Methodology

The findings in this chapter have been derived from a broader case study which explores the translation phenomenon using the case of foreign nationals residing in East Japan at the time of the 2011 disaster. The study aims to better understand how these people gathered information and communicated during ^[1]

the disaster, to find out how linguistic mediation formed a part of these communicative activities, and to explore the importance of linguistic mediation in this context.

As a result of the constructivist philosophical underpinnings of the research project, the goal in participant recruitment was not randomised selection or sampling. The participants were introduced to the researcher by friends and associates or by the participants themselves. In all, 28 participants who identified themselves as a foreign national in a Japanese context and were resident in East Japan at the outbreak of the disaster agreed to participate in the case study. The participants represent 12 nationalities (Irish, Dutch, French, German, Sudanese, Tunisian, Chinese, Bangladeshi, American, Canadian, Australian, and New Zealander) who were resident in 5 different cities in disaster-affected prefectures (Sendai, Furukawa, Tokai, Mito, and Tokyo), who are in their 20s, 30s, 40s and 50s, possess a variety of linguistic abilities, and had varied occupations at the time of the disaster, including language teacher, student, engineer, diplomat, local government employee, company executive, office administrator, interpreter, consultant, and restaurant owner.

The face-to-face, individual interviews that make up the core primary data of this project took place in Ireland, Japan, and New Zealand between July and October 2013, but 25 of these 28 interviews were held in Japan during an intensive, four-week period of data-gathering from September 18 to October 13, 2013. Each interview was scheduled to take about 60 minutes, though the duration was at times shorter or longer depending on the interviewee's convenience. The interviews were largely a semi-structured, dialogic process; the interviewer framed the dialogue in each interview towards certain key topics (the participant's experience of the disaster, their communicative and linguistic needs, and their sense of community) while at the same time encouraging apparently unrelated anecdotes and tangents. It was hoped that these digressions would provide a deeper understanding of how each participant connected the various key topics together in her/his own life world. Ethical approval was granted by the relevant university research ethics committee, and all participants read a plain language statement and signed an informed consent form.

The number of interviews in this case study is not large, but this number was a function of the time and resources available to the researcher, the aims of the research, and the theories of knowledge and reality held by the communities to which the project is being addressed (Fielding 2003). In case studies, particularly those adopting a constructivist philosophy, the number of interviews should continue until multiple perspectives on the phenomenon under examination have been accessed (Stake 1995), and, where possible, until additional interviews are no longer providing significant, new data (Yin 2014). The diversity in participant profiles outlined above, along with the systematic interconnectedness in the interview data for this case study, convinced the author that such conditions had been met. It should be noted, too, that these 28 interviews took place some 30 months after the onset of the 2011 disaster. The timing of research interventions in disasters should not be considered in isolation, but in tandem with issues of safety, access to participants, the readiness of participants to agree to interview, and the ethical appropriateness of the intervention. Thus, while disaster-related interviews can be carried out anywhere from immediately to up to five years post-disaster, it is felt that the early recovery phase of a disaster (usually within a year of onset) is optimal (Stallings 2002, p. 83-86). In short, the timing of the interviews in this case study was less than optimal, but well within the bounds established in other disaster research.

The analytical strategy adopted for this project was a form of thematic analysis operationalised from Braun and Clarke (2006). In this operationalisation, themes were developed over six phases that progressed from participant-led to interpretive to abstract analysis. The first phase entailed multiple re-readings of hard copies of the interview transcripts to generate a list of potential codes that displayed features of interest to the research. The second phase involved creating a rule-for-inclusion for each potential code and then proceeding through the interview data (this time using QSR's NVivo 10 software) to highlight any passages that satisfied the rule. Once the coding had been completed for the interview data, the same coding rules were used to work through all the secondary data. These data included official reports and surveys on the disaster produced by various governmental bodies and research institutes, other grey literature on the disaster written by a number of emergency response and humanitarian organisations, and an illustrative corpus of actual communication dating back to the disaster period created by the author. The third phase involved combining the coded primary and secondary data into larger groups to produce themes. Again, a rule-for-inclusion process was used for this step. The main objectives of the fourth phase were to refine and further define the themes and to re-read the hard-copies of the data with these themes in mind. Following any necessary adjustments, the fifth phase introduced abstract analysis and involved the production of a thematic map and thesis argument linked to the relevant academic literature. The sixth and final phase involved the write-up of the report.

• [1] Community interpreting refers to the interpreting carried out in situations where members ofthe general public come into contact with providers of public services, such as the police, thejudiciary, and the health and social services (Wadensjo 2009, p. 43).