concern translation, i.e. the cognitive work of finding satisfactory answers to the question of “what does this mean?”. As the discussed strands of public administration literature also imply, this question is inevitably experienced as individualised and highly contextual: “what does this mean for me, in my circumstances?” The most common usage of translation signifies linguistic transfer, i.e. text in one language being reproduced in another. This understanding appears most frequently in social policy and social work literatures, e.g. the implications of migrants’ or asylum seekers’ verbal and/or written linguistic performance in the receiving country (e.g. Holzinger, Reference Holzinger2020; Ratzmann, Reference Ratzmann2021; Pötzsch, Reference Pötzsch2025). At the same time, critical social policy scholarship has relied on the sociological tradition of ‘translation’, especially Latour’s (2005) actor-network theory, to engage with, for instance, the ways in which social service reforms are carried out on the ground by frontline welfare workers (Clarke et al., Reference Clarke, Newman and Smith2007; Clarke et al., Reference Clarke, Bainton and Lendvai2015). This tradition is directly relevant to the negotiated meaning-making processes we find frontline welfare workers and private citizens engaging in, although empirical work in social policy implementation relying on Latourian foundations is rare (e.g. Zetterberg et al., Reference Zetterberg, Markström and Sjöström2016).

Translation occurs in relation to the means through which the social is instantiated (Clarke et al., Reference Clarke, Bainton and Lendvai2015: 36–38), inevitably linked to how power is exercised and becomes manifest, including through language via claims, statements, orders, and artefacts (Wæraas and Nielsen, Reference Wæraas and Nielsen2016). Translation is implicit negotiation of meaning in interactions, invariably meant to successfully deploying one’s own claims, statements, orders etc., but which is not necessarily accepted by interlocutors as such. Translation amounts to the transformation of the intents of one’s claims, statements, orders or, indeed, professional artefacts by those on the receiving end. We recognise this process in work concerning frontline workers’ claims of incomprehensibility of migrants’ German (Holzinger, Reference Holzinger2020; Ratzmann, Reference Ratzmann2021), while Hull (Reference Hull2012) relies on Latour’s work in discussing bureaucratic artefacts and how they are put to different uses than intended by citizens in bureaucratic contexts.

The concept of bureaucratic translation we propose highlights the unexpected meaning-making work that is imposed on parents by welfare workers, in their face-to-face encounters. Translation is productive and ends-directed empirical work. The concept is also meant to signify new parent’s contestation of the supposedly helpful ‘information’ the well-intended welfare worker is providing and its partial or total rejection.

Primary data collection

In Denmark fieldwork was carried out over twelve months in 2019 and in 2020, interrupted by the Covid-19 epidemic. Ten nurses were shadowed during their routine working weeks, each for two-week blocks. Following this, seven nurses were shadowed further over a six-month period, their activities encompassing: home visits, primary school clinics, daycare and school meetings, Overweight Clinics (Overvægtsklinikker), group meetings for vulnerable mothers, and evening courses for new parents (Forældreiværksætterne) (for a breakdown of the primary data, see Table 1 in the Supplementary materials). In Romania, fieldwork was conducted between December 2019 and March 2020, when Covid-19-induced lockdown commenced. Relying on an opt-in recruitment strategy at the welfare office of the two mayor’s offices, with the agreement of the public body and the relevant employees, new parents were invited to participate in the study upon arrival. If they agreed, the face-to-face encounter would be recorded and fieldwork notes added, and arrangements made for later visits, if relevant, and/or for an in-depth discussion about the entire claims-making process, usually in person at the participant’s convenience (Table 2 in the Supplementary materials). We were not mandated to undergo formal ethical clearance procedures. However, we observed the ethical guidelines commonly shared in qualitative empirical research during fieldwork, including informed consent, voluntary participation and the right to withdraw, the confidentiality of data and anonymisation, and the protection of participants from any harm.

Data analysis and stylistic choices in reporting findings

For analytical purposes, the data were treated as distinct studies. Processing and analysis were separate, using NVivo for the Danish component and MaxQDA® for the Romanian one. The project PI, not involved in data collection, participated in interpreting the Danish and Romanian findings.

Data analysis was informed by what Hennink et al. (Reference Hennink, Hutter and Bailey2011) call a mainly open – inductive – coding process. Concepts from the literature were applied alongside inductively developed categories based on participants’ experiences and narratives, with the latter forming large sections of both code systems. Coding unfolded in several stages. In-depth interview transcripts and notes were initially read through for the big picture and annotated accordingly. We flagged recurrent themes during this process. In the second stage, a random set of one third of transcripts was chosen to systematically develop the codebook, iteratively. The resulting codebooks were then used to code the remainder of the primary data. The codebooks evolved throughout this process, driven by: i/consistency in applying subcodes and structuring the code system, achieved through the regular ‘auditing’ (Seale, Reference Seale1999); ii/avoidance of over- and under-coding.

The different instances of meaning-making we conceptualise together as bureaucratic translation in this paper are the result of inductive analytical engagement since our study was not primarily focused on communicative practices or bureaucratic information. With both observation and in-depth interviews making evident the centrality, especially for private citizens, these two issues, we sought to conceptualise the informational dimension of encounters. The main rationale of the research project under whose auspices the fieldwork was carried out is discussed elsewhere (Kovács et al., Reference Kovács, Morris and Grauslund2025).

Moving ‘upwards’ in analysing the data, from comparing to defining categories in order to be able to then conceptualise social phenomena (Hennink et al., Reference Hennink, Hutter and Bailey2011), we arrived at five different meaning-making processes, all of which present themselves as instances of ‘bureaucratic translation’. Some are particular to the Romanian context, some to the Danish one, and some to both, in part due to the structural differences shaping the welfare worker-citizen interactions observed. For this reason we contend that our findings are the more valid since they were arrived at through distinct, mostly unrelated analytical processes, yet displaying internal validity and comparability independently of cross-national differences.

In reporting our findings, we rely sparingly on direct quotes from our participants. While common in reporting interpretive qualitative research, quotes do not demonstrate rigour (Corden and Sainsbury, Reference Corden and Sainsbury2006; Parkin and Kimergård, Reference Parkin and Kimergård2022). The second reason is pragmatic: for brevity, we summarise rather than show empirical material.

Seeking clarity and coherence

Citizens often sought clarity and coherence in information. Danish home visits entail the measurement of babies: weight, length, etc. Nurses record measurements in Novax, an electronic system used across Danish healthcare departments. The model displays a graph with the standard curve using the WHO Growth Model (2009). A baby’s specific curve plots against a standard, invariably resulting in deviation. Nurses and mothers often clarify the meanings of this mismatch: nurses often provide context for the reference curve during the first visit, e.g. that it seldom matches real curves; or they emphasise that baby’s curve is an ascending one, showing good development. Yet despite nurses’ insistence that things are as they should be, that the reference curve ‘should be taken with a grain of salt’, some information still remains implicit, e.g. that the reference curve is an international mean derived from breastfed children in ‘ideal’ circumstances (WHO, 2009). Thus, in the absence of detailed information on the methodology of the WHO model, some parents struggle with anxiety stemming from their child’s ‘abnormal’ development. This was vividly illustrated in multiple encounters where parents’ anxiety was not assuaged by nurses reassurance: ‘We don’t care about that, right?’ [about slower gain in length and weight of a baby]. ‘No we don’t, right?’ was the parent’s response, but assertive repetitions by the nurse were needed (who was at the same time filling in the chart showing a deviation from the norm). Translation as cognitively difficult, conflicting meaning-making for the parent was clearly difficult and not even here guaranteed. In short, adequate and sufficient information on how well baby is doing is not straightforward even when nurses readily show and explain the significance of the metrics they use. This is partly because the bureaucratic procedure of measurement has an implicit, symbolic load: it instantiates a clear, universal definition of ‘normal’ and, by implication, ‘abnormal’. The standard used by the professional creates work of clarification and of overcoming incoherence in what is being explained, advised, and assured for parents.

Romanian legislation contains models of standard forms for claiming social entitlements, to be submitted in paper format, dated and signed by the claimant, as well as the list of supporting documents for the application. This makes standard application forms and lists of supporting documentation an indispensable part of claims-making, but inaccessible since private individuals are rarely familiar with the relevant legal text. In the absence of a unified welfare state portal as in the Danish case, individual claimants need to actively seek out ‘the correct’ form and list of supporting paperwork. New parents implicitly trust Google’s list of hits when searching online, opening the sites in the order presented, with some prioritising the links ‘looking more official’. Parents have no choice but to operate with the uncertainty that the form and list of supporting documents might be incorrect or outdated or incomplete. Going in person to find out is, therefore, the safest solution: observation at the two mayor’s offices revealed that a sizeable share of citizens arriving are coming on an initial information-gathering visit, including the standard application form. Frontline workers know this and either make sure the applicant has the appropriate form or provide preprinted blank forms on the spot. Most new parents leave as they do not expect to submit their claim.

But face-to-face encounters are not necessarily expedient in terms of information sought. One of the frontline workers observed routinely indicates the standard list of supporting documentation for the universal child allowance, one recurring problem being her failure to clarify that ‘copies of birth certificates’ in her list refers to the birth certificates of all the children of the claimant, not just the one(s) being applied for. Some parents seek clarification on this plural, but most misunderstand, only to be confronted with the missing copies upon returning with what they expect is a complete application. Seeking clarity is more common in relation to the standard application form and supporting documentation for paid parental leave, where the form is not only convoluted, but requests information in ambiguous terms and/or using legalistic terminology, administrative information claimants rarely have, and information that appears irrelevant.

To sum up, seeking clarification, identifying missing or redundant information and putting one’s finger on incoherence in information provided are all higher-order cognitive processes resting on making new information one’s own – translation as productive meaning-making.

Sorting out contradictions

Notable difficulties and anxiety for beneficiaries stem from situations where the information or advice given is evidently contradictory. Danish parents are often told that infant care is partly a guessing game and there are as many right ways of caring for children as there are children. Nurse Ellen mused: ‘There are so many ways to do it all. You can raise a child in so many different ways.’ Despite this, nurses work by a predefined list of ‘phrases’ (fraser), a checklist of actions which the nurses go through in their examinations during visits in their assessment (see Appendix). The two positions – that ‘all children are different and accordingly need different treatment’ and ‘all children should fulfil the prescribed phrases’ – are understood as contradictory by parents. This creates an ambivalent space for parents: it demands negotiation and a decision as to how to position oneself and act – another form of translation required of parents. When the matter at hand is as high-stakes as infants’ feeding, cognitive development, or motor skills, parents want direct instructions rather than having to decide. Consequently, parents encouraged to follow their gut or pay attention to their own child while being presented with the necessity of reaching specific milestones felt insufficiently guided. Advice meant as encouragement to help parents become authoritative and self-sufficient carers was seen as ‘lazy’, as one frustrated mother put it. Some parents felt abandoned and thus anxious, instead of being helped, the nurses’ chief role.

Partially contradictory, incoherent information is also present in Romanian claims-making processes. This is a recurrent problem when initial information-seeking efforts prove useless on subsequent trips, when parents are asked for slightly different paperwork and the submission of their application depends on this. ‘This is why it is crucial that you deal with the one person’, one mother explained, referring to bureaucratic encounters more generally. Indeed, in exploring challenges around claims-making, information from multiple sources for the same entitlements or bureaucratic process was a recurrent theme. Needing some supporting documentation or not is understood as contradiction rather than inconsistency, and puzzled citizens often feel abandoned to resolve such contradictions unassisted when needing to make a decision. In rarer cases, frontline workers offer advice, but tentatively, and refer claimants to the higher authority, the ANPIS county office, for a reliable answer, but the cited legal provisions or advice can turn out to contradict what the local frontline worker stated. Romanian parents also criticise staff for not being helpful enough or not being adequately knowledgeable. Parents object not only to uncertainty about how to proceed and the related anxiety, but also to what these contradictions reveal about the (welfare) state and its agents more generally: ‘laziness’, incompetence, unacceptable discretionary interpretation of what should be universally applicable rules, and incoherence in how welfare programmes operate were mentioned as the sources of contradictions.

Corroborating and synthesising information from multiple sources

Insufficiently clear or ‘translated’ information necessitates sometimes active reliance on knowledge that private citizens already have about the matter at hand. Danish and Romanian parents, whether first-time parents or not, know certain things about children’s care and family benefits. Indeed, Danish parents are encouraged to rely on bookish, ‘scientific’ knowledge by the nurses. Thus, home visits and Romanian parents’ initial information-gathering trips ahead of applying for benefits requires that citizens identify what is distinct information from what they already have; what is useful, redundant, outdated, irrelevant; and that they boil it down to decide how to move forward. Danish parents discussing their child’s development were observed incorporating nurses’ advice into a body of knowledge that they have built from other sources: they showed recollection of earlier acquired knowledge and confronted new advice with older advice.

Similarly, Romanian parents were often observed mentioning knowledge of specific details, seeking confirmation from the bureaucrat or, on the contrary, expressing incongruent knowledge to corroborate and arrive at the ‘right’ information. This is because encounters with frontline welfare personnel inevitably require that new parents corroborate and synthesise different types or information. This is even more obvious for paid parental leave applications if, say, regulations have changed or parents find that they are an ‘atypical case’. Becoming aware of being ‘atypical’ often rests on corroboration and synthesis because this is not spelled out: parents glean it piecemeal based on bits of information they piece together – eventually. This matters because ‘atypical’ cases signify different or more paperwork and different steps for and duration of claims-making, i.e., additional administrative burden.

Uncertainty dominates the lives of new parents: welfare encounters add another dimension to uncertainty, in terms of information. Although the bureaucrat’s explanation and advice are accepted as the right one, some parents’ display of the corroboration work reveals the demands of translation. These processes are revelatory not just of meaning-making, but also negotiation of what is necessary, most useful, for the particular citizen making her claim.

Interpreting and couriering information as a third party

Yet another instance of translation some parents find they must engage in is when benefit receipt is predicated on collaboration between public bodies or claims-making necessitates paperwork from different institutions. Although in Denmark this is the case for some social programmes, the structural features of home-visiting nurses’ service provision makes this type of ‘translation’ absent in the Danish data. In the Romanian context, however, parents applying for universal child allowance for children born abroad, but now living in Romania are required to submit documentation from foreign public bodies demonstrating that the child is not in receipt of an equivalent benefit. Furthermore, the application for paid parental leave and related benefits includes supporting documentation issued by the employer and/or various other public bodies, e.g. the Unemployment Office, the Health Insurance Fund, the Tax Authority, etc. Citizens obtain such documentation to submit with the application even when the public body addressed has institutional access to – and even relies, in the form of background checks, on – the same bureaucratic information.

Not only are citizens made into couriers of official documentation which could be requested inter-institutionally on their behalf, but couriering inevitably rests on citizens interpreting for the two public bodies in a mediated communication process. Citizens address the second public body to obtain documentation or some procedure required by the initial one and explain what is needed. Inevitably, the production of this explanation rests on citizens translating what the requesting public body expressed. If the response is other than ‘yes’, the claimant needs to explain further, negotiating the obtainment of the requested documentation or procedure. If the response remains nay, the citizen returns to convey the refusal. Interpreting continues with the frontline worker of the requesting institution. If the response is positive, the new parent couriers back the positive outcome or news of the procedure completed, where the message might prove adequate or incomplete, again highlighting the ubiquity of translation burdening the citizen. Almost half of Romanian participants related experiences of being a ‘commuting citizen’, as a father formulated, interpreting and couriering information and/or documentation in interinstitutional disagreements and brokering solutions for oneself, reminiscent of emissaries’ work. Parents are highly critical of this necessity, seeing it as a structural problem, and expressed objections to being cast in the role of inter-institutional interpreter-courier-emissary.

Turning experience into useful knowledge

A final iteration of translation is turning first-hand experience into useful knowledge. Hull (Reference Hull2012) discusses how bureaucratic artefacts are strategically relied on by coalitions of seasoned citizens to successfully navigate bureaucratic processes, illustrating how experience can lead to useful knowledge. Skills-(deficit)-based concepts used to explain private citizens’ differentiated access to and take-up of entitlements also suggest that experience with bureaucratic processes equip private citizens with knowledge that can make their efforts expedient (Holzinger, Reference Holzinger2020; Döring, Reference Döring2021; Bisgaard, Reference Bisgaard2023; Tarshish and Holler, Reference Tarshish and Holler2023). Our findings unpack the complexities of deriving useful knowledge from experience, suggesting it is more demanding and riskier than suggested in the literature.

Earlier we explored the challenges of following nurses’ advice of following one’ gut feeling. Many parents, especially first-time parents, objected to this advice as it contradicted their ideas of universally ‘good’ and ‘bad’ caring. Following instinct posed another problem for many parents: being unable to draw on anything instinctive. They did not know what ‘to feel for’, as one mother expressed it, as this lay outside her prior experience. The intuition nurses encouraged is, unbeknownst to both nurses and parents, rooted in hands-on experience of reading baby cues: it is not innate. What nurses call instinct or intuition misrepresents what it, in fact, is: acquired knowledge gained with experience. This type of knowing requires a different kind of translation than the knowledge acquired from health professionals, books or other sources. Moreover, terming it instinct is a denial of the repeated translation of the precise meaning(s) of baby’s messages and of the learning process it necessitates. The meaning-making and negotiation process between nurses and especially new mothers concerning ‘intuition’ and reliance on it is, perhaps, the most illuminating case of bureaucratic translation work imposed by welfare state agents on users.

In the Romanian case, older parents explicitly highlight the usefulness of first-hand experience dealing with street-level bureaucracies generally. Comparing observations of different claims-making experiences, parents with university education are invariably more adept: they need less time to fill out forms and are less likely to make mistakes. They remember more details about the application package partly because they are less flustered and are able to retain more information, including in specialised terminology, more easily; and they typically arrive with supporting documentation to begin with, suggestive of experience and habit in engaging with Romanian bureaucratic bodies. In short, university-educated citizens navigate claims-making more easily because their professional lives form an experience-based pool of knowledge to draw on, something that some of these parents made explicit in interviews. This finding validates the skills(-deficit)-centred concepts of differentiated success in navigating welfare bureaucracies, for instance ‘legal literacy’ (Holzinger, Reference Holzinger2020), ‘administrative literacy’ (Döring, Reference Döring2021), or citizens’ ‘self-efficacy’ (Bisgaard, Reference Bisgaard2023), discussed earlier.

However, knowledge derived from hands-on experience may also hinder bureaucratic encounters. Some Romanian parents related experiences illustrating how familiarity with legal provisions and bureaucratic procedures create hurdles when cited as objections to what frontline workers advise or request. Earlier we detailed how Romanian citizens sometimes make evident their corroborating and synthesis while being advised. Formulating objections to the bureaucrat’s information, advice or request by referring to previously acquired knowledge can be judged as misplaced, unacceptable by frontline workers. The most dramatic bureaucratic encounters detailed by participants featured citizens objecting to frontline staff’s information, advice, or requests citing bureaucratic and legal knowledge derived from experience, with bureaucrats taking offense and turning unhelpful, even vindictive. Learning from experience may be a double-edge sword, suggesting that information acquisition – learning – in relation to the programme does not necessarily reduce ‘costs’ related to access and take-up, as the administrative burden literature suggests.

Deriving useful knowledge from first-hand experience is imposed on Danish and Romanian parents, but remains invisible and, for citizens, challenging. Knowledge from experience is recognised as a resource especially by Romanian parents, but its value materialises when knowing how to use it and when: if relied on to object to or challenge the bureaucrat’s information, advice, or request, chances are that being knowledgeable backfires. The matter of lesser difficulty in navigating bureaucratic processes among those with more formal education remains inconclusive: we cannot assert unequivocally that greater familiarity or training in cognitive tasks eases the process of translation universally. Rather it seems that educational differences bring with them different challenges of translating and producing usable knowledge.