From Experience to Text: Issues of Representation, Disclosure, and Understanding in Ethnographic Social Work Research

ABSTRACT

This paper starts with the ethical dilemma that appears when researchers end data collection, start to analyse the material, and construct the narrative of the thesis. This is the moment when the research subject might become objectified. Although this is a well-recognised dilemma, the aim of this paper is to investigate this in relation to ethnographic studies in social work in order to further develop some aspects of this ‘old’ dilemma. Three prominent texts on the topics of representation, disclosure, and understanding are used in an analysis of two ethnographic observations in different social work settings. A conclusion is that the main challenge for the ethnographic researcher in social work is the fact that they often write concurrently about people in unequal positions. Hierarchies and power relations have to be part of the question about how to write without objectifying. This fact places the researcher in a position of ambiguity because they act on both sides, including both those who are in privileged positions and those who are considered vulnerable groups. Hence, the ethical dilemma not only includes the relation between the researcher and the research subject, but also inter-party relations between research subjects in different power positions.

KEYWORDS:

• Ethnography
• social work
• representation
• disclosure
• research ethics

Introduction

There is something deeply troubling, as we all know, about joining with people, apparently in good faith, only later to turn your back on them so that yours becomes a study of them and they become a case. (Ingold 2017, 23)

The quote above pinpoints a classical research ethical dilemma that appears when researchers end data collection, start to analyse the material, write about the results, and construct the narrative of the thesis. This is the moment when the research subject might become objectified (Ingold 2017). It is also a moment that is difficult, if not even impossible, to anticipate from a research ethics perspective, for both the researcher and for the research subject. Before the start of a research project, several ethical considerations have to be taken. In this process informed consent is a core virtue for research that involves human subjects. According to international research, ethics and national legislation on informed consent mean that each research subject ought to get clear information about the project, purpose, methods, risks, and how personal data will be handled in order to decide if they will participate or not. The consent has to be required in advance. Furthermore, during the research project ethical dilemmas might arise and these should also, as far as possible, be anticipated in a risk analysis that precedes the project. Dingwall (2006; 2008) and Katz (2007) have fruitfully discussed the difficulties that qualitative researchers in human and social sciences might have with this kind of informed consent in comparison to researchers in medicine. The latter often know a lot about the intervention, treatment, or diagnoses they want to investigate. At least they know more than the possible research subjects. In human and social sciences it is often the other way around. The researcher approaches the research subject with wonder about a particular part of the research subject's life and circumstances, and the aim is to create understanding through interviews, observations, and other methods. Hence, the researcher knows much less about the topic of the research than the research subject, and expected results and the narrative of thesis are not possible to anticipate. Consequently, such obscurity might create insecurity among the research subjects already during the fieldwork. This has been discussed by Leigh et al. (2020) who shadowed social workers at two units for more than 15 months. The research team perceived that several of the social workers were not comfortable with having them around because they were not sure about what they were going to write about them.

In contrast to such discussions about ethical considerations before and during the fieldwork, we want to move forward to the moment when researcher writes about what they have understood about the research subject's life and circumstances. This is an act that involves delimitations and choices. One choice is to select what parts of the data should be presented, i.e. quotes from interviews, extracts from field notes, or something else. Another choice is to decide how to represent the research subjects. The researcher also has to consider what language to use, representing the vernacular of the research subjects or turning it into scientific terminology. Hardesty and Gunn (2020) provide us with an instructive example of such ethical considerations as they reflect upon how they used the term ‘survival sex’ instead of ‘prostitution’ as used by the research subjects. The change of terminology was possibly a violation of the research subjects’ authentic ‘voices’. It was simultaneously a result of the researchers’ analysis and their aim to conduct anti-oppressive social work research. This process, when experiences from interviews and observations are turned into text, therefore opens up far more and sometimes new ethical considerations that cannot be dealt with before or during the fieldwork (Murphy and Dingwall 2001). One main ethical dimension to handle is the power of the researcher, who is the one to decide who is represented and how. Other ethical questions are about how to avoid objectifying, how to let the researched persons ‘speak with their own voices’, and how to write with respect for the research subjects without failing to comply with scientific demands for accuracy and reliability. In human and social sciences there are several methods for dealing with this problem. One is the principle of mercy (Gilje and Grimen 1992). This principle asks for compassionate interpretations and that researchers should aim to present all research subjects as rational actors with agency. Hence, this principle should guide the way researchers present their analysis, results, and research subjects in their thesis and is one way to keep a critical distance to the field and still avoid the objectifying problem and the risk of reproducing simplified narratives about ‘good’ and ‘evil’. Another method is the principle of reflexivity, a principle that is intrinsically embedded in qualitative research (Davies 2008).

Although these are well-known ethical dilemmas, they might appear differently in different research areas and disciplines. Starting with experiences from our own fieldwork and text production, the aim of this paper is to dwell on some aspects of the ethical dilemmas and problems that might be significant for writing ethnography within the area of social work.

The main question is:

• How can we produce just and balanced research reports that represent plural and often concurrent perspectives?

This paper is based on an analysis of two ethnographic observations. These are part of larger studies that took place in different social work settings where policy rules and where professionals encounter individuals in precarious situations. The analysis is guided by concepts and theories presented in three texts written by Linda Alcoff (1991), Don Kulick (2015), and Iris Marion Young (1997).

Alcoff's text The problem of speaking for others (1991) provides us with theory and arguments about how to ethically deal with the power of the researcher and issues of representation. Kulick's 2015 text, which is a response to Alcoff, introduces and discusses more thoroughly the researcher's disclosure, i.e. the confession of the researcher's experiences and their relation to their research subjects as one way to balance the researcher's power and relation to the research subject in the final text. Young’s (1997) text provides us with tools for discussing the results of an ethnographic study in terms of understanding. She contributes with constructive arguments that in one way undermines the commonly stated goal of ethnographic research, namely to understand people's lives and circumstances, while in another way she offers new options that include listening with respect as a core ethical virtue.

In the following sections, we will first provide some more background about ethical considerations in ethnographic research in social work. Thereafter follows a short introduction to the texts by Alcoff (1991), Kulick (2015), and Young (1997). That section is followed by a description of the methods and the context of the two selected ethnographic observations. Then follows three sections with the analysis of the two observations. In the final section, some conclusions are drawn about how researchers in social work who use ethnographic methods can handle ethical dilemmas that arise when turning experiences from the field into text.

Background: ethnographic social work and ethical dilemmas

What, then, is significant for ethnographic social work research? And what ethical considerations are related to this kind of research? To begin, we can establish that there is a long relation between social work and anthropology/ethnography. Pfeilstetter (2017) discusses the common roots of the two disciplines and ends with the first urban studies and the famous Chicago school (cf. Shaw 2014). In early twentieth-century Chicago, several anthropologists for the first time in history used ethnographic research ‘at home’. At the same time as the Chicago school began investigating urban life at the margins, Jane Addams laid the ground for social work theory in the same city and with similar methodological approaches.

Pfeilstetter explains that although close in nature and ground breaking, the rise of both disciplines was embedded in a historical process of western ethnocentrism and administrative political control and are therefore considered as contributing to the oppression of ‘others’ (cf. Thomas 2009). Drawing on the historical background explained by Pfeilstetter, we can stipulate that the issue of representation could still be considered as a prominent problem within ethnographic social work research because the discipline often seeks to investigate and do research on vulnerable and subordinated groups in society (cf. Chambon 2013). Hence, the risk of ethnocentrism, oppression, and mis-representation is always present because the researcher in social work has a privileged position in relation to the research subjects. This is a relationship that also includes intersecting powered dimensions of, for example, gender, ethnicity, and race (Thomas 2009; Miller and Glassner 2011).

Furthermore, researchers in social work often include people in more privileged positions, e.g. social workers, managers, or politicians, in their studies. Therefore, vulnerable target groups in society, social workers, and welfare institutions are equally interesting as research subjects. This fact might add some aspects to the initial agony that comes with writing about others and the risk of objectifying the research subject described by Ingold (2017) in the introduction. What happens when the researcher writes about research subjects who have different power positions in the same texts?

Researchers in social work using ethnography have certainly dealt with this question through methodological discussions about reflexivity, power, and ethics. Mikkonen, Laitinen, and Hill (2017) and Warren (2020) write instructively about the ethical dilemmas that arise from the asymmetric and powered relation between the researcher and the research subjects. Their main focus is on people in vulnerable positions, i.e. women in communities in Nepal and older prisoners. In contrast, Blaisdell (2015) and Hickson (2016) discuss asymmetries that appear in fieldwork among professionals, i.e. preschool teachers and social workers. These discussions are, as stated above, motivated by their fieldwork, which often involves vulnerable people and marginalised life situations. This is put in contrast to the privileged position of the researcher and the power asymmetries that come with it. Furthermore, their research aims are to understand and in a long-term perspective improve these life conditions by enhancing anti-oppressive social work research (Hardesty and Gunn 2020). In line with our question above, Leigh et al. (2020) discuss the problem of doing field work in a context that involves both social workers (their main research subjects) and vulnerable families. However, they do not approach the problems of writing text and representation. Rather, they are occupied with the delicate dilemmas of gaining access to the field, blending in, and dealing with the researchers’ effects (ibid.). The way a researcher deals with these issues before and during fieldwork might have an impact on the final texts, but we argue that the process of reflexivity and considerations about power and ethics cannot stop there (cf. Hardesty and Gunn 2020). In this paper we will add to these previous discussions and share some experiences from the writing process and how we can continue to ethically explore the act of turning experiences from the field into text. By doing this we will hopefully tap in to the development of research ethical discussions. Reflexive and open-hearted ethical and methodological considerations about the arbitrary realities of ethnographic fieldwork are often left out due to space limitations, i.e. the format of a scientific article, and/or the dominating positivistic imperative model for research (Burke 2007; Leigh et al. 2020).

Three texts to guide the analysis

In order to answer the research question of this paper three texts providing us with concepts and theories have been selected. The selection was based on the author's previous long-time experience of work on research ethics related to ethnography in research, teaching, and practice.¹ The first text is Linda Alcoff's classical paper The problem of speaking for others (1991), where she discusses the problem of representation and power, primarily when people in privileged positions, represent, ‘give voice to’, or ‘speak for’ underprivileged groups. Alcoff (1991, 6) quotes Trinh T. Minah, who explains the problem by describing ethnographic research as: ‘mainly a conversation of “us” with “us” about “them”, of the white man with the white man about the primitive man … in which “them” is silenced’. The quote is a critique towards the reproduction of stereotypical and normative perceptions of different cultures and peoples as in the old school of imperialistic ethnographies (Ashcroft, Griffiths, and Tiffin 2013). The critique is still relevant and is one reason for writing the present paper. To avoid such imperialistic positions, Alcoff presents some interrogative practices that might inform scholars who speaks for or on behalf of others. One position is the ‘reduction response’ that starts from the claim that the speaker is right because they are located within the group they are speaking for and therefore know the truth about this group. Through the in-group position, the researcher becomes a legitimate speaker and not an oppressive outsider. However, Alcoff dismisses this response as ‘reductionist theories of justification and essentialist accounts’. Her point is that inside voices can be just as oppressive as outside voices. Another position is the ‘retreat response’, meaning that the researcher only speaks for themself, their own individual perceptions, and ‘truths’. Such a perspective might undermine reliability and validity and perhaps also trust and respect for the research (cf. Davies 2008).

In the second text, the anthropologist Don Kulick expands on Alcoff's arguments. He also relates to the reduction response when he introduces a critical discussion about disclosure in the paper The problem of speaking for others REDUX, (Kulick 2015). He argues that within his own research field of ability/disability it is almost a rule that the researcher discloses their relationship and engagement in the research field. Although critical of the in-group idea, he writes extensively about his own studies among disabled people and discloses that he has a disabled sister. In reading about his studies, it becomes obvious, though, that the main research subjects are rather the social workers and their shortcomings. Hence, he excludes them from his disclosure and does not recognise the social workers as equally important research subjects.

The third text by Iris Marion Young (1997), Asymmetrical reciprocity: on moral respect, wonder, and enlarged thought, was chosen with the purpose of looking beyond disclosure and different reductionist or retreat responses. Young starts with the philosopher Seyla Benhabib's ideas about communicative ethics and symmetrical relations between people. Young's idea about relations is that they are always asymmetrical and that it is impossible to understand or even imagine the situation and experiences of another person even if you ‘share’ a position, e.g. have experience with a disability or are a social worker oneself. Instead of striving for avoiding or dissolving asymmetries between the researcher and the research subject, we should recognise and even include them but still put effort into listening to others (rather than speaking for them). Importantly, in this paper we are focusing on power asymmetries between the researcher and the research subject on one hand and different powered positions between different groups and individual research subjects who participate in research on the other. The latter are predominately defined in terms of power relations between clients (vulnerable individual and groups) and social workers. By doing that we risk losing the focus on how complex dimensions of power include intersections of gender, ethnicity, race, ability, and so on (Thomas 2009). Therefore we urge the reader to also keep these critical aspects of power and oppression in mind while reading the paper because they are also part of the rationale of Alcoff's, Kulick's, and Young's texts as well as of our paper.

Method

Two ethnographic observations were selected for analysis in this paper. Besides forming the basis for the analysis, they will also give the reader a more concrete idea about the core of what we are discussing, namely the outcome when experiences from the field are turned into text. The first example is part of a research project called the Interpreter project conducted in 2008–2012 (Gustafsson, Norström, and Fioretos 2013).² Together with two co-researchers, Gustafsson conducted extensive ethnographic fieldwork including observations and interviews with interpreters, service users, and professionals in different public service institutions (Gustafsson, Norström, and Fioretos 2013). The fieldwork took the researchers inside otherwise closed rooms where encounters between staff working at various welfare institutions e.g. social services, health and medical care, police departments, migration agencies, and their clients take place. Gustafsson had no former experience of public service interpreting before the research project, except for the use of interpreting in interviews with non-Swedish-speaking people. Her colleagues and herself had found that there was very little research in the area of public service interpreting, although it seemed to be such an important tool for many professionals in their everyday work with clients. The research was driven by curiosity and the aim of creating a better understanding of the impact of interpreting on legal security and integration as well as improving interpreting services.

The second example is part of Nordstedt's dissertation. The purpose of Nordstedt's study was to investigate the social meaning of a needle exchange programme for individuals who inject drugs.³ The study took an everyday approach, and she used ethnographic methods. Nordstedt spent time at the clinic, and she followed participants in their everyday life, including injecting drugs, over a period of 16 months. One of the reasons why she took an interest in these programmes was an ambition to find out about how to best support and help the clients. Nordstedt is a social worker and has extensive experience working with such clients. Clients as well as several different kinds of professionals, including enrolled nurses, nurses, doctors, therapists, and social workers, were part of her study.

The intention in this paper is not to provide a complete description of the methods and research ethical dilemmas that appeared during these studies. Instead, we reflect upon the fact that both projects are examples of typical long-term ethnographic field work taking place in the middle of the ongoing everyday lives (including professional work and criminal acts) of the research subjects. Both studies included a broad spectrum of people, including both vulnerable individuals and groups and people in more privileged positions. Both are examples of field work in delicate environments, with Nordstedt's taking place in criminal situations and Gustafsson's taking place in settings protected by laws of secrecy. The conditions for informed consent differed from time to time and between settings. In the waiting room of the needle exchange where Nordstedt observed and made contact with research subjects, she used a sort of ‘opt out’ model for consent. Everybody who entered the waiting room was informed about the research that was going on through written information on a notice board. Nordstedt also gave verbal information to those she talked to, and with time visitors became familiar with her presence. If they did not want to be part of the research they had to tell her or had to leave. In the settings that Gustafsson observed, often in environments protected by laws of secrecy, another model was used. Gustafsson shadowed a public service interpreter for some weeks and went with him to various places where welfare professionals had encounters with non-Swedish-speaking clients. As in the observation presented in this paper, Gustafsson actually asked for consent on the threshold, only minutes before the meeting. If someone declined, she had to wait outside and could not fulfil the observation. However, none declined.

As discussed in the introduction, informed consent is a core virtue of research that includes human subjects. Furthermore, it has to be well grounded. Our two studies show how complex the procedure for consent might be in ethnographic social work research. In both cases we can see how the topic of this paper, the ethical dilemmas that arise when the researcher turns experiences into text, becomes delicate and difficult when consents are based on such circumstances as described above where most research subjects have little or no possibility to take part in the process of writing or in the long-term outcome of the analysis. Of course there is always the possibility to return to the waiting room or to the settings where interpreting took place, but there is no guarantee that the same people will be around. The end result – a thesis, report, or paper – is often completed years after finishing the field work. Still, the researcher has a duty to follow research ethics, and it is this duty and how it can be dealt with in the process of writing the thesis that will be discussed based on the two selected ethnographic observations.

Analysis and discussion

The problem of speaking for others in plural

In this ethnographic narrative, Gustafsson is observing a meeting arranged by the child welfare services in a small municipality. The meeting concerns the resettlement of a young boy who had been separated from his parents for some months under the custody of social authorities at the Swedish National Board of Institutional Care (SiS).

The interpreter had an assignment for three hours, and when he arrived at the social services office he was directed to a conference room at the third level in the building by the staff in the reception. When we entered the room, a lot of people were already gathered. The interpreter introduced me (Gustafsson) to the case worker, who was in charge of the meeting, and to the main client, a young boy and his mother. They accepted my attendance. I was briefly introduced to the reason for the meeting. The young boy had been forced into custody and placed outside his home for some months. Now he was going back to his home municipality, but not to his father and mother. The plan was to keep him in a transit home for some weeks, ‘for adjustment and observation’, before returning to his parents. Besides the case worker, the mother, the son, the interpreter, and myself, there were six more people in the room, including one man who worked at SiS where the boy had been in custody, one woman who worked at the transit home in the home municipality, two people who worked with the field team in the municipality, a friend to the mother, and the friend's baby daughter. The room was quite crowded. I was placed between the case worker and the interpreter, who had the mother and the son next to him on the other side. The case worker started the meeting and explained that they should go through the ‘care plan’ for the boy. After that, they should drive to the transit home so that the mother could meet staff and see the facilities.

That was the plan, but the mother immediately declared in Arabic translated by the interpreter: ‘Why are we here?’ The case worker repeated her initial words. The mother continued in Arabic: ‘But why are we here? Why is my son not coming home?’ The case worker explained that if she just could be patient, she would get the answers from the ‘care plan’. ‘Just listen’, the case worker said and glanced at her clock. The interpreter translated and the mother answered, this time she talked for a long time, only interrupted by the interpreter who translated her words. The case worker looked displeased, checked her clock again, and interrupted the interpreter. ‘Please, let's go through the care plan’, she said and started to read from her documents. Words like responsibility, identity, and life vision passed by and gave me the impression of a quite abstract and in many aspects intangible language that was used in the care plan for the boy. Could this be a reason why the mother did not understand? The mother stopped the case worker again and again with the same question and longer and longer sequences when she tried to express her frustration. The case worker checked the clock and interrupted the interpreter while translating the talk of the mother. The situation became awkward. Suddenly the man from SiS asked for attention. He said: ‘I will explain, can you interpret?’ The last part of the comment was directed towards the interpreter who already translated his question. The man started to explain, in a patient and clarifying mode, the whole process from the day the boy was forced into custody and placed at SiS until the situation now, when he was supposed to be resettled in his hometown. He explained why the boy had to be under observation at the transit home and that eventually the boy would come home. I am not sure, but the mother seemed content with this. The case worker was not. She was overruled.

When writing about this observation once more, it becomes obvious how the participants in the meeting have different positions when it comes to power and the possibility to decide the form and content of the encounter. The boy and the mother are the vulnerable ones, and in the long run it is their situation that the researcher (Gustafsson) is considerate about. But Gustafsson and her colleagues also write about other individuals in other positions at the same time. The interpreter is presented in the text, as well as the case worker and the man from SiS. Power relationships take place in the situation and change during the meeting. Most obviously, the case worker, who is in charge of the meeting, fails in her engagement with the service users, the mother and her son. She might seem unprofessional and lacking empathy in her repeated wish to follow the bureaucratic procedures. In the meantime, we can also see how she struggles to follow the procedure and not go beyond the scheduled time. And in the end she is disempowered and overruled, if only temporarily. There is obviously also a long pre-history, with repeated encounters between the case worker and the mother and her son, that might explain some of her impatience.

From the perspective of representation of the different research subjects in the ethnographic narrative, it is important to also take a look at the position of the researcher in this case. Gustafsson clearly comes from the outside. The only person she has a long-term relationship with is the interpreter, and just like him she becomes a witness in this particular situation, but again, what they are witnessing is a short chapter of something that has a long pre-history. Just like the interpreter, she can only translate the situation into a narrative and try to present a relevant interpretation of what happened. She can assume that the mother was aware of the power relations in advance because the mother had prepared by bringing her friend, who also brought her baby girl. Gustafsson can experience the situation when the man from SiS starts to speak as relief and support for the mother. In addition to the power asymmetries between the position of the client and the professional, other intersecting dimensions of power can be further analysed. This can be from a gender perspective, because the situation reproduced gendered power relations where a man takes the right to speak away from a woman, or by implying critical race theories or postcolonial studies because additional dimensions of oppression become tangible (Thomas 2009) Furthermore, dimensions of gender and ethnicity also add important aspects to the position of Gustafsson as an outsider and her position as a researcher (Miller and Glassner 2011). Later, while working with the analysis and writing of the text, Gustafsson and her research colleagues realised what a complicated situation it was to narrate about without taking a stand (Fioretos, Gustafsson, and Norström 2020). With the aim of following the principle of mercy (Gilje and Grimen 1992, 244), they tried to give a comprehensible description and to represent all who took part as rational individuals. In that sense, the researchers tried to speak for everybody and to understand them as rational and reasonable. One way to do that is to position all participants and describe how they are involved in different dimensions of oppression and power relations and how some of them have a dominant position and authority while others are silenced and oppressed. In addition, it is also crucial to analyse how these positions and power relations can change and be destabilised in the interaction between the participants in the meeting. Otherwise there is a risk that the narrative will be simplified based on the shortcomings of the case worker.

If we compare these thoughts to Alcoff’s (1991) text about the problem of representation, we find that her arguments are based on the dual relationship between a privileged researcher and exposed and vulnerable research subjects. Hence, her concerns raised about representation originate from this kind of asymmetrical relationship where the researcher has most (or even all) the power, especially in the act of writing about the other(s). This fact has on the one hand generated a lot of research that does not recognise the problem and therefore has ended in oppressive ethnocentric narratives where researchers, represent, ‘give voice to’, or ‘speak for’ underprivileged groups. On the other hand, these kinds of ethnocentric oppressive narratives have been extensively criticised, and Linda Alcoff's article The problem of speaking for others is one example.

In reading her text we can assume that the issue of representation and ‘speaking for the other’ is a common problem within social work research because the discipline often seeks to investigate and do research on vulnerable and subordinated groups in society (cf. Chambon 2013; Mikkonen, Laitinen, and Hill 2017; Hardesty and Gunn 2020; Warren 2020). As becomes obvious in the situation described above, social work studies often also include people in more privileged positions, e.g. social workers, managers, or politicians. Vulnerable target groups in society, social workers, and welfare institutions are equally interesting as research subjects. Our analysis of this complex web of different, often asymmetric, and sometimes changing positions might therefore add some aspects to the debate about ‘the problem of speaking for others’ because the researcher ‘speaks for’ and has to relate to multiple experiences and concurrent narratives of different individuals and groups that might be in opposition and/or various power relationships simultaneously.

The problem of the researcher's disclosure about a privileged position

In the ethnographic narrative below, Nordstedt makes observations in the waiting room at the needle exchange. She has made more than a hundred observations in this room, and the particular situation is not extraordinary or special, but is a rather ordinary situation at the clinic.

I [Nordstedt] have placed myself in a couch with my back towards the room where the staff is situated. There is a huge window, which gives the staff a complete view into the client waiting room. The social worker is placed along one of the walls. We are drinking coffee out of small white plastic cups and ‘chit-chat’ with each other because there are no clients present. The television is on, and we sometimes glance at a show about how to decorate the home. A man enters the waiting room carrying a plastic bag filled with groceries. He places himself beside me and starts a discussion. It is almost two months since we met last time. He lost his mobile phone when he had a relapse and has been without it ever since. He has not been able to visit the clinic because he has tried to stop using drugs. When we talk, he seems exhausted and he explains that he has been precluded from the housing that social services had organised. Therefore, he had nowhere to go the following night. In the plastic bag he keeps the groceries that he had bought on a requisition from social services. But without a stove to cook them or a fridge to keep them, they are useless. His situation is extremely hard. I talked to him and after a while I tried to involve the social worker, the one who has the assignment and knowledge about how to support the man. No conversation took place though, and neither the social worker nor the man showed any interest in talking to each other. The man starts to talk in short, single words, and the social worker answers my questions but does not ask anything himself. After a while, the man turns back to me and the social worker keeps watching the TV show. The situation was unresolved.

In this example, one research person is a social worker and the other is a client. Nordstedt gets involved with both of them in different ways, and the example describes a difficult situation for the researcher because none of the research persons act the way the researcher would have wished. The social worker seems uninterested in talking and listening to the client. The client is more interested in explaining his troubles to Nordstedt. Nordstedt experiences a double role and position because she is an independent researcher, but also a sort of colleague to the social worker (or might have been because they have the same profession). Most of all she perceives a friendly relation with the client in that he talks to her as a friend rather than as a researcher or social worker.

The observation can be analysed in many ways. Here we want to pinpoint the situation Nordstedt found herself in as the client obviously would rather explain his problem to her than to the social worker, while the social worker seemed totally uninterested in having a conversation with the client. It is not possible to fully explain the rationale of the social worker, but it demands some kind of effort in order to not highlight the shortcomings of the situation where the social worker takes no initiative and seems to take a greater interest in the TV show than in the miseries of the client.

Here we would like to turn back to the discussion about disclosure discussed in the text by Kulick (2015). Basically, Kulick is sceptical about the use of disclosure as a way to legitimize the ‘right’ of the researcher to speak for the research subjects. His arguments are in line with Alcoff's reduction response. Interesting, though, is the fact that he discusses disclosure only in relation to the vulnerable group, in his case people with disabilities, while his critical analysis is actually occupied with the role and actions of the social workers. Kulick does not write about it at all, but the way he argues he seems not to recognise the social workers as vulnerable research subjects subjected to the interpretations and power of the researcher. As in many cases in social work research, Nordstedt's relation to the field is that she is also a social worker (cf. Mikkonen, Laitinen, and Hill 2017; Leigh et al. 2020; Warren 2020). Hence, if we follow Kulick's discussion about disclosure, her interest in the topic emanates from that position rather than closeness to the experiences of the clients. She is obviously close to the field and part of an in-group, but the privileged group. In this context, could a disclosure about the relationship to the social worker support her analysis and give Nordstedt the legitimacy to speak for him? Perhaps, but it might also lead to a narrow interpretation that she can only ‘speak for’ the social worker, while she actually also aims to ‘speak for’ the clients.

We would like to add a reflection of ethical value about why a disclosure about the relationship to social work and the privileged position would be important for the analysis and representation in the text. Again we return to the principal of mercy (Gilje and Grimen 1992, 244). In relation to research in social work, we find Kulick's arguments about disclosure limited because he only includes some of the research subjects, those who are disabled, but not the social workers. One argument for why we should also disclose about the social workers is the above-described situation where the social workers in Kulick's study and in Nordstedt's and Gustafsson's observations often become the target of critical investigation. It is their shortcomings that are analysed. There is therefore a risk that the professionals in the field, the social workers, appear as the ‘evil ones’ while the vulnerable target group appears as the ‘good ones’. Despite the power and privileged position of the social worker, the risk is that this lack of engagement and disclosure about the social workers might strengthen and support already existing prejudices.

The risk of ethnocentrism and oppression that is part of the ethical dilemma of the representation of research subjects in reports or theses, and to which the whole discussion about disclosure and legitimacy is supposed to safeguard the researcher from, is therefore turned upside down. Neither Kulick nor Nordstedt or Gustafsson want to write in an oppressive manner about the research subjects, but there is a risk that this is not fulfilled if they fail to also recognise the vulnerability of those in privileged positions. This does not end or contradict the fact that social workers are in a dominant position in relation to their clients. Among other things, their vulnerability lies in the fact that they have authorised the researcher to include them as research subjects by giving their informed consent. Therefore, the researcher has the power to represent them in their writings. However, as argued in the introduction, when it comes to the outcome of the research, i.e. the written thesis, the research subjects, give their consent more or less ‘in blanco’. This argument leads to the conclusion that, for ethical reasons, a researcher needs to have a broad idea about representation that also includes privileged groups. In our cases, as often is the case in social work studies, the relationship between the researcher and the privileged groups is closer than it is to the vulnerable ones. Another question, therefore, is whether this means that the researcher who is a social worker understands the social workers better than the clients do. In the narrative above, it seems to be the other way around.

The problem of understanding others

As stated in the introduction, to reach understanding is often a common goal for ethnographic social work studies (Blaisdell 2015; Hickson 2016; Mikkonen, Laitinen, and Hill 2017; Leigh et al. 2020; Warren 2020). Kulick (2015) seeks to understand the situation of disabled people, and in the argumentation by Alcoff (1991) she discusses the role of understanding another person or group as a fundamental value when it comes to issues of representation. In the observations above, Nordstedt seeks to understand the everyday life of individuals who inject drugs and Gustafsson tries to understand the role of public service interpreting for legal security and integration.

To further develop the dilemma of turning experiences into text, we will introduce some arguments about understanding the other raised by Iris Marion Young (1997). Young states that the idea that one person can imagine and get into another person's situation is overestimated. She introduces the philosopher Seyla Benhabib's somewhat hopeful idea about communicative ethics and symmetrical relationships between people who recognise each other as equal in their encounters. Young states that this idea might work in theory if we imagine each other as ‘generalised others’ (the liberal idea of each person's equal value). In reality, she argues, all kinds of communicative relationships and encounters are marked by the background, history, and social position of those who meet. It is never equal ‘generalised others’ who meet, but ‘concrete others’ (cf. Benhabib 1992). Instead of trying to deny or overcome this fact by the reduction response or disclosure, Young states that the asymmetry of all relationships has to be recognised and dealt with. Otherwise there is a huge risk that unequal positions and prejudices about other people or groups, based on, for example, stereotypical and hierarchical perceptions about gender, ethnicity, and race will be reproduced (cf. Thomas 2009).

Young exemplifies this by referring to a telephone survey in Oregon in which a group of people were asked about quality of life among disabled people. They were also asked to imagine themselves as wheelchair bound, blind, or deaf. The result of the survey was that ‘the majority of respondents said that they would rather be dead than wheelchair bound or blind’ (Young 1997, 344). Young continues by discussing the fact that the rate of suicide actually is much lower among people with these kinds of disabilities than those who do not have them. The result of the survey therefore says more about the quality of life of those who answered than about the target group they were asked about. Young's conclusion is that whenever we try to understand and imagine the life situation of another person or group, there is a great risk that we reproduce our own perceptions rather than an understanding of other people's lives and situations. This creates a high risk of reproducing ethnocentric stereotypes and narratives that say more about the ethnographer than about those being studied (Davies 2008; Mikkonen, Laitinen, and Hill 2017). Young (1997) states that the best thing we can do if we seek understanding is to listen to the other rather than trying to imagine ourself in their position or taking their standpoint. Young (1997) therefore declares that it is neither possible, nor desirable, to entirely understand another person or group.

How can we take this argument under consideration if a central part of our research is to understand the life of other people and our research methods are supposed to give us access to the internal perspectives of those whom we study? If we look back on the two ethnographic narratives from the needle exchange and the interpreted encounter, we have to admit that we are writing about our own (the researchers’) experiences and understandings of the situations. This is indicated by the positioning of the observer in the texts. Both observations are written in the first person presenting us as eye witnesses but also as participants in the situations that we observe. We share our interpretations with the reader. The aim is not to write texts about the experience of the observer, however, but to present a more general analysis of the observed situation. Young introduces the ideal of ‘moral humility’, which means that the researcher embraces the idea that they understand things from the perspective of the other, but still expects to learn more about the life of the other by listening. The driving force should be to wonder rather than seeking to understand, and the narrative of the other should be considered a gift that the researcher has to take care of with respect.

Referring to the philosopher Hanna Arendt, Young calls this act of wonder, listening to, and being offered the gift of the other person's narrative a road to enlarged thought (1997, 358):

The concept of enlarged thought is supposed to explain how a person moves from a narrowly subjective self-regarding perspective on action to a more objective and socially inclusive view.

Enlarged thought is perhaps also what the researcher is actually looking for when they say that they seek understanding. Enlarged thought is perhaps also the result that the research can present in the act of turning experiences from the field into text.

Conclusion: the problem of writing about others

The discussions, arguments, and examples above leave us with several ethical and methodological considerations that we have to relate to during and after ethnographic fieldwork and in writing ethnography. We can conclude, though, that the discussions and advice we have found in the literature often derive from a dual relationship between the researcher, who has power and is privileged, and the research subject, who is subordinated and represents individuals or groups living in exposed and vulnerable situations (cf. Ingold 2017). This is also a common discussion in ethnographic social work research and is a driving force for researchers to also engage in methodological and ethical discussions (Mikkonen, Laitinen, and Hill 2017; Hardesty and Gunn 2020; Leigh et al. 2020; Warren 2020). Ethnographic studies as well as social work research often include such vulnerable research subjects, but at the same time other individuals and groups are also included. The most obvious are social workers, a group in power that normally has privileged positions. But other powered dimensions are also at play, for example, gender, ethnicity, and race, and these need to be recognised both in relation to the researcher and to the research subjects (Miller and Glassner 2011). Relations of power therefore appear in an intricate web of relationships between different research subjects and groups in the field, and such inter-relations have to be handled by the researcher, which might lead to some ethical dilemmas in the act of writing. Here we have paid extra attention to privileged research subjects (often social workers) and the risk pointed out in our examples is the reproduction of stereotypical interpretations of how social workers fail in their efforts to support their clients. We discussed this based on thoughts from Alcoff; about the possibility of ‘speaking for others’ in plural. In addition, based on Kulick's arguments, we considered the implications of disclosing one's background as a social worker.

Rather than thinking about representation and disclosure, the researcher has to realise that they cannot represent anything else other than what has been learned by listening to others. Hence, in that sense, the researcher speaks with their own voice and presents their own story (Young 1997; Ingold 2017). This standpoint is not the same as the retreat response described above where the researcher withdraws and only speaks for themself and their own individual experiences and ‘truths’. Instead, we would like to frame it as a process of making contact with different individuals and groups in the field and of receiving their confidence as a gift. The crucial thing is to manage this gift with respect. In both observations, unequal power relations take place and it becomes difficult to remain impartial and keep up the wonder in relation to all of the participants. Instead of becoming a critical judge, the researcher has to rethink their position and embrace the ambiguity of listening to others. Ambiguity is embedded in the act of doing field work in social work settings. The researcher acts on both or even many sides, including those who are in privileged positions and those who are considered vulnerable groups. This does not mean that the researcher should let go of their critical analysis and instead means that the same moral respect should be paid to all research subjects before, during, and after the fieldwork when experiences from the field are turned into text.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by Linnaeus University, Sweden.

Notes on contributors

Maria Nordstedt

Maria Nordstedt is a Ph D Student in Social Work at the Faculty of Social Sciences, at Linnaeus University, Sweden and her dissertation is about investigating the social meaning of needle exchanges for those who inject drugs. She has a Master's degree in Social Work from Malmö University, Sweden.

Kristina Gustafsson

Kristina Gustafsson is associate professor at the department of Social Work at the Faculty of Social Sciences, Linnaeus University in Sweden. Her main research interests include cultural encounters, democratic practices and postcolonial perspectives. Gustafsson is a member of the steering group of Linnaeus University research center, Concurrences in colonial and postcolonial studies.

Notes

1 Gustafsson is a member of and scientific secretary at the National Ethical Review Agency since 2014.

2 The study was approved by the Regional Ethical Review Board in Sweden (Dnr 117/2008).

3 The study was approved by the Regional Ethical Review Board in Sweden (Dnr 2018/225–31).