Language and translation strategies in researching migrant experience of difference from the position of migrant researcher

Introduction

In an increasingly mobile world characterized by ever more diverse communities, social scientists inevitably find themselves working with their fellow nationals outside the country of origin. As they share nationality and mother tongue with their study participants, researchers frequently collect data in a language different from the language of its subsequent report. Moreover, since they conduct fieldwork in their native language, they are often assumed to share a language experience with the migrant compatriots they work with. This raises very significant considerations related to the validity of the research and its dissemination on the one hand (Birbili, 2000; Temple, 1997, 2008; Temple and Young, 2004; Wong and Poon, 2010) and the commonalities and differences in language experience on the other. While crucial for understanding research process and production of knowledge, these issues have, nonetheless, not yet been given adequate attention.

In response, in this article I engage with my personal research experience as a Polish migrant and translator researcher researching my own ethnic migrant community in my mother tongue (Polish), while being affiliated to an English institution and expected to disseminate research findings in English. I recognize my specific positionality as a young, female, white, non-religious, middle-class academic of complex migration history (including living in several European societies), committed to values such as equality and diversity. It is necessary to stress that while I explore my language and translation experience, I do not have a background in linguistics or translation studies. This paper should rather be understood as a critical reflection on the moments when I felt strangely detached from my compatriot informants and struggled to translate such moments into the language of report. As the study this paper draws upon focuses on migrant encounters with difference, in the article I consider the examples which relate to the language of ‘othering’. Although some readers may find this discussion uncomfortable due to the racist character of the terms involved, drawing academic attention to such issues is necessary to understand the nature of the challenges that migrant and translator researchers encounter.

Specifically, I explore the challenges related to the assumed shared relationship with language between migrant researchers and their migrant informants. In doing so, I contribute to the debate about positionality and reflexivity in researching migration experience from the position of a migrant researcher (Carling et al., 2014; Kim, 2012; Matejskova, 2014; Nowicka and Cieślik, 2014). Furthermore, I discuss the implications of collecting data in one language and presenting the findings in another, and seek to extend the discussion on translation in a research process (Claramonte, 2009; Kim, 2012; Smith, 1996; Squires, 2009; Temple, 1997, 2006, 2008; Temple and Young, 2004; Venuti, 1998; Wong and Poon, 2010). In sum, I address essential methodological queries many migrant researchers are likely to face when conducting projects involving their compatriot communities.

Social scientists have paid much attention to complexities of positionality in terms of ethnicity, religion, class, sexuality, gender, age, disability or personality (e.g. England, 1994; Haraway, 1988; Kobayashi, 2003; Ley and Mountz, 2001; McKay, 2002; Moser, 2008; Rose, 1997; Ward and Jones, 1999). Nonetheless, there seems to remain an assumption that migrant researchers of the same national/linguistic background as their migrant respondents are able to elicit ‘better’ responses from their participants and provide more authentic reading of the research data (Nowicka and Cieślik, 2014). In the context of dual-language studies there are strong voices which discuss the translation of research data for dissemination and challenge the common practice of conducting research using interpreters (Squires, 2009; Temple, 1997; Temple et al., 2006). However, so far less attention has been drawn to migrant researchers studying their native populations abroad in their native languages. It is important to acknowledge that a consequence of this lack of reflection is that to date little has been said about the practicalities and challenges of translation that migrant and translator researchers frequently face.

Importantly, there is a growing need for refreshing and deepening the discussion on translation of research specifically into English. Indeed, English is currently the lingua franca of global academia (Ferguson et al., 2011; Hamel, 2007). Disseminating knowledge in English became a part of a publishing routine and – on a more negative side – the only chance for many scholars to promote their studies to a wider international audience. In addition, many senior researchers, and international research teams, tend to have the collected data translated for them by professional translators external to the academia. In doing so, these scholars ‘risk’ being excluded from a significant part of data analysis process as translation of research data requires making crucial decisions about how to represent ideas and people (Simon, 1996).

In the article, I draw upon 32 in-depth interviews with a group of 14 Polish migrants who arrived in Leeds, UK, post-2004 when Poles were granted full access to the British labour market. The interviews lasted between 70–180 minutes and explored the nature of encounters with difference prior to and after moving from Poland (a postsocialist society relatively homogeneous in terms of ethnicity, nationality and religion) to the UK (a diverse postcolonial state). The study sought to explore how migrant values and attitudes towards difference in terms of ethnicity, religion, class, sexuality, gender, age and disability develop, alter or evolve through the meaningful contact with increased diversity. As such, it did not intend to address language and translation issues. However, my cooperation with the research participants has drawn my close attention to the role of a migrant and translator researcher in the process of interpreting of highly contextualized data.

I begin the article with exploring the complexities of positionality in cross-cultural research. Then I focus on translation dilemmas I have faced and discuss a simple data translation procedure which contributes to the practice of conducting dual-language studies. Finally, I investigate the assumed shared language experience between the migrant researcher and the migrant researched.

Position of a migrant researcher

The last three decades brought a revolutionary turn in understanding the production of knowledge in social sciences. The previously dominant paradigm, which conceptualized knowledge as universal and value-free, and researchers as objective experts who ‘remain detached and neutral from their subjects’ (Kobayashi, 2003: 346), was challenged by vast – particularly feminist – critique (e.g. Oakley, 1981; Stanley and Wise, 1983). As a consequence, a major shift towards situated knowledge produced in highly contextualized circumstances (Rose, 1997) and by non-neutral researchers embroiled in changing power relations was acknowledged. In her influential paper Haraway (1988) famously argued that assumptions about the alleged researcher’s ability to see everything from nowhere, is simply an illusion, a ‘god-trick’. Recognition of knowledge filtered through individual biographies, lived experience, the ‘embeddedness [of knowledge] in practice’ (Moss, 1995: 446), has drawn academic attention to the meaning of the ‘position’ or ‘positionality’ of the researcher. In this context, McDowell claimed that ‘we must recognize and take account of our own position, as well as that of our research participants, and write this into our research practice’ (1992: 409). This resonates with Rose’s calls to view the researchers as subjectively positioned individuals who inevitably ‘see the world from specific locations, embodied and particular, and never innocent’ (1997: 308).

The voices advocating that race, ethnicity, age, gender, sexuality, religion, class, disability, personality (and a unique intersection of these categories) may not only determine the outcomes of a fieldwork, but also affect research-related interactions or relationships with informants, have been widely accepted. Cohorts of young researchers have been advised to be self-reflexive at every stage of the qualitative research process, ‘to link the idea of self to the process of knowledge construction’ (Avis, 2002: 205), ‘reflect upon oneself and one’s experiences’ (Falconer Al-Hindi and Kawabata, 2002: 104) and be aware of ‘our consciousness [being] always the medium through which the research occurs’ (Stanley and Wise, 1993, cited in Rose, 1997: 309).

Even though ‘open’ European borders encourage academics to conduct research abroad and numerous legal solutions tempt scholars from all over the world to work particularly within the Anglo-American zone, there is less discussion about the challenges related to the positionality of migrant researchers studying their compatriot migrant communities. This concern is shared by Kim (2012) who confirms that the academics have so far conceptualized such positioning in simplistic terms and viewed the relationship between the migrant researcher and the migrant researched as equivalent to the relationship between the researcher and the researched per se.

Against this backdrop, a complex positionality of migrant researcher researching his or her own migrant population has been so far underexplored and discussed largely with reference to the researcher’s insider/outsider status (Ganga and Scott, 2006). A telling example here is Mohammad’s (2001) research experience of the Pakistani community in the south of England. This British (Pakistani by birth) female researcher noticed that due to her ascribed ethnic identity she was automatically positioned as an insider by her informants. This positioning determined the interactions with her research participants as well as the data she collected. The author admits:

I was seen to be an ‘insider’, as someone who was from and hence ‘belonged’ to the local Pakistani ‘community’. […] This belonging was seen to endow me with a superior, almost organic knowledge of the ‘community’ not accessible to ‘outsiders’. (Mohammad, 2001: 101)

While still frequently recalled in migration studies, the insider/outsider binary is, nonetheless, a contested one as it inevitably simplifies the intricate and flexible nature of relationship between the migrant researcher and the migrant researched (Carling et al., 2014; Ergun and Erdemir, 2010; Matejskova, 2014; Merriam et al., 2001; Nowicka and Cieślik, 2014). Drawing on my own fieldwork experiences, the fact that I am Polish and I speak Polish definitely helped me gain ‘insider’ access to some people (e.g. people with poorer English) and made it easier to establish contacts during the initial stage of my fieldwork. Nevertheless, what became meaningful during the further stages were precisely those of my features that differentiated me from my research participants (e.g. my complex migration history, class background, education, the intersection of my gender and age, my approach to religion/belief or the values I live with). During my fieldwork I had many opportunities to find out that a presumption that individuals of the same ethnic background have a greater understanding of each other’s experience or views is based on a false premise that a single ‘truth’ or ‘reality’ shapes human reception of the outside world (Kim, 2012; Rhodes, 1994; Twine, 2000). Based on this research experience, I agree with Valentine (2002) who suggests that the assumption that people perceived as insiders can interpret informants’ stories more correctly is a dangerous form of essentialism as it (re)produces binary categorizations which do not capture the intersections – complexity and diversity of experiences and views – within as well as between various groups.

Translating research

While significant for broad social sciences, debates on language and contextualized translation as part of a research process are largely held within translation studies, sociology of translation and the studies of bilingualism (cf. Temple, 2008; Temple and Young, 2004; Wolf and Fukari, 2007). In these discussions, the act of translation is argued to be a socially driven process (Wolf and Fukari 2007). Given this understanding, translators undergo an intricate socialization that affects the way they interpret and translate concepts and meanings (Wolf and Fukari, 2007). Thus, the discussion of the language experience and translation of research is closely connected to the positionality debates. Nonetheless, although translators are inherently involved in the knowledge production process, their role in shaping data remains underappreciated (Fathi, 2013; Temple and Koterba, 2009).

I understand translation as ‘a multidimensional process that involves different epistemological layers’ (Fathi, 2013: 55). ¹ In this article, I am particularly interested in the issue of ‘interpretative decisions’ (Riessman, 2008, cited in Fathi, 2013: 55) involved in producing a transcript from a translated interview. In this context, Temple (2008: 362) calls for what Venuti (1998) has described as an ethics of translation – an awareness of a translator of having ‘responsibilities to research participants regarding the way [he or she] represents them in writing’. This refers to Simon’s (1996) argument that translators are continuously involved in making choices about how to represent people.

Representation of others is not the only ethical problem involved in translation. Equally important is the awareness of possible change in meaning or potential construction of new meanings (Claramonte, 2009; Fathi, 2013; Kim, 2012). Therefore, the discussion often concentrates on the notion of quality of translation and the translator’s skills. Birbili (2000: 1) claims that the quality of translation tends to be affected by ‘the linguistic competence of the translator/s; the translator’s knowledge of the culture of the people under study; the autobiography of those involved in the translation and the circumstances in which the translation takes place’. All these factors are even more evident in the case of migrant researchers.

Given the thorough theoretical debates on language and translation, less attention is paid to a practical side of translating data as part of a research process. In this article, I address this issue by presenting a fairly simple procedure I developed through the course of my research and as a result of my ongoing translating/language dilemmas. I am aware that similar practice is implied by language and translation theoreticians (e.g. Pavlenko, 2006; Spivak, 1993; Temple, 2006, 2008; Temple and Young, 2004, 2014; Venuti, 1998). Nonetheless, social scientists without a background in translation studies may be less aware of the significance of such approaches for the production of knowledge. Thus, I believe, a reflexive discussion of the practicalities related to translating research data (e.g. particular translation decisions) is so crucial. Even though in my discussion I focus on translation from Polish into English, I hope non-Polish-speaking researchers involved in dual-language studies will find it meaningful and will be inspired to develop similar approaches including other languages.

The undeniable problem of translating research data is a risk of getting lost in translation. Indeed, the translation process is a hybrid cultural production (Smith, 1996) and ‘involves translating lives rather than simply words’ (Temple and Koterba, 2009: 2). This is mainly a matter of the so called conceptual equivalence (Birbili, 2000; Temple, 1997), or – in other words – comparability of meanings between the original utterance and the translated transcription. In this article, I suggest that if translation is prepared with extra care so as conceptual equivalence is maintained, the quality of transcribed (and translated) data can remain high and fully appropriate for further analysis.

Over the course of my study, I developed a procedure which allows me to gain conceptual equivalence by including in every translated transcription detailed notes clarifying the context and supplementing them with the original utterance. I rigorously produced such notes whenever an interviewee a) left a message unsaid, b) his or her utterance carried emotional connotations, a set of cultural or personal assumptions and/or values impossible to translate, c) he or she used a grammar form and/or an expression which was hardly translatable into English. These notes were introduced in square brackets and included my comments related to the usage of particular language-specific expressions, grammar forms, indirect translation or an interviewee’s behaviour as well as the assumptions he or she made. They also contained the word/phrase/utterance in the original language (Polish in the case of my study). This procedure has provided me with a translated transcription very close to the linguistically nuanced record in the original language. Having in mind the justified warnings of more experienced researchers (Erickan, 1998; Temple, 1997, 2008), I tried to be most careful when transcribing and simultaneously translating, and repeatedly compared my transcription with the original recording.

Below are two extracts ² from my transcripts which shed light on the logic of the translation procedure. I realize that the sequences of narratives in the original language are only useful for the people who speak Polish. However, my aim is simply to show the results of the procedure. Getting familiar with it does not require the knowledge of any particular language. In the first extract the interviewee, Ela, speaks of her group of Polish friends:

So, I have few friends … Generally, we’re all more or less – how to say this – we’re all the same/similar story [PL: Generalnie, wszyscy jesteśmy mniej więcej – jakby to powiedzieć – z tej samej półki]. All girls, all after ogólniak [general profile secondary school – ENG: high school/grammar school], the same age more or less, 28–34 years old. Most of us with families, married, kids, no kids. (Ela, migrant, female, aged 31)

After saying that she has ‘few friends’, which was literally translated into English and did not require any comments on language, Ela tries to describe this group and uses a colloquial expression which would not make much sense if translated from Polish in a direct way. In order to gain conceptual equivalence the part of the sentence was translated into: ‘we’re all the same/similar story’ whereas if translated literally it would rather mean: ‘we’re all from the same shelf’. Since the translation is not a direct one, there is a line in original language – in italics and in square brackets. The previous line did not require to be supplemented with an original utterance as it was directly translated. Similarly, the subsequent lines are not supplemented with original utterances. However, as the interviewee uses a colloquial name for a type of secondary school in Poland – ogólniak – this is also explained in square brackets and supported with the closest equivalent(s) in English – ‘high school/grammar school’. The Polish term ogólniak is left in the quote also to indicate that the interviewee’s group of friends comprises of individuals with specific general secondary education, which in the Polish context is perceived as more prestigious than the vocational education, for instance. This way the whole quote captures all personal and social assumptions the speaker leaves unsaid about her group of friends.

Another extract I explore is a short story of moving out from a Northern English town. The interviewee, Marek, speaks of the problems he encountered in his neighbourhood and refers to the people who live in council accommodation. He borrows an English word ‘council’, which is then inflected (declension) and used in the Polish fifth grammatical case (instrumental case). ³ It is impossible to express such a language structure in an English translation. Thus, the phrase is simply translated into ‘council-flat people’ and requires an original line which carries a set of unique meanings. A researcher studying migration will be fully able to analyse this data in the next step and consider it, for instance, as an example of flexibility of native language in transnational settings. Moreover, a researcher interested in difference and ‘othering’ will be able to notice how migrants possibly conceptualize people of different social status within the host society:

When I moved to England my company rented a flat for us – so, we lived there. But, we didn’t stay long because there was a situation with council-flat people [PL: zaistniała sytuacja z councilami]. Because, we lived in this neighbourhood with many young Brits [PL: Angolami]. They knew we were Polish and they were a pain in the neck for us [PL: uprzykrzali nam życie]. There were always some minor tensions. So, there was no point in staying there. (Marek, migrant, male, aged 32)

In the quote, the interviewee uses another interesting expression Angole which derives from a Polish word Anglia – meaning: England – and refers to English people. Yet due to its slightly pejorative meaning the word was translated into ‘Brits’, which embraces this deeper sense much better but refers to a much broader group – British people. Since there is a risk of change in meaning the original expression in square bracket and italics was included. In the case of the last square bracket a rather colloquial Polish expression was exchanged with an English idiom which stresses the specific bitter-sarcastic undertone of the whole sentence.

These two examples demonstrate that a nuanced translation of data is relatively easily achievable provided that a translator researcher is highly sensitive to maintaining conceptual equivalence and supplementing a translated transcription with additional explanatory notes. Such comments introduce further contexts that data refer to. However, on a critical side, there is an obvious limitation of this translation procedure. Kim (2012: 140) argues that ‘translation is a job that involves a speaker, a translator, and listeners (audience)’ and suggests that translators should keep their potential audience in mind. These readers cannot be perceived as a homogeneous group as they possibly come from various cultural, social and linguistic backgrounds. In this respect my procedure is only a partial solution since it can be fully appreciated only by the readers fluent in both used languages. Nonetheless, I believe it serves a very useful purpose especially for lone researchers expected to work with and disseminate data in a language different from the language of its collection. It can be also beneficial for international research teams involving several multi-language scholars working together in a selected language (e.g. English), some of whom collect data in other language(s).

Shared language experience?

Migrant researchers (Kim, 2012; Temple, 2011) frequently stress that research participants tend to identify their and the researcher’s (non)belonging to home and host societies. This is most often expressed by the usage of the pronouns ‘we/us’ and ‘they/them’. It could be argued that these are verbal expressions of imagined communities (Anderson, 1991) or the so called we-images (Mennell, 1994) formed by the migrants who share a language and a nationality. Unsurprisingly, this tendency was also noticeable in my study. The research participants rarely spoke of Poland/England/the UK or Polish/English/British people. Many of them would rather speak of ‘our country’/‘our place’ (Poland) versus ‘this country’/‘here’ (England or the UK) and how things are done by ‘them’ (English or British people).

Particularly interesting was the usage of the pronoun ‘us’ meaning – depending on the context – either Polish migrants to the UK or Polish society in Poland. My informants not only assumed I shared a sense of belonging to the Polish nation, but also the same migration experience. In other words, I was assumed to share a we-image (Mennell, 1994) with the research participants. In his study of Welsh people in London, Segrott (2001) found that a shared language is an important element of an assumption of a common identity and values. In the case of my research, (un)conscious banal expressions of shared national identity and migrant status were sometimes accompanied by an assumption of a shared religious identity. Such an assumption is most probably made due to the fact that Poland is predominantly Roman-Catholic (Eberts, 1998). Even though there is a number of religious minorities (as well as non-religious people) in Poland, I noticed that the majority of the Polish migrants in my sample (no matter whether believer, non-believer, practising religion or non-practicing) assumed that their migrant compatriots were Catholic. This is exemplified in the extract below in which one of my interviewees, Magda, a Roman-Catholic married to a Muslim person, describes how she realized there were many commonalities between Christianity and Islam. In the quote, she speaks of ‘our religion’ meaning Catholicism:

The story from the Old Testament is the same in our religion and theirs. […] So, at first we were both very surprised because we realized that we both knew Abraham, Moses … that we both knew all these stories. We had thought we came from two completely different worlds and it turned out that we had the same bases. And, for example, in our religion, the most important person was Jesus. In their religion it was Mohammed. (Magda, migrant, female, aged 28)

The pronoun ‘our’ may indicate two interconnected assumptions the speaker makes. The first one is an assumption of religion shared by the interviewer and the interviewee. The other is an assumption of Polish society being originally Christian/Catholic. In the case of the latter, the interviewee may not assume that that the interviewer is Catholic herself, but since she is Polish she must consider this religion as an element of her Polishness. In other words, even if the speaker did not assume I was Catholic myself, she assumed that since I am Polish I must be familiar with what she calls ‘our religion’. Moreover, the pronoun also plays an ‘othering’ function – the interviewee opposes ‘us’ (Christians) to the imagined ‘them’ (Muslim people). This extract illustrates that an assumption of shared language constitutes the assumption of a shared identity and values.

During my study, I observed that the presumptions research participants tend to make are in the first place related to nationality, language and migrant status. This is not to say migrant informants never made assumptions according to my gender, age, appearance or any other guessable or ascribed features (e.g. education, sexuality, marital status) as well as intersections of a few of these categories. Indeed, a few of my female informants spoke of ‘our bodies’, meaning female bodies, while discussing gendered upbringing or certain professions dominated by men. Other interviewees would repeat ‘you don’t remember how it was’ referring to the communist regime in Poland and thus making age-based assumptions about me. Nevertheless, I felt these features were definitely less frequently subject to various presumptions in comparison to my nationality, native tongue and migrant status. On many occasions, the research participants assumed that since I am a Polish migrant resident in England, I would easily read between the lines and immediately understand their own experience of migration. That was particularly noticeable in the way I was provided with research data. Some of my interviewees spared me certain context or culture specific explanations by simply saying: ‘we both know what I mean’ and giving me a knowing look. In such moments I felt these informants strongly believed in a symbolic spiritual alliance between the two persons who were both raised in the same country, with the same language and now share all pros and cons of the migrant life.

Below, one of my male informants Marek, quoted earlier, describes his experience of Bradford, UK, which is home to a substantial Pakistani community. After mentioning Bradford he just says: ‘so – you know’ implying that Bradford is for whatever reason a specific place. In doing so, the interviewee assumes that I share his understanding of Bradford and all the personal connotations this name carries for him:

We went to a playground – it was in Bradford, so – you know [gives a knowing look]. So, we went there and there were hell knows how many of these women in letter-boxes [PL: w diabła kobiet w letter-box’ach; he means burqas/abayas here] – and we both know what that is … plus, [women] in scarves only – typical Muslims [PL: typowych Muslimów] with their children. (Marek, migrant, male, aged 32)

This extract is very interesting for several reasons. The speaker, for instance, does not say women in burqas/abayas but uses a discriminatory-grotesque expression ‘women in letter-boxes’. In doing so, he borrows an English word ‘letterbox’ and uses it in the fifth Polish grammatical case (instrumental case). I remember thinking for a while before I realized who he was referring to. This is explained in the sentences that follow as he speaks of women in scarves and ‘typical Muslims’ (again, he uses an English borrowing here instead of the Polish equivalent of a word ‘Muslims’). Even though the informant used a discriminatory language, I could tell from the context that he did not intend to offend Muslim people. Yet, he took the liberty to express a comment with a racist undertone because he was there sharing a moment with me – another Pole in exile who furthermore expressed a deep interest in his personal experience of difference. This discriminatory linguistic construction appears here to be a verbal form of normalizing a dissonance that comes with an encounter with difference rather than an expression of hostility and racism. Moreover, it demonstrates that in the pseudo-intimate company of a fellow national who will obviously get the joke specific data are (re)produced as a result of the assumptions of a shared language and migration experience. This is not to infer that my informant does not use certain rhetoric apart from the interview situation or a Pole-to-Pole situation. My point here is that assumptions of shared language and migration experience affect power relations between the migrant researcher and the migrant researched. As a result, and in line with the (feminist) positionality debates (Katz, 1996; McDowell, 1992; Moss, 2002; Nast, 1994), highly specific research data are generated.

The ambivalent language of difference was one of the biggest translating challenges I faced. I was particularly confused when some of my most open-minded and respectful informants would all of a sudden use such expressions as, for example, Pakole, ciapaci or ciapasy, describing Pakistani people. The closest translation of all the three words would be ‘Pakis’ – however, this word was definitely too strong for the non-prejudiced or even positive tone of some of the utterances. I also struggled trying to distinguish between them. Whereas Pakole is closest to ‘Pakis’ and is definitely very negative, needless to say racist, ciapaci or ciapasy is much more urban-slangish, quite pejorative yet not as strongly derogatory as Pakole. I believed that if I stuck to a fixed dictionary and translated ciapaci into ‘Pakis’ each time, I would misrepresent some of my informants in writing and that it would be ethically wrong. Therefore, recalling Simon’s argument (1996), every time I had to make a separate and independent translation decision based on a specific context. Many times though, I would leave the original word ciapaci or Pakole and supplement it with an explanatory comment describing the context as I felt I was not empowered to make any choices related to the translation of these terms.

I had a similar dilemma with another word – Murzyn – indicating a Black person. This word appeared in Polish language in the14th century as a linguistic borrowing coming from the Latin maurus meaning black and has been used ever since with reference to a person with dark skin colour (Ząbek, 2007). However in contemporary Polish language it introduces a specific ambivalent context. It has largely negative social connotations as many folk sayings or idiomatic expressions involving it refer to a situation in which somebody is a servant, a slave, a cheap work force or a representative of a group ‘believed’ to be backward (Ząbek, 2007). Moreover, these sayings address racist stereotypes. An interesting case here might be a small poem/rhyme written by an interwar-period Polish poet Julian Tuwim. The poem titled Murzynek Bambo used to be learnt by heart by many Polish children during the 1980s–1990s. The title of the poem could be translated into ‘Bambo the little Black boy’ or, more accurately, into ‘Bambo the Nigrette’. Moskalewicz (2005) argues that the rhyme has a strong colonial undertone and refers to simplistic and essentialist representations of people of African-Caribbean descent. The Bambo boy is indeed a joyful little savage-kid who is a diligent pupil in a school somewhere in Africa; a prankster who climbs a tree when he wants to run away from his mother; or a funny exotic figure who is afraid his skin will became white while bathing.

In a survey of a representative sample of adult Polish people 20% of respondents regard the expression Murzyn insulting, while another 12% – sometimes offensive and sometimes not (CBOS, 2007). Nevertheless, a substantial portion of survey participants consider the word inoffensive, which illustrates a broader tendency within Polish society. The influential Polish–English PWN-Oxford Dictionary (2004) mentions few possible English equivalents of the word: ‘a Black(man)’ [sic], ‘a person of African/Caribbean descent’ and ‘Negro’. Nevertheless, considering the associations the word Murzyn may evoke and the historical translations, the closest or most traditional equivalent would be ‘Negro’ or a more contemporary-slangish, highly offensive and prejudice-loaded term ‘Nigger’ (Rahman, 2012).

The translation of the word was a tough task indeed as my sample involved a few types of linguistically (un)aware informants. The first cluster includes the interviewees who did not use the word Murzyn at all as they were aware of its racist undertone. The second cluster embraced the people who would use this word in an unprejudiced context being unaware of its negative connotations. The last cluster were the informants who would use it in the context of their prejudice either being aware of the word having pejorative meaning or not. Whereas the first usage of the word is beyond the interest of this article, the other two became the subject of my great ethics-versus-good-translation dilemma. Initially, I thought I should stress the context and translate the word in two different ways choosing between the words ‘a Black person’ and ‘Negro’/‘Nigger’. However, I abandoned this idea as I realized that even though I could define the context (favourable or unfavourable attitude, racism, etc.), as a translator I was simply not empowered to make such choices due to ethical reasons. This is why, unlike the Pakole/ciapaci/ciapasy dilemma, I decided to translate the word Murzyn into ‘a Black person’ every time. Naturally, each usage of the term was supplemented with the original Polish word and, if necessary, additional comments on context.

The limited awareness of the discriminatory undertone all the described words introduce remains most intriguing. Even though some of my informants were fully aware of this rhetoric being inappropriate, others seemed to have no sense of that at all. Importantly, from the context I could tell that many participants from the latter group used the language of prejudice without an intention to offend. While this obscure practice needs further academic attention, it is partly related to the relative ethno-national and religious homogeneity in the Polish context, which affects how, for example, equality legislation is understood and exercised. It has been suggested that although equality is rather broadly addressed legally, the degree of legal awareness, social censure and the penalizing of abusive and discriminatory language are relatively low in Poland (Gołębiowska, 2009). Meanwhile, it is apparent that migrants who share a language can have extremely diverse awareness and understandings of this language.

Conclusions

This paper draws upon and contributes to the key positionality and translation debates by exploring the contested shared language experience between the migrant researcher and the migrant researched as well as the translation challenges migrant and translator researchers are likely to face. In the context of increased international mobility of academics and the dominant role of English in disseminating knowledge, I argue for more recognition of translation issues. Since my main research interest is in migrant encounters with ‘otherness’, I focus on the language of difference.

On the one hand, we are encouraged to ‘giv[e] as full and honest an account of the research process as possible, in particular explicating the position of the researcher in relation to the researched’ (Reay, 1996 cited in Falconer Al-Hindi and Kawabata, 2002: 104). On the other, the assumption that researchers from the same backgrounds as their respondents are able to elicit better data from their informants and provide a more authentic reading of it is persistent in the global academia (Nowicka and Cieślik, 2014). Migrant researchers’ and translator researchers’ positionings are highly specific and transgress the contested insider/outsider status or ascribed positionalities based on ethnicity, age, gender, sexuality or social class. Valentine (2002) claims that a research encounter is often a game of similarities and differences. Most frequently it is both parties – the researcher and the researched – who demonstrate multiple (dis)identifications and make various assumptions about each other. My study evidences that in the case of migrant researchers studying their own compatriot communities, such presumptions are likely to be based on nationality, language and migrant status in the first place. In particular, they may be linked with assumptions of a common we-image (Mennell, 1994), religious identity or the same understanding of difference. Moreover, such assumptions may influence the rhetoric used and the data produced during a research encounter. This may result in strongly contextualized sets of information which must be carefully and ethically translated into the language of dissemination.

The language we use is a product of all the individual experiences we have collected, our age, social background, education, religious attachment, or importantly, the values and beliefs we live or choose to live with. It is unique as much as the speakers, their experiences of migration and ‘otherness’, are unique. Therefore, conscious research translation should always be a priority while collecting data in a language different from the language of its subsequent report. Translator researchers must never underestimate the risks and the challenges translation involves. The examples I explore in this paper are evidence that researchers translating data are often forced to make significant choices about how to represent their informants in writing (Simon, 1996). For this reason, the role of a translator researcher is very distinct from the role of a researcher per se.

In the paper, I also discuss a simple data translation procedure which allows me to gain conceptual equivalence and explore nuanced language contexts. I am aware many senior researchers frequently have the collected data translated for them and it is not my intention to discredit such a solution. Nevertheless, I would like to draw their attention to the very powerful position of translators and their role in shaping research data.

Finally, this article is first and foremost a reflexive account of a personal language and translation experience. It is crucial to reiterate that my reflections are connected to my ‘complicated’ positionality and my sense of (non)belonging in relation to research participants. As my research sample was diverse in terms of social positionings and attitudes towards difference, it is unsurprising that I experienced many moments of multiple (dis)identifications (Valentine, 2002). Nonetheless, what made me feel particularly detached from some of my informants was the language of difference drawing upon orientalizing, essentializing and racializing discourses of the imagined Other. Despite the fact that it was sometimes used to express favourable feelings, I found this language stigmatizing and discriminatory. In exploring it I considered that how difference might be described and related to is a consequence of socio-historical circumstances in the Polish context (e.g. relative ethno-national and religious homogeneity of Polish society, social limitations of equality legislation).