Swazi co-researcher participants’ dynamic preferences and motivations for,representation with real names and (English-language) pseudonyms – an ethnography

Racial representation and pseudonym choice

In cross-cultural research selecting pseudonyms may be particularly complex, because naming participants is a (potentially harmful) form of racial labelling (Foster, 2006). In these contexts participants may be represented using Indigenous mother tongue or colonial second language names. The choice of language alternatively emphasises or masks participants’ racial or ethnic ‘otherness’ (Brunger and Russell, 2015). In post-colonial contexts individuals have often historically been forced to adopt colonial-language names. Not using Indigenous-language names might represent, or be interpreted as, a subtle form of racism and White supremacy (Foster, 2006). In addition, the assumption that pseudonyms should be used to protect privacy may be considered paternalistic, because colonized people marginalized due to race, have historically been exploited by White researchers. For example, during the colonial period research was used to steal Indigenous knowledge, justify dispossession and misrepresent cultural identities (Svalastog and Eriksson, 2010).

Pseudonym use, cultural identity and recognition

The expectation that pseudonyms will be used and the assumption that using them will always protect participants, is increasingly problematized. In post-colonial contexts where this history of exploitation continues to influence people’s lived experiences, participants may be more concerned about receiving recognition for the knowledge they contribute to research, than having their privacy protected (Svalastog and Eriksson, 2010). Using pseudonyms may undermine the human right to freedom of expression, which includes the right to impart information and ideas (Krog, 2011). This right is denied when a named author/s ‘writes up’ and attributes another’s ideas to a pseudonym because it shifts ownership of and recognition for the ideas to the author. The associated economic and other benefits of owning the information, whether gained through career advancement or commercialisation of Indigenous knowledge, also shift to the researcher-author (Svalastog and Eriksson, 2010).

Participants may perceive being named integral to having a ‘voice’ (McCormack et al., 2012) or conceive naming as a form of public acknowledgement associated with ‘justice, pride, and empowerment’ (Yanar et al., 2016: 125). Even if participants want their identity hidden, protecting participant privacy might be problematic. For example, it may conceal problems people experience due to marginalization and in turn protect the dominant group from confronting these problems (Baez, 2002). Protecting participant privacy may also reduce research quality, if it involves abstracting data from its socio-historic context (Svalastog and Eriksson, 2010). The notion that researchers can ‘protect’ participants by keeping their identity a secret is particularly likely to be problematic in participatory action research (PAR), a paradigm for conducting scientific enquiry in equitable partnership with members of a marginalized community (Yanar et al., 2016). In this context, there are potential tensions between the ethical principle of protecting privacy, and PAR principles like equitable recognition of community co-researchers, and their ownership of the research project.

In recognition of these ethical complexities Svalastog and Eriksson (2010: 104) have suggested a need for research participants to engage in ‘careful deliberations’ about the use and choice of pseudonyms. Documenting and sharing how such deliberations unfold in ‘real world’ research projects is a potentially important strategy for developing evidence regarding: participants’ perspectives on, and how researchers can facilitate careful deliberations about, pseudonym use. My aim is to share my experiences deliberating pseudonym use with research participants, in order to enhance understanding of: (1) the motivations which underlie academics’ and participants’ pseudonym-related preferences; and (2) how researchers might ethically facilitate careful deliberations about pseudonym use in qualitative research.

Methodology

I do this through a theoretically-informed, critically-reflective analysis of ethnographic data, detailing a PAR process in Swaziland. The ethnography was one component of a multiple methods research project, which was approved by the Monash University Human Research Ethics Committee (CF12/2637 – 2012001422) and the Swaziland Scientific and Ethics Committee (MH599C). The project ethnography focused on participation and empowerment in PAR about health in a rural Swazi community caring for children affected by AIDS. The model of care for children affected by AIDS in sub-Saharan Africa relies heavily on communities ‘helping themselves’. Studies of community-based care are well suited to PAR because this methodology aims to create locally actionable evidence and build on existing community assets. As a privileged White woman, born in a high income country (Australia), I further considered the PAR methodology suitable for conducting research in Swaziland because of its focus on empowering marginalised community members, and developing culturally sensitive research procedures (Israel et al., 2013).

Country context

The Kingdom of Swaziland, a former British ‘protectorate’ in southern Africa, is now the continent’s last absolute monarchy (Sihlongonyane, 2003). The Swazi monarchy’s strategic decision to peacefully collaborate with British missionaries and colonial officers was integral to them maintaining uninterrupted rule throughout the colonial period (Matsebula, 1972). Although Swaziland peacefully gained independence from the British in 1968, peaceful collaboration with foreigners continues to be upheld as a virtue of Swazi culture. It is best conceptualized as ‘a virtue made of necessity’ (Bourdieu, 1984: 372). The only other option for Indigenous groups like the Swazis was engaging in colonial war on highly unequal terms. Peaceful collaboration with colonizers represented ‘resignation to the inevitable’ (Bourdieu, 1984: 372).

Swaziland’s survival as an independent nation in the post-colonial era of neo-liberal economic globalization is highly dependent on international collaboration. The British-colonial legacy continues. English remains an official language of Swaziland, and the medium of instruction in the nation’s secondary and tertiary institutions. It is common for Swazis to have multiple names, including an English name that they use in formal interactions, especially at school and work.

Research setting

My project ethnography was conducted in a rural subsistence farming community, isolated by mountainous terrain and marginalised due to poverty and lack of basic infrastructure (Brear et al., 2017b). I selected the community purposively because I had been participating in development activities in it for over five years when the study commenced. Specifically I had participated in establishing a preschool (in 2007) and ‘neighbourhood care point’ (soup kitchen) (in 2012). In the first year of the preschool’s operation, I assisted a local teacher in the classroom, and lived in the community. Thereafter I continued to participate in these projects as a committee member, fundraising coordinator and regular visitor, while residing in the nearest major town.

The preschool was the partner organisation for the PAR. The preschool classroom was the setting where the co-researchers and I participated in all aspects (as defined by Brear et al., 2017a) of the PAR process. This involved 96 participatory learning and reflection workshops in which we co-developed the methodology, methods, tools and ethical procedures for the PAR. In 23 workshops the co-researchers analysed transcripts of focus group discussions I collected as part of the ethnography (i.e. their own data) to learn about qualitative thematic analysis. They also participated in 36 days of fieldwork and structured opportunities to reflect on our experiences designing and implementing the PAR. I documented these interactions through the ethnography.

Methods

I collected the ethnographic data using unstructured participant observation and focus group discussions. I developed the ethnographic methods independently of the participants, however I purposely shifted my enquiries to respond to issues, which were revealed to be relevant to my broad focus on participation and empowerment, during the iterative PAR process. Investigating pseudonyms use and choice was not an initial aim of the study. Rather, its importance was revealed by spontaneous contributions from the co-researchers, and feedback from academic colleagues.

In response I incorporated specific enquiries, and provided opportunities for the co-researchers to engage in ‘careful deliberations’ about, pseudonym use and choice during eight participatory ‘member checking’ activities (Table 1). Member checking, sometimes referred to as participant validation, is a strategy to increase rigour in qualitative research (Miles et al., 2014). It typically involves researchers providing participants opportunities to read, comment on the accuracy of, and if desired expand, the data they provided (for example, in an interview). Less commonly it involves providing lay summaries of, and opportunities for participants to comment on, data analysis. I used the latter approach.

OPEN IN VIEWER

Table 1.

Summary of participatory activities in which the co-researchers deliberated pseudonym use.

Activity	Date	Description	Number of Participants
1. Focus group discussion1	25/6/2013	Participant discussion about how they felt being written about as research participants.	8
2. Group discussion2	19/7/2013	Discussion about ethical considerations surrounding identifying the community inadvertently by using co-researchers’ real names and how to obtain community-level consent.	6
3. Agree and disagree activity (adapted from (International AIDS Alliance, 2006)	27/9/2016	Co-researcher participants indicated whether they agreed or disagreed with the statement, ‘Using a pseudonym to write about me in research reports is disempowering’, and to share their reasons.	5
4. Small group presentations	27/9/2016	Two groups of co-researchers discussed, prepared posters and presented their answers to: • Why did you choose an English pseudonym? • How would you feel about using a siSwati pseudonym?	5
5. Answer questions in writing	27/11/2016	Individual co-researchers prepared written answers to the following questions: • What do you remember Michelle telling you about the pseudonyms when you started the research? • What contributed to your decision to choose the pseudonym you chose? • What does this name mean to you? • What were you thinking the name would mean to other people who read it in a research report?	7
6. Small group presentations	6/12/2016	Participants further discussed their answers to activity 5 questions in small groups, and prepared posters and presentations of their answers, which were discussed. I explicitly asked if they remembered me telling them: (1) they could choose their own pseudonym; (2) that I thought using English-language names might be better in English-language reports.	4
7. Agreeing and disagreeing activity	6/12/2016	The co-researchers discussed if they agreed or disagreed with the following statements and shared their reasons why: • I changed my mind about using my real name because other researchers said they wanted to use a pseudonym. • I changed my mind about wanting to use my real name because I realised the data Michelle was collecting was sensitive. • It is more important to use a pseudonym in reports that people in the community will read, compared to reports that people in Australia will read.	4
8. Small group presentations	6/12/2013	The co-researchers imagined siSwati was an international language. They then discussed, prepared posters about and presented their answers to the following questions: • What pseudonym would you choose? • How would you feel about being written about in siSwati? • Having thought about that, what does using an English pseudonym in the research that Michelle is writing about in English mean?	4
9. Small group presentations	10/1/2017	The co-researchers read and discussed a plain language summary of this article. They then prepared posters and presented answers to the following questions:	5
1. The focus group discussion was audio-recorded. All other activities were documents through participant observation. 2. Groups discussions involved all participants and I and were conducted in English. Small group discussions were conducted in siSwati and the co-researchers worked independently. The posters and presentations resulting from these were prepared and made in English.

International AIDS Alliance (2006). Tools together now! 100 Participatory tools to mobilise communities for HIV/AIDS.

Participants

Ten community members participated as co-researchers in the PAR and research participants in the ethnography. I selected the co-researcher participants in partnership with three female community members who worked for the community preschool. They were selected purposively for age and sex diversity, from amongst a group who responded to advertisements and announcements made in the community. After receiving information about the study all potential participants provided written informed consent to participate. The co-researchers participated for a median of 14 (range 2–14) months in the PAR process. They were compensated for the time and knowledge they contributed, in accordance with regional ethics guidelines (NHREC, 2012).

Data collection and preparation

Data were collected in 76 participatory, research learning and reflection, and 6 member checking workshops in which the co-researchers and I participated. I played the role of facilitator and collected participant observation notes conspicuously (i.e. I recorded notes in front of them, as we worked together). I expanded and transcribed my hand written notes within 24 hours. I conducted 22 focus group discussions (mean 62, range 48–72 mins) periodically during the workshops. These were audio recorded and transcribed. All data were collected in English, the co-researchers’ second language.

All but one of the co-researchers participated in at least some of these participatory member checking activities (Table 1) and three (all women) participated in all the member checking activities. Each co-researcher selected the pseudonym which I use to refer to them in this article. Three chose English-language first names with biblical connotations (Mary, Simon and Samuel). Five chose English-language first names with no obvious biblical connotations (Adrianna, Sophie, Petunia, Lucy and Jackie). One provided a first, middle and surname (Colin Trevor Dutoit). The remaining co-researcher chose an English-language word I had not previously heard used as a name (Pretty).

Data analysis

I utilized QSR NVivo (QSR International, Victoria, Australia) to manage the data analysis process. Primary analysis of the entire ethnographic dataset involved generating and coding data to broad, recurring themes that were salient to the overarching research aim of understanding participation and empowerment in the PAR process. The use of pseudonyms was one such theme. This article details the sequential analysis (further analysis using a different analytical technique, to enable deeper insights about a subject) (Simons et al., 2008) of the pseudonyms dataset, using an iterative, abductive (inductive/deductive) (Onwuegbuzie and Combs, 2010) approach. This involved shifting between generating analytical ideas and interpretations from (1) the data and my experiences (inductive); and (2) academic colleagues and literature (deductive).

Dynamic perspectives on pseudonym use

Upon commencing the project ethnography I assumed that protecting participants’ identities by using pseudonyms was appropriate. I informed the co-researchers, in written explanatory statements and verbal discussions during the informed consent process, that I would use pseudonyms when collecting participant observation notes and writing research reports. Before and on several occasion during the research, I mentioned that I would use English-language pseudonyms I had selected for data collection purposes, but invited each co-researcher to choose a preferred pseudonym for research reports. All the co-researchers consented to participate on these terms and none provided me a preferred pseudonym. However, only one of the co-researchers who completed activity 6 (Table 1) recalled me inviting them to choose their own pseudonym for report writing.

The co-researchers spontaneously initiated our first deliberation about pseudonyms, approximately six months after the project commenced, in a member checking workshop (Activity 1, Table 1). At the time we were still co-designing the PAR. I facilitated the workshop to obtain feedback on analysis I was presenting in a conference abstract. I had not intended, or imagined a need, to discuss pseudonyms. Instead I simply informed the co-researchers of my intention to use siSwati-language pseudonyms, as the English-language presentation would be made in neighbouring South Africa, where siSwati is an official language. All the co-researchers who participated in activities 1 and 6 (Table 1) recalled me giving them this information. I also mentioned that using English-language pseudonyms might work better in other English-language publications intended for international audiences, however none of the co-researchers recalled this.

I was (pleasantly) surprised when the co-researchers spontaneously expressed a desire to use their real names. They said:

Pleased that the co-researchers wanted to use their own names and interpreting it as an expression of ownership of the research, I assumed that naming the co-researchers’ was ethical because it was their stated desire. However, as an academic researcher I was obliged to obtain permission from ethics committees before doing so. Discussions with academic colleagues about obtaining ethical clearance, prompted me to reflect on the ethical complexities surrounding the use of real names.

In particular these discussions enlightened me to the possibility that naming the researchers might inadvertently identify the community that was being studied in the PAR, and the need to obtain community-level consent before doing that. Assuming (perhaps paternalistically) that the community was unable to make an informed decision about being identified until they had actually participated in, and seen the results of the PAR, I requested the co-researchers wait to use their real names. They reluctantly accepted.

Problematising the use of English-language pseudonyms

Having invited the co-researchers to provide preferred pseudonyms and not received a response, I assumed they were satisfied with me using the English-language names I had selected for them. I proceeded to draft my PhD thesis using these pseudonyms. However, feedback from several White academic colleagues suggested that using English-language names to represent the Swazi co-researchers was inappropriate and might be interpreted as racist. I then explicitly requested the co-researchers select their own pseudonyms rather than simply telling them they could do so if they wanted to. I used email to ask one of the co-researchers (Sophie) to coordinate this process, as at the time I was residing in Australia.

None of the co-researchers chose the pseudonym I had selected for them. All chose English-language pseudonyms, with the exception of Pretty, who initially chose a siZulu-language name (a related but distinct southern African language) ‘Zondwako’, meaning ‘one that everyone hates’. Sophie independently denied Pretty permission to use their chosen siZulu name. Sophie only advised me they had done that, after I requested Pretty select a third preferred pseudonym, concerned about the interpretations readers might make of their second choice ‘Bin Laden’. Pretty complied and finally chose her English-language pseudonym.

After completing my thesis I returned to Swaziland, intending to work with the co-researchers to develop a strategy for obtaining community and ethics committee approval to use their real names in research publications. However all six co-researchers who participated in activity 3 (Table 1) reported they would now prefer me to use their self-selected pseudonyms. This group included four co-researchers who had previously stated explicitly that they wanted me to use their real names. I shifted my enquiries to understanding what motivated the co-researchers to change their preferences, and their reasons for choosing English-language pseudonyms.

Reasons for pseudonym-related preferences

I was concerned that my initial (subtly racist) choice of English-language pseudonyms had influenced the co-researchers’ preferences. Indeed, when I asked Sophie by email why they had chosen an English name, one of five reasons they offered was, ‘the pseudonyms that you have used for us before were in English . . . so why this time give myself a siSwati name?’ However when I asked (in Activity 7, Table 1) other co-researchers if I had influenced their choice, they rejected the suggestion. Lucy stated explicitly that the idea to use English-language pseudonyms had come from the co-researcher group, not me. They also rejected the suggestion that they had changed their mind simply because other co-researchers decided to use pseudonyms. Instead they reported their changing preferences were motivated by reflection about the potential consequences of using real names, particularly after becoming aware of how sensitive my reporting of the data could be. For example, Sophie said she started wanting to use a pseudonym when reading about something I reported she did in a chapter of my thesis. The co-researchers also reported thinking that readers (particularly other Swazis) would be more enthusiastic about reading the report if English-language names were used. According to Petunia, this was because Swazis look down upon each other. The co-researchers also believed the English-language pseudonyms would represent them as people who were ‘smart and clever’. Finally they mentioned wanting to protect their own privacy and the identity of the community.

I facilitated two activities (4 and 8, Table 1) to investigate how the co-researchers would feel about using siSwati pseudonyms. All the co-researchers indicated they would not like to use siSwati pseudonyms. They reasoned that: (1) English-language names were consistent with the (‘international’) language in which I was writing; (2) most Swazis had English names that they used in formal contexts, especially in writing (although only two of ten co-researchers used their English names in our interactions in the PAR); (3) Swazis were proud of their English names; and (4) they disliked hearing their siSwati names pronounced incorrectly. In their presentations the co-researchers parodied White people trying to pronounce the ‘click’ sounds in their real siSwati names. They also acted out Swazi children brimming with pride when given (their typically ingratiating, as in the pseudonym Pretty) English-language names on their first school day.

The co-researchers provided diverse reasons for choosing their particular English-language pseudonyms. Some said the names just sounded nice. Others reported their pseudonyms were the names of people who were role models for them. Several indicated they had chosen their pseudonym simply for the sake of giving me a name to use. One pair of co-researchers reported they had chosen gender-specific names so that people would know they were females. They said that they did not think their gender would be obvious to people who could not speak their language, if they chose siSwati names. None explicitly indicated that their chosen name ‘meant’ anything about their race or ethnicity. Their comments indicated that the nature of racial labelling through use of English-language names was complex. For example, Sophie told me her namesake, ‘was one of my favourite [Black South African] actresses’. However, the co-researchers’ comments about how Swazis liked White people and English-language names suggested their choices might have been intended to mask their racial identity as Black Africans.

Subjective motivations for pseudonym-related preferences

I now turn to deciphering the co-researchers motivations for pseudonym-related preferences, informed by the Bourdieusian theoretical framework. The co-researchers’ dynamic preferences regarding pseudonyms and the complex and diverse reasons they provided for their choices, suggest complex motivations. Their reasoning suggests their preferences were influenced by their recent direct experiences of being research participants, particularly seeing how I interpreted and wrote about implicit meanings in data they provided. However my interpretive analysis suggests that the co-researchers’ pseudonym related preferences were also motivated by longstanding tacit dispositions (i.e. habitus) characterised by ‘internalized oppression’ (i.e. a belief in their own racial inferiority) (Chavez et al., 2011). These predisposed them to make a virtue of English-language names. The results further suggest that I reinforced this disposition by suggesting, even if they did not recall it, using English-language names.

My analysis suggests that the co-researchers’ lived experiences as Swazis, an ethnic group oppressed by their history of British colonization, markedly influenced their preferences. Specifically it indicates that their decision to use English-language pseudonyms was partly motivated by a disposition to ‘peacefully collaborate’. I had suggested I would only be ethically able to use their real names if all the co-researchers and the community agreed. They perhaps (tacitly, even though they reflexively rejected my suggestion) did not want to disagree with each other, or create unforeseen negative consequences for their community. It is conceivable that, knowing obtaining ethics approval to use real names would be challenging and time-consuming for me, the co-researchers simply did what they thought I wanted them to, ‘choose’ to use pseudonyms.

Virtue and necessity

The interpretation that the co-researchers’ dispositions were constituted by internalised oppression is supported by the ‘virtue’ they made of the ‘necessity’ of using English-language names. The results would not be published in their mother tongue, a ‘necessity’ created by my academic careers goals, in a post-colonial reality where English was the international language). It was further necessitated by my inability to speak or write fluently in siSwati, a language spoken by just a few million people globally. The co-researchers simultaneous excitement and pragmatism about choosing English-language names is shown through this interpretation to be a ‘virtue made of necessity’. Their ‘choices’ were adaptations made to cope with an oppressive social structure they did not have the power to change. They had accepted and saw as inevitable, a practice (communicating in English) born of necessity. Adopting English-language names was a necessary step in accessing economic, social and cultural capital in the colonial period. It remains a necessity in a world where the history of colonialism has produced one (or several) international language(s), rather than international bi- or multi-lingualism. Given the international nature of the English language, my academic colleagues’ assumption that using Indigenous-language pseudonyms for Black people is appropriate, must be interrogated.

I could have chosen to make a ‘virtue’ of using siSwati-language pseudonyms to give my English-language research reports a more authentic ‘Swazi’ feel. Doing so would have been based on the incorrect assumption that the co-researchers preferred to be represented with a name that identified them as a member of a specific cultural group (rather than for example by a name that indicated their gender). It would have made this ‘virtue’ of the ‘necessity’ created by the academic expectation of ‘cultural-sensitivity’ in cross-cultural research, and the related assumption that it is possible for White academics to make their research about Black participants culturally sensitive. However it would have been a superficial strategy that glossed over the culturally insensitive system through which academic research reproduces privilege. Academic institutions distribute cultural capital. It is primarily accessible to the White people who have historically had, and are therefore predisposed to, access cultural capital in the form of university degrees (Bourdieu, 1986).

Unmasking racism and White privilege

My analysis supports my academic colleagues’ suggestions that my initial approach to choosing pseudonyms was subtly racist. However it indicates that the racism was enacted in my choosing, rather than using, English-language names. It further suggests that selecting myself, or restricting the co-researchers’ choice to siSwati-language pseudonyms, would have been equally problematic. Had I done so, I may have remained oblivious to the subtly racist, tacit disposition that motivated me to choose and use English-language pseudonyms in the first place. I would not have needed to confront my own racist assumptions, nor have been motivated to discuss them with the co-researchers. In the absence of deliberations about pseudonyms, the subtle ways in which I enacted racism would not have been revealed. The internalised oppression exposed in the co-researchers’ desire to be represented with ‘international’ colonial-language names, would likely have remained hidden.

The transformative potential of deliberating pseudonyms would have been missed had it been glossed over by using siSwati pseudonyms. The process was confronting and transformative for me because it required I reflect on my own racial privilege. This privilege was fundamental to me having the opportunity to represent a group of marginalized Black co-researchers by writing about them. The process was confronting because it challenged my assumption, informed by PAR principles, that it was possible to establish an equitable partnership with a group of marginalized co-researchers. The deliberations required me to acknowledge that despite my good intentions, I had put my ‘need’ to expand my own cultural capital by attaining the degree of Doctor of Philosophy, ahead of the co-researchers desire to use their real names, when I asked them to wait. Waiting to name the co-researchers enabled them to develop a better understanding of the potential consequences of revealing their identities, and ultimately led to them changing their minds. However, my reasons for asking them to wait were all related to my need to finish my degree and fulfil my academics requirements. Interrogating my research practice revealed the subtly racist assumptions I brought to the study and the ways these reproduced my power and privilege.

The process of deliberating pseudonym-related choices was also potentially transformative for the co-researchers. It required them to question assumptions about the superiority of White people, that they had ‘taken for granted’. It provided opportunities for me to name my own racism, a process which the researchers probably found confronting because they liked me and did not want to think of me, someone they perceived had ‘empowered’ them (as detailed in Brear, 2017) as a racist. Although they disagreed with my analysis of my own racism when I suggested it to them, having the direct experience of hearing it named, reconstituted their habitus. It is this habitus, now constituted by the experience of hearing a White person they liked, acknowledge their own privilege and racism, that will structure their future thoughts and actions (in more or less subtle ways, depending for example if it is a topic we continue to discuss). It will potentially make them more tacitly aware of the oppressive race relations which structure their society and their internalized feelings of inferiority.

The results highlight the cultural insensitivity and subtle racism inherent to the process of White researchers representing non-White participants in research that they control, and reports they write in ‘international’ colonial-languages. The implication is that these colonizing effects are present, regardless of pseudonym language choice. ‘Cross-cultural’ research may be a necessity of our socio-historic context. Certainly we cannot undo colonialism. However the reproduction of privilege and marginalization cross-cultural research involves must be acknowledged. My analysis indicates that such acknowledgment, although confronting, has transformative potential because of the new necessities and possibilities it creates. Acknowledging masked forms of racism exposes the way it perpetuates marginalization of non-White people and White privilege (Baez, 2002). It makes confronting subtle forms of racism necessary and changing them possible. On the contrary while ever the subtly racist assumptions White researchers bring to cross-cultural research remain hidden, they will be reproduced, tacitly, and perpetuate historic forms of racial oppression.

Strengths and limitations

The results of my Bourdieusian analysis provide novel insights into the complexities of using and choosing pseudonyms. However the implications of this must be drawn out with careful consideration of the study’s strengths and limitations. The multiple method study was a partnership approach which incorporated relatively extensive interactions and member checking processes (compared to most research). It is becoming more common for qualitative researchers to provide participants opportunities to read their own interview transcripts and/or lay-summaries of the findings. However, I am not aware of another study which has provided structured opportunities for participants to critically reflect on results in participatory workshops. Although this strategy cannot eliminate my privilege or the co-researchers’ marginalization, I believe our close and continuing critically reflective interactions greatly strengthened this research process.

Combining data documenting both the tacit motivations that structured the co-researchers’ practice, and the reflexive thought processes I stimulated through member checking, also enhanced the dependability of my results. However, some of my interpretations were rejected by the co-researchers. According to the Bourdieusian theoretical framework I utilized this is unsurprising because of the different assumptions and dispositions we brought to the research. It does not negate the dependability of the results (e.g. I can still be racist even if the co-researchers did not perceive it). It does however necessitate the results I have presented being interpreted as one possible, theoretically-informed, version of the ‘truth’. The results are the truth as I see it, from my standpoint as a White academic with an interest in interrogating my practice and acknowledging my privilege.

As with all qualitative research, the results are specific to the context and cannot be statistically generalized. However this theoretical analysis of my own experiences and mistakes provides important insights which are ‘transferable’ (Miles et al., 2014: 314), that is have implications for other settings with theoretically similar conditions. The findings are particularly relevant to other post-colonial, bilingual, cross-cultural, qualitative research settings. They also have implications for using and choosing pseudonyms in qualitative research in which inequalities of privilege and power between researchers and participants are related to characteristics other than race (e.g. education, economic status).

Researchers

The primary implication of this study is that, as suggested by (Svalastog and Eriksson, 2010), qualitative researchers should plan opportunities for participants to carefully deliberate the use and choice of pseudonyms. More specifically the results of this study indicate that cross-cultural researchers should create opportunities for participants to deliberate: (1) whether or not pseudonyms or real names should be used; (2) discuss the assumptions about ethnicity and racism that might be made when colonial-language or Indigenous-language names are used; and (3) select their own pseudonym in a language of their choice. Inviting participants to select their own pseudonyms or use their own names, without providing specific opportunities to deliberate the implications, is insufficient.

The results further indicate that to enhance the transformative potential of deliberations about pseudonyms, they must not only be careful, but also critically reflective. My experience demonstrates that critically reflective deliberations which interrogate the reasons for pseudonym-related preferences may enable participants to recognize their marginalization and researchers’ privilege. On the other hand deliberations which unquestioningly accept marginalized participants’ preferences, for example colonised people’s tendency to make a virtue of the necessities of post-colonial realities, will likely reinforce and further mask the dynamics of marginalization and privilege inherent in cross-cultural research. I therefore propose that qualitative researchers further develop participatory activities like the ones used for member checking in this study, to facilitate critically reflective deliberations about pseudonyms. Compared to opportunities to check the accuracy their own data, member checking that involves critical reflection on researchers’ interpretive analyses has greater transformative potential. Documenting the deliberative process using ethnographic techniques enables long term, theoretically-informed analysis. It has greater transformative potential, because theoretical interpretation enables participants to ‘see’ their experiences differently, and potentiates transferring the findings to other similar settings. Critical ethnography may also be a useful technique for documenting and analysing broader issues of power and ethics in research.

This study concurs with and advances other researchers’ (see Grinyer, 2002) findings that participants’ pseudonym-related preferences are dynamic. The implication for future research is that opportunities for careful deliberations should occur throughout the research process. They should commence during initial informed consent procedures. At this stage my findings suggest that academic researchers should refrain from sharing their own pseudonym-related preferences, to avoid biasing participants’ choices. Careful deliberations should ideally continue throughout the research process, including during the writing and member checking of results. In particular researchers can enhance the ethical rigour of their research by creating explicit opportunities for participants to change their minds about pseudonym use, after they have seen how academic researchers choose to represent them in research reports.

Finally, the findings of this study demonstrate that researchers must be prepared to respond to participants changing pseudonym-related preferences, if deliberations are to serve any purpose but to further mask their privilege and power. This may simply require academics engage in research with an explicit plan regarding how they will respond if participants’ request, for example, to use their real names, in a study that has ethical approval to use pseudonyms. However, they might take additional steps such as flagging the possibility in their ethics review applications, and enquiring in advance from ethics committees regarding the procedures for changing pseudonym-related practices.

Ethics committee members

Ethics committees could facilitate researchers’ attempts to incorporate careful deliberations about pseudonyms into their research processes by questioning the implicit assumption that pseudonym use is ideal. They might instead, as Svalastog and Eriksson (2010) have previously suggested, adopt an expectation that qualitative researchers detail how they will facilitate careful deliberations about pseudonyms with their participants. They might also provide guidance regarding ethical procedures for facilitating deliberations, including through the use of participatory activities.

Readers of research reports

A further implication of this study is that readers of research reports should not judge cultural-sensitivity based on the language used for pseudonyms. Rather they should make these judgements based on whether or not researchers provide evidence of careful, critically reflective (and potentially transformative) deliberations about, and demonstrations that participants selected their own, pseudonyms. Readers should be aware that the practice of researchers representing non-White participants may be subtly racist and expect authors to reflect on their own privilege. Although it may be a necessity carried out with good intentions, ‘cross-cultural’ research should not be made a virtue. It should be seen as a necessity that can and must be transformed, by naming, acknowledging and problematizing the racist notions and White privilege that constitute and are constituted by White (or otherwise privileged) academics researching non-White (or otherwise marginalised) participants.