Challenges and strategies of translation in a qualitative and sensitive research

Introduction

This study examines how qualitative and sensitive research can translate the data from a different language into English. This is based on the doctoral research of investigating Turkish perpetrators’ engagement in domestic violence interventions in the UK. This research explored how these men engage in domestic violence interventions by conducting semi-structured interviews with Turkish men who attended domestic violence interventions and the professionals who worked with this group. Some participants who emigrated to the UK had poor English communication skills and so preferred to speak in Turkish during interviews. This illustrates the power of language on the data. Twelve out of 20 semi-structured interviews were conducted in Turkish, which was the participants’ native language. After each interview, the interviews were transcribed in Turkish and translated into English by the researcher. There were issues of transcription and translation in terms of producing acceptable data and findings. The audio-recorded interviews were listened to at least three times to catch every word and achieve a comprehensive data set. Giving participants’ original phrases or sayings in Turkish was beneficial to illustrate the complexities of the language. Furthermore, sharing the same native language increased the awareness of the meanings of certain phrases and sayings in their social and cultural contexts was useful. This article explores the key challenges and strategies around the translation regarding the time of the data collection and analysis.

Several researchers (Barrett, 1992; Simon, 1996; Spivak, 1992) point out that translation of the text in a qualitative study cannot be done through the simple translation of the data because they argue that ‘meaning is constructed through a discourse between texts’ (Temple and Young, 2004: 165). Therefore, the translation of qualitative data is not a straightforward practice, especially in a sensitive and complex research environment. This research is sensitive as it explores the multifaced subjective lived experiences. For example, Johnson (2018) paid attention to the importance of considering ethical issues and the engagement with gatekeepers during sensitive research. Any attempt at a translation of data can be questioned because of the complexities around the concepts that can be understood from the participants’ words based on the readers’ social and cultural contexts (Temple and Young, 2004). Specifically, examining Turkish men’s engagement in domestic violence intervention processes was complicated due to the participants’ minimisation of their violence and justifications for their inadequate engagement in interventions by blaming women and the legal system. Furthermore, there were strong links made between race, gender, class, migration and programme related conditions in their stories. This complication increased the sensitivity of the context and language in the data. For instance, the majority of the men used Turkish phrases to describe their justifications for their actions. Whilst there were complexities in the processes of providing a high-quality translation in the presentation of the findings, the benefits of reflexivity and other strategies on the translation should be considered.

Translation issues are critical for the validity of the data and for presenting the findings (Birbili, 2000; Temple, 1997, 2008). Gawlewicz (2016) noted that the circumstances of translation during data collection and analysis should be considered. Some studies also paid attention to the importance of reflexivity of the researcher’s position as a migrant on language issues (Carling et al., 2014; Kim, 2012). The discussions on methodological issues and implications of using a different language during data-gathering to data presentation are helpful for those researchers who translate to present their data in English. Nevertheless, there are challenges around the researcher’s epistemological positions in terms of how to represent the individuals’ meanings by using words that are ‘best in a sentence in a language’ (Temple and Young, 2004: 164). Therefore, this article addresses the following research questions: (1) What are the epistemological implications of a researcher holding migration status? (2) How does the translator/researcher apply strategies to avoid missing the meaning of the participants? (3) How does the researcher get involved in the translation process? In responding to these questions, the influences of language and strategies to provide effective translation will be clarified. These strategies can help illuminate participants’ meanings in their stories in qualitative research. The complex translation experiences will be presented by giving some examples from the data and findings.

Epistemological implications of the researcher’s position as a translator

It is important to discuss how race, gender, class and other social structures might have impacted the researcher’s relationship with the participants by concentrating on the translation. Gawlewicz (2016) pays attention to the roles and influences of the translator’s positions on data analysis and the outcomes of the research study. Likewise, Fathi (2013) and Temple and Koterba (2009) highlighted how holding any preconception about the data is inappropriate. The translation process should not lose the meanings in the text or produce new meanings. A good strategy to create high-quality translation is for the researcher to be aware of their positionalities. An effort to understand the best way to collect data from a different language and translate it into English must focus on the researcher’s epistemological and ontological position and the power of the language during data collection and analysis. As the researcher’s positions are often identified as key in gathering data and analysing it (Temple, 2008), these critical issues and under-researched conditions around translation will be discussed.

Being aware of insider position in terms of sharing the same language was beneficial for reaching in-depth interviews. For instance, one participant started to respond to the questions in English at the beginning of the interview. However, he changed the language when he started to share more intimate and personal experiences. This participant used mixed language during the interview, and the major language of his talk was Turkish. Therefore, having the same native language as an insider position was helpful to access in-depth experiences. Insider position may have helped the researcher to build rapport and mutual understanding by using a similar vocabulary. Also, this position might have increased their cooperation during interviews. This might have impacted translation in positive ways because they were able to express their feelings or experiences by using phrases or sayings in their native language. This is frequently enhanced to clarify their experiences in effective ways.

While the researcher’s insider position was helpful to gather data in terms of sharing the native language, the outsider position also impacted on accessing data. For example, many participants might have been unwilling to communicate with someone from the outside of their community. The researcher’s educational attainment was one of the outsider positions. A few participants did not permit audio-recording because they might have mistrusted the educational affiliation. In this position, understanding the potential experiences of racism, discrimination and oppression in the community was an important element to increase the researcher’s sensitivity to the data. Therefore, this awareness was critical to achieving a high-quality translation. Moreover, Twine (2000) emphasises that sharing the same race and ethnic background with research participants does not always have a positive effect on the translation. Being an insider or outsider position includes complex circumstances as these positions are perceived in different ways based on individuals’ unique conditions.

The participants were more willing to explain the researcher’s further questions about confusing phrases when they trusted the researcher. For example, the first author’s positionality was configured as a Turkish migrant doctoral student who seeks help for her thesis. Some participants might have perceived this was a vulnerable position due to the difficulty of accessing a hard-to-reach population and they were the only individuals who can help to gather data. In this sense, many men were willing to share their experiences regarding the feeling of helping a student who held the same racial and ethnic background and moved to the UK for her degree. Importantly, this position sometimes increased their participation as well as responses regarding further clarification about vague meanings. On the other hand, a few participants held very difficult experiences and the sentences about their struggles were often unclear. Within these struggles, they were not able to express their feelings and experiences exactly. Therefore, participants’ difficulties with their sensitive experiences were reflected in some uncompleted sentences and difficult translation. The researcher managed her migrated student position by concentrating on how their words are important and vital regarding their sensitive lived experiences. Moreover, being a female researcher was a difficult and uncomfortable position due to hearing ideas about women blaming. The first author worked on focusing on what they tried to tell her as a researcher without perceiving these ideas towards the female gender position. Participants may have felt uncomfortable in talking about domestic violence with a female researcher. This position is also recognised by using reflexivity.

Challenges and strategies of the translation

Throughout the translation of the data, key strategies were adopted to provide the meanings of the texts and bring out the voices of the participants. For instance, the researcher’s efforts to map the journey of how to access the participants and the detailed notes around the process of gatekeeping were helpful to realise how they were the hard-to-reach groups (Turhan and Bernard, 2020). When untranslatable sentences were phrased in English, the original Turkish was provided in brackets. Many researchers highlight the importance of providing the source of the text (Erickan, 1998; Temple, 1997, 2008). The process of translating and comparing the original text illustrates the researcher’s careful translation process. Birbili (2000) describes back translation as a strategy of returning to the source language to reduce and remove uncertainties or inconsistencies in the meanings of translated transcripts (Ercikan, 1998). This helps to remove misinterpreted words. When the first author applied line-by-line coding during data analysis, she often implemented back translation to ensure the proper meaning of the words emerged. Back-translation was useful to increase the quality of the translation by maintaining clear meanings in the text.

Having the same native language and similar social and cultural background might be beneficial during the data collection process (Vulliamy, 1990). Based on the researcher’s own experiences, being a translator and researcher was beneficial for providing a more comprehensive analysis. For example, this contributes to the knowledge in the field because working on both transcribing, translating and interpreting the data increased the researcher’s familiarity with and sensitivity of the data set. The circumstances of translation were related to the location, accompaniment, contingency and role (Tilusubya et al., 2018). A few interviews were conducted at the restaurant where the participants were working. This circumstance was challenging as they were not comfortable speaking about their sensitive experiences. The researcher solved this uncomfortable environment to ask general questions at the beginning of the interview and not push the participant to explain in detail their obstacles by being respectful of their positions. After the interview, the researcher took notes about these circumstances to recognise their unique interview environment. Birbili (2000) argues that conceptual equivalence should be considered during translation. Word-by-word translation of the data might reduce the readability of the texts because of the complex topic. For instance, to increase the readability of the extracts, a freer, more ‘elegant’ translation should be implemented (Birbili, 2000). It is believed that the researcher’s knowledge about the participants’ positions in their social and cultural context and her command of the language prevented this from occurring.

Meanings of ‘here’, ‘there’, ‘we’ and ‘they’

The indexicality of language use recognises how some expressions such as ‘this’ and ‘that’ are conditional based on individuals’ positions and their sensitive experiences. For example, many men often preferred to use ‘that’ or ‘this problem’ rather than stating violence or violent behaviour. The reason for that they might have felt shame about their violent behaviour. Thus, ‘that or this problem’ indirectly indexes violent behaviour. Moreover, personal pronouns and deictics (there, here, now) are important indexicals to recognise the relationship between the situated interaction and reproducing that action. For instance, many participants use the word ‘here’ to mean the UK. Even though there is heterogeneity among Turkish groups within their religious and cultural perspectives, the participants frequently generalised their social and cultural values to all Turkish people in the UK. The majority of the participants used specific expressions when they compared their experiences regarding culture and legal issues in their home country to those in the UK. For instance, some participants often used the word ‘us’, ‘our place’ and ‘we’ to mean their home country and people who hold the same religious, cultural and ethnic background. The first author put these explanations in brackets in the extracts:

In general, the stressful events are financial problems, difficulties of adaptation and homesickness in the UK. When these three come together, everything falls apart. We [Turkish families] always take heed of what our mothers, uncles and other relatives have to say. As you know, we have a feudal structure.

Phrases and expressions based on cultural values

One participant expressed the difference between Turkey and the UK based on his social and cultural background by using a few difficult words. In the quote below, the researcher put extra explanations for the participant’s meanings around vague word and phrases:

For instance, they say, ‘Yes, you are my mother and father, but you are not everything to me.’ This is unheard of in Turkish culture. Our fathers and mothers are everything. ‘Paradise lies at the feet of the mother’, as the saying goes [Cennet annelerimizin ayaklarının altındadır: the children need to be very respectful of their mothers]. This is our attitude but people who grow up in the UK become ignorant of their cultural values.

The participant used some phrases related to cultural and religious values. Translating these into English was challenging because word-by-word translation did not adequately convey the meaning of the participant’s words. As such, they were translated by giving the actual meanings of these phrases in English and then the original text in Turkish was provided. This strategy was helpful for the readers who could then understand the complexity of the language and the sensitivity of the context of domestic violence perpetrator intervention among marginalised ethnic groups. Likewise, some expressions were also needed to extra explanations for the clarifications. For instance, the explanations were added to clarify the actual text:

Of course, I think it is so important to take help from a person who is Pakistani if you are from Pakistan as [someone of the same nationality can understand my problems/issues more clearly].

Moreover, the participants often used untranslatable Turkish expressions when discussing sensitive subjects. For instance, participants used sensitive and complex language when they resisted accepting women’s freedom and rights. However, they did not directly share such ideas around their power and controlling behaviour in their language. These issues increased the challenges of translation. For instance, one participant said:

She was saying like, ‘you are leaving me and you come back whenever you want’ [dingonun ahırı gibi girip çıkıyorsun]. Can you imagine? Still, the British facilitator is telling me to avoid confrontation, yet she does not. She wants to work herself up and make herself angrier.

When the participants used Turkish proverbs, word-by-word translation was inappropriate because the meaning was lost, the readers’ understanding was reduced and this increased confusion around the extract. In the example below, the phrases in Turkish were presented and the meanings in English were given. This illustrated the complex language issues and increased the quality of the translation:

They are just following procedure [Kağıt kürek işleri işte]. In other words, you have this problem so take this, and then return this to me etc. . .It is not very reassuring. It does not make you relax psychologically.

It was sometimes challenging to present the participants’ voices when sensitive and complex topics arose. However, giving the meanings within the sentence by providing some words and explanations was helpful to clarify a quote, and thus increase the readers’ understandings. While translation techniques increase the readers’ understanding, the reflexive work is also conducted by the first author. For instance, researchers can increase their awareness of the participants’ social and cultural backgrounds and how they impact the data gathering and analysis processes when they apply reflexivity. Gawlewicz (2016) emphasises that migrant researchers are more likely to be aware of their participants’ culture in their research. Similarly, Temple (2008) notes that the quality of translation can be achieved by the researchers’ knowledge about the culture of participants and their professional understanding of the language. Gawlewicz (2016) suggests that ‘if translation is prepared with extra care so that conceptual equivalence is maintained, the quality of transcribed (and translated) data can remain high and fully appropriate for further analysis’ (p. 32). The examples of translation strategies with some extracts from the data were provided to illustrate how to reduce the challenges around translation and increase the readability of the extracts. When researchers become aware of translation strategies as well as challenges, they can present their findings in effective and appropriate ways. Therefore, detailed procedures on how to overcome the participants’ assumptions are needed to achieve conceptual equivalence.

Conclusion

The complex and unique task of translation in qualitative and sensitive research and the importance of reflexivity in the creation of high-quality translation have been presented. In the literature, discussions on the challenges and methodological issues of translation have been limited. This article has sought to understand how best to hear the voices and meanings of the participants in a translation. However, there are obstacles to translating the texts regarding the power of language. For instance, there were difficulties in translating the participants’ voices in their social and cultural contexts in sensitive and qualitative research. These issues were inextricably connected to the highly sensitive environment created by the relationship between the men and a female researcher. Adopting reflexivity was helpful to improve understanding and increase the awareness of insider and outsider positions and their influences on the data gathering and analysis process. While there are a small number of studies that examine the relationship between the researcher and participant in terms of translation of the data (Temple and Young, 2004), this article contributes new knowledge to the topic of translation of interview data within qualitative and sensitive research. Importantly, key strategies of translation including providing untranslatable phrases with the original language, back-translation, the awareness of the key circumstances, and positionalities, recognising indexicality of language, and supporting some explanations with brackets were discussed by highlighting how to provide effective and appropriate translation in a qualitative data. Therefore, a researcher involved in the translation process needs to take into account methodological questions and epistemological/ontological positions by being aware of their roles as a translator and a researcher to avoid losing the meanings of the text and voices of the participants.