Interviewing practitioners

Semi-structured interviews (Kvale & Brinkmann, 2009) were used to include administrators’ and musician-teachers’ experienced meanings and perspectives. This was done with the understanding that knowledge was socially constructed through the

์๒ It was explained to me that although the piano is a western instrument, even in the schools promoting Nepali folk music, piano/keyboard/melodica is taught due to its relationship to the harmonium.

interactions between myself as interviewer and the interviewee (Kvale & Brinkmann, 2009), and the others participating in the interview, such as the research assistant who acted as a translator during musician-teacher interviews or the co-author of Article I.

I conducted semi-structured interviews with eight school administrators (e.g.

founders, chairmen, directors or principals) from the six private schools in which I was observing music teaching and described above. The primary intent of interviewing administrators from these schools was to inquire into the general background of the schools and the role of music within them (see Interview Guide in Appendix 10).

0RUHRYHUDIWHUWKHVFKRROVSHFL¿FVRQJSUDFWLFHHPHUJHGDVWKHIRFXVRI$UWLFOH I, administrators’ knowledge of the history of the song in their school and their perspectives on the song were essential. As this practice had not been a focus during P\¿UVWYLVLWWZRRIWKHWKUHHDGPLQLVWUDWRUV¹⁸ interviewed then were re-interviewed during my second visit exclusively about this practice. In addition, one school founder and one composer who was suggested to us as he was known to have composed VFKRROVSHFL¿FVRQJVZHUHDOVRLQWHUYLHZHGH[FOXVLYHO\DERXWWKHVFKRROVSHFL¿FVRQJ practice. 7KHIRXULQWHUYLHZVH[FOXVLYHO\DERXWWKHVFKRROVSHFL¿FVRQJSUDFWLFHZHUH conducted with the co-author of Article I, Professor Heidi Westerlund. Administrator interviews were conducted in English in a location of the administrators’ choice in their schools.

I also held semi-structured interviews with seven musician-teachers¹⁹ from three of the private schools I had been visiting. Time restraints during my second visit were the cause of this restricted school representation. However, all of the interviewed PXVLFLDQWHDFKHUVKDGH[SHULHQFHWHDFKLQJPXVLFLQRWKHUVFKRROV2QHKDGWDXJKWLQ DWOHDVWGL൵HUHQWVFKRROVDQGRWKHUVPHQWLRQHGWKDWDWWKHWLPHRIWKHLQWHUYLHZ they were teaching part-time in four or seven other schools. Some also taught in homes. In addition, musician-teacher interviews began to demonstrate saturation UHJDUGLQJWKHVFKRROVSHFL¿FVRQJSUDFWLFH$UWLFOH,DQGDVVHVVPHQW$UWLFOH,, DQGZHUHWKXVGHHPHGVX൶FLHQW7KHLQWHUYLHZHGPXVLFLDQWHDFKHUVKDGDUDQJHRI teaching experience, with at least four mentioning having taught 12 to 16 years prior to the interview. They taught compulsory or optional (extra- and co-curricular) music lessons and their teaching included classroom music, singing, folk dance, bansuri,

์๓:HZHUHXQDEOHWRFRQWDFWRQHRIWKHVFKRROVDJDLQDIWHUP\¿UVWYLVLWVRWKHDGPLQLVWUDWRUIURPWKLV VFKRROZDVQRWUHLQWHUYLHZHGUHJDUGLQJWKHVFKRROVSHFL¿FVRQJSUDFWLFH7KLVDOVRSDUWO\H[SODLQVWKH GL൵HUHQFHLQWKHQXPEHURIDGPLQLVWUDWRULQWHUYLHZVFRQWULEXWLQJWR$UWLFOHV,DQG,,VHH)LJXUH

์๔ In an attempt to maintain anonymity, information about the teachers is combined here rather than presented individually. For the same reason, the schools in which they taught at the time of the interview

pianica, sarangi (a bowed chordophone), and Nepali percussion. In addition to their WHDFKLQJDOORIWKHPZRUNHGDVSHUIRUPHUVVRPHKDYLQJUHOHDVHGDOEXPV2QHDOVR LGHQWL¿HGDVDFRPSRVHU0DQ\DOVRGHVFULEHGKDYLQJZRQYDULRXVFRPSHWLWLRQVHLWKHU as a performer or director/teacher working with students. At least three noted how their work as musicians had provided them with opportunities to travel and perform RXWVLGHRI1HSDO7KHVWRULHVRIKRZWKH\EHFDPHPXVLFLDQVGL൵HUHG6RPHFDPHIURP families of musicians–one identifying at least 5 generations of musicians in his family–

and began learning music in childhood. These musician-teachers described informal OHDUQLQJVLWXDWLRQVVXFKDVRQH¶VIDWKHUDVD¿UVWPXVLFWHDFKHUOHDUQLQJDQLQVWUXPHQW by listening to the radio as no one in the family or village played it, and participating in the community bhajan events. They also described more formal instruction like taking daily lessons during school holidays and on Saturdays while school was in session, studying with a classical singing teacher, or studying music in India. The musician-teachers who did not have other musicians in their families described primarily formal approaches to learning music, such as lessons, or pursuing a diploma or bachelor’s degree. Some also described informal learning. All of the musician-teachers were asked to teach based on their success as performers, and with the exception of one musician-teacher none described having had any music teacher education. The one H[FHSWLRQKDGSDUWLFLSDWHGLQPXVLFWHDFKHUWUDLQLQJR൵HUHGE\RQHRIWKHPXVLF LQVWLWXWHVLQ.DWKPDQGX2QHRIWKHPXVLFLDQWHDFKHUVKDGEHHQWHDFKLQJDGL൵HUHQW subject in schools prior to beginning to teach music. Most, however, were performers ZKROHDUQHGRQWKHMRE2QHPXVLFLDQWHDFKHUWKXVXQGHUVFRUHGWKHLPSRUWDQFHRIWKH mentoring received from the person who asked him to start teaching. In addition to teaching in private schools, at least one of the interviewed musician-teachers said he had worked in 3 or 4 government schools on a short-term basis, for example, being hired for two months to prepare students for a competition.

Interviews with musician-teachers from the schools in which I had been

observing deepened the understandings being developed through observation, allowing me to ask questions about themes not always easily observable. In line with the

methodological framework of this research project, the interview guide for musician-teacher interviews was created by adapting the generic AI questions (Cooperrider et al., 2005, p. 25; Watkins, Mohr, & Kelly, 2011, pp. 155–156), with additional positively-framed questions related to the themes of the research project such as diversity, the VFKRROVSHFL¿FVRQJSUDFWLFHDQGDVVHVVPHQWSUDFWLFHVVHH$SSHQGL[0\GHFLVLRQ to frame musician-teacher interviews using AI was also related to my attempt to

emphasize my stance, which regards the musician-teachers as experts, and to counter

any feelings that this cross-cultural research project involved foreign evaluation.

The interviews were semi-structured in the sense that I followed the interview guide ÀH[LEO\IUHHO\SXUVXLQJDGGLWLRQDOWKHPHVDQGIROORZXSTXHVWLRQVDVQHHGHG7KH

¿UVWPXVLFLDQWHDFKHULQWHUYLHZVHUYHGDVDSLORW,QDGGLWLRQWRWKHSLORWOHDGLQJPH to revise the interview questions, it also reinforced the need to encourage musician-teachers to speak Nepali, rather than English, so that they could express themselves with greater detail and ease. Interviews with musician-teachers took place in a location agreed upon by the musician-teacher and research assistant, either at their school or in my hotel lobby. They ranged from approximately 30 to 75 minutes.

Transcription

Transcription varied for administrator and musician-teacher interviews, due to the ODQJXDJHLQZKLFKWKHLQWHUYLHZVZHUHFRQGXFWHG$GPLQLVWUDWRULQWHUYLHZVZHUH¿UVW WUDQVFULEHGE\H[WHUQDOWUDQVFULEHUVLQ)LQODQGZLWKZKRP,KDGDFRQ¿GHQWLDOLW\

DJUHHPHQWDQGWKHQYHUL¿HGE\PHE\FDUHIXOO\UHOLVWHQLQJWRWKHDXGLRUHFRUGLQJV while reading and, when necessary, correcting the transcripts. The transcripts were then sent to the administrators to provide them with the opportunity to read, comment on, and make changes or elaborations to their individual transcripts (Bresler, 1996). My LQVWUXFWLRQVZKHQVKDULQJWKHVHWUDQVFULSWVKRZHYHUPD\KDYHEHHQLQVX൶FLHQWDV one of the administrators also corrected the wording of my questions in a few instances when the questions and responses did not clearly match. In total there were 102 pages of transcriptions with the 8 administrators.

%HFDXVHWKHPXVLFLDQWHDFKHULQWHUYLHZVZHUHLQ1HSDOLWKH\ZHUH¿UVW transcribed and then translated by the research assistant. In instances where words or FRQFHSWVZHUHGL൶FXOWWRWUDQVODWHLQWR(QJOLVK,DVNHGKHUWRUHWDLQWKHRULJLQDO1HSDOL DQGSURYLGHVRPHH[SODQDWLRQRIWKHLUPHDQLQJVLQ(QJOLVK/LDPSXWWRQJ2QFH the interviews were transcribed and translated, we met to discuss them, including her perspectives and interpretations, as well as the transcription process (Liamputtong, 2010). As this research assistant was not a musician or a teacher, I decided to have WKHWUDQVFULSWLRQVDQGWUDQVODWLRQVYHUL¿HGE\3UHP*XUXQJ²⁰, the musician-teacher who was hired in 2016 to work with me as a co-facilitator (see Section 4.3). Like the UHVHDUFKDVVLVWDQWKHDOVRVLJQHGDFRQ¿GHQWLDOLW\DJUHHPHQWEHIRUHFRPPHQFLQJRXU work together. In addition to verifying the translations, this process provided him the

ํ๋ Prem Gurung’s name is used with his permission, and to give him credit for his immense contribution to

opportunity to familiarise himself with my earlier work and prepare for his work in the research project. There were also 102 pages of transcriptions in total from the 7 musician-teacher interviews and 1 composer interview.

Analysis

Interview transcripts served two purposes in this research project: preparation for later stages and reporting. Rereading the transcripts helped me prepare for Stage Two, by revisiting the various themes or issues that had arisen. Moreover, I selected a number of anonymous quotes from the musician-teacher interviews to share during WKH¿UVWZRUNVKRSVLQ6WDJH7ZRDVDZD\RIUHSRUWLQJEDFNWRWKHFRPPXQLW\DQGRI stimulating discussion. In preparing Articles I and II, interview transcripts were read DQGFRGHGIRUUHODWHGWKHPHV7KHFRGLQJFRXOGEHFRQVLGHUHG¿UVWFRQFHSWGULYHQ (Kvale & Brinkmann, 2009) in that certain themes were selected ahead of time based on the planned articles. Thus, close attention was given to interview excerpts related to DVVHVVPHQWSUDFWLFHVWKHYDOXDWLRQRIPXVLFLQVFKRROVDQGWKHVFKRROVSHFL¿FVRQJ practice. Data-driven coding (Kvale & Brinkmann, 2009) was then used to further develop the codes through careful re-readings of the interview transcripts. While I was the author primarily responsible for the initial analysis and interpretation, having been the main interviewer in all the interviews, relevant and extended interview excerpts were anonymized and shared with co-authors to enable co-interpretation. All of this was done with awareness that the knowledge produced during the interviews was further coloured by the transcription, analysis, and reporting that followed (Kvale &

Brinkmann, 2009).

Ethical considerations

Some of the ethical issues attended to in relation to conducting these interviews ZHUHLQIRUPHGFRQVHQWFRQ¿GHQWLDOLW\FRQVLGHULQJWKHSRVVLEOHFRQVHTXHQFHVRI participation, and the researcher’s role (Kvale & Brinkmann, 2009). Information letters and consent forms were provided for and signed by all interviewees in both English and Nepali (Liamputtong, 2010, see Appendices 8 and 9²¹). I did not discuss the names of the schools I was visiting or individuals I was interviewing with anyone not present in the interviews. The only exception was co-facilitator Prem Gurung (see Section 4.3) who checked the musician-teacher interview transcriptions. Prior to doing so, however,

ํ์7KHFRQVHQWIRUPVIRUPXVLFLDQWHDFKHUVFKDQJHGEHWZHHQ6WDJHV2QHDQG7ZRDVDUHVXOWRIWKH application process to the university’s ethical board. I felt that the recommendations in their application form were better than those I had followed previously.

ZHGLVFXVVHGFRQ¿GHQWLDOLW\DQGDVDOUHDG\PHQWLRQHGKHVLJQHGDFRQ¿GHQWLDOLW\

agreement. Although there were no foreseen negative consequences for participation in the interviews, I aimed to ensure that all interviewees were presented respectfully when reporting on the research project, and to protect their anonymity to the extent possible, considering the small size of the music community in Kathmandu. For this reason, as described above, the schools and musician-teachers are not presented individually, nor are associations made between musician-teachers and schools. I also aimed for the interviews to be a positive experience for the interviewees, being aware WKDWIRUPDQ\WKLVZDVWKH¿UVWWLPHWKH\KDGEHHQLQWHUYLHZHGDVSDUWRIDUHVHDUFK project, and using AI to frame the musician-teacher interviews as one way of doing so. I also remained aware of the power asymmetry of interview situations (Kvale &

Brinkmann, 2009). The interviews were primarily one-way and instrumental dialogue, and I and my co-authors had monopoly over the interpretation. At the same time, I was aware of the interviewees’ power in choosing how and what to share, acknowledging also the potential for being told what interviewees believed I wanted to hear (Kvale &

Brinkmann, 2009). Moreover, in at least one case the interviewee spoke at great length with few pauses, making it challenging for me to guide the conversation towards my interview questions. Due to the cross-cultural nature of the interviewes, ethical issues also arose from the process of translation. To address these, I had Prem Gurung verify the translations, and whenever quoting a musician-teacher the words were always taken from the translation of their words in the transcript, not the transcription of the translators in the moment translation. Finally, one unforeseen ethical issue that arose was that without my prior knowledge, one of the interviews occurred at a very challenging time emotionally for the interviewee. Due to personal issues, rather than the content of the interview, the interviewee began to cry. This was very challenging for me as a researcher as I felt unprepared for the situation. All I could think to do was R൵HUWRVWRSWKHLQWHUYLHZKRZHYHUWKHLQWHUYLHZHHZLVKHGWRFRQWLQXH

4.2.4 Documentation

In addition to observations and interviews, policy documents were also used as empirical material. In particular, in Article II government policy documents including the National Curriculum Framework for School Education in Nepal (NCF, Government of Nepal, 2007) and the Nepali Music Curricula documents for grades 1-5, 6-8, and 9-10 were used. These documents were not analysed in any systematic way, rather they were read thoroughly with particular attention to the article’s themes.