“Interview research is a craft, which is learned through practising interviewing.” (Kvale xviii)
There comes a moment when preparing and thinking about doing something has fulfilled its purpose and you need to do said thing; for me, moving from interviews in theory to conducting interviews in practice was a nerve-wracking experience. Even now, over 30 interviews later, I have a healthy respect for the situation and for the many in-the-moment decisions that need to be made in order to make the interview into a generative and ideally enjoyable experience. And I am lucky enough to get to say that mine have been wonderful encounters that I am grateful for. Following on from the first blog post about the Poetry Off the Page (PoP) interviews, this piece outlines many of the practicalities of conducting interviews as part of poetry performance research, including the stages of the interviewing process and the uses of these interviews in relation to the public.
The Interviewing Process
As a team, we discussed various facets of the interview process, both during our workshop with Rosa Reitsamer in November 2021 and during the development of our ‘critical path’. This document outlines the steps in our interviewing process, from approaching interviewees to organising and preparing for the interviews, conducting the interview, and processing it afterwards. Team members with interview experience shared their experiences with the rest of the team during this process. We also included our reflections about potential difficulties which could emerge in the process and how to address them. This guideline proved invaluable as more members joined the team in 2022 and 2023. Many of the points in the critical path are explored below.
The next step was to set the interviewing process in motion. As the opening quote by Steinar Kvale expresses, in order to learn how to interview, we had to actually approach interviewees, conduct interviews, and learn from those experiences. The active phase of the interview inquiry began in January 2022 when I conducted the first interview with the Irish poet Hazel Hogan via Zoom. This interview served as a trial run that we reflected on as a team afterwards. I was also the first person to conduct in-person interviews. In the spring of 2022, I interviewed 26 more individuals, both in person and online via Zoom, during a research stay in Ireland. As I was conducting these interviews, the initial interview guides were being revised, which accounts for some differences in questions between my early interviews and those conducted subsequently by myself and others.
Pre-Interview Procedure
Before interviews could take place, interviewees needed to be selected. This process was collaborative for those PoP team members focused on the UK spoken-word scene.[1] At various points throughout the project, researchers created lists of potential interviewees together, combining insights and interests from their own research strands with the overall aims of creating an oral history that included both well-recognised figures as well as more marginalised voices. There is no such thing as a complete (oral) history; nonetheless, we attempted to balance the perspectives in our collection by being attentive to differences in participants’ ages, periods of activity in the scene, regions of activity, performance styles, genders, and ethnicities.
My process of selecting interviewees was somewhat different from that of the rest of the team. Being the only researcher with a specific regional focus, namely on Northern Ireland and the Republic of Ireland, I was more independent in my choice of interview participants. During the initial months of my PhD project in 2021, I immersed myself in poetry performance recordings, read extensively, and created a comprehensive Excel sheet to track various performers and event series. Building on this foundational work, I began contacting poets who seemed to be central figures. Subsequently, I supplemented my judgement calls with the snowball sampling method (c.f. Noy), asking interviewees during the process to direct me towards other potential participants. As the events central to my work became clearer later in the project, I naturally interviewed participants involved in those events. Like the rest of the team, I also remained attentive to providing a broad sample of interviewees in relation to geographic and demographic features.
We had different approaches to contacting prospective interviewees, depending on whether we had previous contact with them before requesting an interview. If we did not have a connection to the participant, we sent an email based on a template which introduced the PoP project, explained our interest in speaking to them, and outlined the procedure. If the person expressed interest in the interview, we sent them the informed consent form and the participant questionnaire. We informed them that they could receive either the questions or the topic areas in advance if they wished, although we made it clear that they did not have to prepare anything ahead of time and were free to skip any questions they did not want to answer. We were unable to remunerate participants for their time, which we acknowledge made participation unfeasible for some. We are incredibly grateful to all those who donated their time to provide us with their invaluable contributions.
After this stage, the time and location were agreed upon. Considerations relating to the meeting place for in-person interviews were manifold. Ideally, interviews were conducted in quiet, neutral, safe, and easily accessible spaces. The best locations for audio recordings involved minimal outside noise and avoided echoes; however, since we lacked funding for studio recordings, small rooms in university buildings often turned out to be the best options in terms of sound quality. It was also crucial to consider the comfort and safety of both interviewees and interviewers when selecting a location. This ensured that interviewees did not have to travel extensively and that the spaces were suited to their needs. We conducted interviews in university rooms, libraries, cultural centres, and occasionally in cafés, although these were not ideal environments for sound. Other interviews involved travelling to participants’ homes when requested by the interviewee and deemed comfortable for the interviewer.
Before conducting interviews, each researcher performed additional research on the interviewee in relation to their work, including watching performances, reading their written works, and listening to other interviews, among other activities. As Ritchie recommends in his work on oral histories, “[i]nterviewers first get acquainted with the outline of interviewees’ lives and then allow them to fill in the details,” or more bluntly: “do your homework” (Ritchie 74). Based on our research, questions from the guide were adapted where necessary. Furthermore, the other members of the PoP team were informed about the interview and invited to submit specific questions they wanted the interviewer to ask. At this stage, we also identified questions that we considered particularly important based on prior research. This was essential because the interview guide was extensive, and if we had always posed every question, the interviews would have become unreasonably long; keeping them under two hours was challenging.
The Interview
On the days of the individual interviews, I reviewed the checklist that was part of the critical path. Depending on whether I was conducting the interview in person or online, the necessary steps I needed to take varied. For in-person interviews, I made sure that I had printed copies of the informed consent sheet and declaration of consent, as well as the interview guide. I usually brought a bottle of water for participants too. Importantly, I always double-checked the technical components: Did I have backup batteries for the audio recorder?[2] Was there sufficient space on the memory card? For Zoom-based interviews, I rechecked the technology: Was my internet connection stable? Had I sent the correct type of Zoom link (note: fully encrypted Zoom meetings cannot be recorded)? Sometimes, I triple-checked the recordings by conducting a trial recording with a friend before the interview. Preparing a list of last-minute checks was helpful, even if I couldn’t anticipate every circumstance. Then I waited at my computer or at the agreed-upon venue.
Once the interviewee arrived, either in person or online, the briefing phase began (cf. Kvale 55–56). In this phase, before starting the recording, I welcomed and thanked the interview participant, discussed the informed consent sheet with them, and answered any remaining questions before they signed it. In person, this was a relatively easy task. For virtual interviews, the signing of the documents usually took place before or after the interview. Processing paperwork digitally was sometimes challenging, as participants had different levels of familiarity with various technologies, which occasionally led to insufficiently filled-out documents or delays in response time. Technical challenges like this should be considered when planning an interview inquiry.
Often the briefing phase included explaining the origins of the PoP project and my connections to Ireland and poetry, as participants were interested in learning more about me. Sharing more about myself was essential for building a trusting relationship with the interviewees. I also detailed the recording process, either by explaining the Zoom procedure or by gesturing to the audio recorder positioned on the table between us. I informed the participants that they could pause the interview at any time for any reason and decline to answer any question. Moreover, mindful of the interviewees’ time, I made sure to check how much time they had for the interview, if this hadn’t been established previously. Once all these preliminaries were clarified, the interview officially began, and I started the recording.
The interview always began with the same introductory sentence: “This is [Interviewer’s Name] interviewing [Interviewee’s Name] on [Date] at [Time], and we are meeting at [Location].” After a brief welcome and thanks, I began with the introductory question about the interviewee’s history with poetry performance, which usually elicited a lengthy response that could then be built upon throughout the conversation. The rest of the interview followed the interview guide in terms of the questions and topics covered, but it was up to the researcher’s “judgement and tact how closely to stick to the guide and how much to follow up the interviewees’ answers and the new directions they may open up” (Kvale 56). For me, this freedom meant that I frequently changed the order of topics to accommodate the flow of conversation and asked follow-up and clarifying questions when appropriate or paused after an answer, since “[t]he quality of an interview relies not only on the questions posed; the way the interviewer reacts after an answer may be just as important,” as Kvale (65) reminds us.
In a similar vein, due to the nature of the qualitative interview as a conversation, it is the interviewer’s responsibility to make ethical decisions in the moment and navigate the conversation carefully (Kvale 29). This included sharing our own perspectives, allowing space for interviewees to express what they considered relevant, and responding sensitively to vulnerable acts of sharing without, as Jean Duncombe and Julie Jessop put it, “‘faking friendship’ in order to encourage the interviewee to open up” (119). As a team, we discussed various situations in advance that might require particularly sensitive responses or be especially challenging. Additionally, as a closely-knit team, we provided each other with a space to reflect on our interview experiences, the choices we made, and the questions that arose for us.
I usually ended the interviews by asking for suggestions of other people to talk to, as well as offering a final chance for poets to talk about their most recent and upcoming projects. This was followed by a thank-you, after which I stopped the recording.
After the interview recording was ended, there was a debriefing (Kvale 56), or a short period of informal chatting. During these moments, I reiterated my thanks, and the interviewees sometimes commented on how they felt about the process. When it felt appropriate, I inquired about potential archival material the interviewee had or had mentioned in the interview and whether I could receive access to it. Before parting ways, I told interviewees that they would hear from me in the future regarding the recording and that I would keep them updated on developments in the project.
Post-Interview Procedure
After the interview, the audio files were stored on a password-protected data server of the University of Vienna, accessible only to the PoP team. The researcher then listened to the recording before returning it to the interviewee via a password-protected online folder for their perusal. The interviewee had 14 days to decide how they wanted us to proceed with the data. Participants were informed about this in the consent sheet, which detailed the process as follows:
The post-interview emails sent to participants explained this procedure once more and included time stamps for passages that the individual researchers deemed especially worth re-listening to.
This procedure was developed to give participants control over their data (including the interview material itself), support our adherence to the General Data Protection Regulation (GDPR) of the European Union, and help us avoid potential libel claims. In accordance with the GDPR, we were transparent about the data we collected and its uses, and in alignment with the data minimization principle, we ensured that the data collected was related to the stated purpose (meaning, directly connected to the interview subject) (see also “GDPR” art. 5) . To ensure participant control, we provided the opportunity to approve the recording without additional changes (aside from those we deemed appropriate); they could have certain passages anonymised, allowing us to refer to them in publications without including those passages in the final public recording; they could have passages removed from the public files, and these would not be referenced by the PoP team in any publications; and, of course, they could withdraw their recording entirely.
We felt it was particularly important to give interviewees various ways of interacting with the recorded material, given the interview’s nature as an oral source and a more-or-less spontaneous offering of perspectives, insights, and stories. By providing these ways of interceding, we also hoped to retain as many interviews as possible.
In my experience across 30 interviews, most participants only asked for minor changes. None withdrew from the process, and almost none used the pseudonymisation clause. The changes requested often involved removing opinions they felt were not stated in a representative way, streamlining their stories and arguments, removing inaccurate information, or smoothing over certain passages.
For the interviewer, the cutting process meant that while listening to the interview recording, we took notes on passages that we felt were particularly important to flag for participants to re-listen to, as well as those we wanted to exclude from the final file. Regarding the former, I often highlighted passages to interviewees when they had made comments during the recording that seemed to signal a reluctance to have this material published (e.g., “off the record”) or when they had shared something they might reconsider having disclosed so candidly. Concerning the passages we as researchers wanted to remove, these decisions were based on the GDPR data minimisation principle, potential libel, and other instances we judged to be sensitive. In practice, there were not many instances in which this type of intercession was necessary. The most common situation in my interviews was the use of incorrect pronouns for other poets. This was pointed out to the participants, and with their consent, we attempted to correct these instances by splicing sentences or removing the relevant passages. In my experience, most of the cuts I made were related to interruptions during the interview itself, such as closing windows, getting glasses of water, taking short breaks, or long pauses. In a few cases, I removed passages that referred to personal matters concerning third parties when the information shared seemed too private to be shared by the interviewee without the third person’s consent.
In addition to our editing process, which involved listening to our interviews carefully, we also created numerous supplemental text documents for our research. During the period when only my interviews had been conducted, I matched each of my questions with a timestamp in writing and added rough notes about the interviewee’s responses. At this point, we began thinking about transcripts, which had not been part of our original critical path but would surely make our interviews more easily searchable for our future research. In spring 2022, Marie Krebs, our research assistant, and I spent many hours transcribing two of my interviews in detail for our second workshop. However, we quickly discovered that this resource-intensive practice was not sustainable, given our aim of conducting 120 interviews for PoP. Therefore, we attempted to utilise speech-to-text software to create transcripts; however, this also proved rather difficult. Most options involved a breach of GDPR (data protection) regulations or did not produce transcripts of remotely usable quality. Eventually, the English Department’s IT expert at the time, Alexander Kurmayer, was able to create a private transcription server based on a speech recognition model called ‘Vosk’, which compiled semi-accurate text transcripts. We later progressed to using Microsoft Azure’s Batch Speech Recognition at Kurmayer’s recommendation, which further improved the quality of our transcriptions. These required only minimal editing, either by the interviewers themselves, sometimes in dialogue with the interviewees, or by Marie Krebs, which created helpful supplemental documents for each researcher. When it came to the actual interpretation and incorporation of interview material, however, we frequently returned to the audio files themselves to preserve the oral quality of our history and to pick up on nuances that cannot be perceived through written text alone.
We revisit our interviews often, especially when they become relevant to a smaller research idea, such as a conference paper, a podcast episode, a journal article, or a book chapter. For instance, the Poetry Off the Page Podcast serves as an ideal platform for sharing extracts from the interviews because the audio files can be integrated, allowing all the nuances of spoken language to come through. Incorporating interview extracts not only lets the interviewees speak for themselves but also creates a compelling listening experience.
In our written work, integrating interview quotes requires a more complex approach, as it involves a medium change. As Linda Shopes demonstrates in her work on oral history publishing,
editing for publication aims at making the spoken word […] accessible to the reader, more or less according to the conventions of written language. Yet it aims at doing so in a manner that remains faithful to the oral, to the narrator’s words and word order, speech patterns and rhythm, as well as to the sense of what she is trying to say and the way that sense unfolds (472–73).
While Shopes discusses longer publications that undergo an extensive editing process, we can nonetheless take the above passage and see how it applies to PoP’s work. When we identify a passage we want to include in our written work, we return to the audio recording along with the partial transcript and re-transcribe it, including the relevant timestamp. During the final quotation process, we attempt to clarify meaning and address the imbalance between our own academic writing and the spontaneous spoken utterances of the interviewees. This involves removingcertain elements of spoken language, such as false starts and filler words, as well as adding punctuation to the transcribed passage. Before submitting this work for publication, as per the guidelines described in the informed consent form, participants are offered a clearance read of their passages; this includes a brief contextualisation of the passage followed by the transcribed version of what they said. In this way, we adhere to best practice models regarding participant control (Shopes 472).
The Interviews and the Public
Our interviewing process is still ongoing at the time of publication. However, we have been working with the completed interviews since mid-2022 and have been able to begin publishing both our research and the interviews themselves. I have already shown how the interviews have been integrated within our research publications. In the following section, I explore the functions of the interviews as a tether between ourselves as researchers and the scene(s) we are researching, as well as the process of preserving the interviews for posterity.
Fostering Relationships
As previously indicated, one function of the interviews was to foster contact with the communities at the centre of our research. Conducting interviews during our research stays in the UK and Ireland has led to a significant connection to the scenes we are researching. As Kat François mentioned in the consultation session, it is vital to spend time connecting with the people whose work we analyse and, in a way, (re)present in our academic work. In early 2022, when I spent two months in Ireland, the poetry performance scene was gradually emerging from the Covid-19 pandemic lockdowns. I attended as many performances as possible, but it was the interviews that were vital in involving me and informing me about the poetry performance scene in Ireland. The interviewing process accelerated my understanding and knowledge of the scene beyond what would have been possible through participation or reading alone, especially since there is little published scholarship on the Irish spoken word scene. The interviews were a way of reaching out to the communities and giving them opportunities to shape the (academic) conversation about the spheres in which they work and create.
Another benefit of the interviews was their ability to foster extended relationships and contact with participants, which led to more trust. Shefali Banerji, the second PhD researcher on the team, has explained how conducting the interviews and building relationships with poets was essential. Through these connections, Banerji was granted access to archival recordings of spoken word theatre shows, which were rarely found in institutions or official archives. Thus, the interviews served an additional purpose, helping them to create a more extensive corpus. An upcoming blog post by them will discuss the issue of archiving spoken word theatre soon!
Preservation & Access
Providing access to the interview material has always been central to the PoP project, as it aims to serve the poetry performance community, researchers, and the wider public. As I mentioned in my previous blog post, interviews play a vital role in preserving the history of this ephemeral art form. However, many interviews conducted by other poetry performance researchers in the UK remain inaccessible for various reasons. In contrast, the PoP team has already begun publishing our recordings with UK poets through the University of Vienna’s digital repository, Phaidra, where they are publicly accessible. On the website, interviews are attributed to the participants by name, assigned to the appropriate researcher, and include metadata about the interview circumstances along with keywords related to the content. In the near future, links to these recordings will also be featured in Apples and Snakes’ Spoken Word Archive. The interviews I have conducted with poets working primarily in Ireland and Northern Ireland will be available through the National Library of Ireland’s Born Digital Collection. Those already cited in publications have been temporarily made accessible via a folder on the University of Vienna’s cloud service.
A final note on the public availability of the transcripts: As indicated above, the machine-generated transcripts are only partial and frequently inaccurate. Making them publicly accessible would require not only more resources than the project possesses, but also re-approval from the interviewees, as this was not included in our consent form. We acknowledge that not publishing transcripts does mean more work for future researchers and that some people may lack the time to listen to the interviews, leading them to exclude these sources from their research, which is unfortunate yet understandable. The absence of transcripts also means that we must trust in future users’ (ethical) standards of transcription and citation (Good 462). All this said, there is an additional reason for not publishing transcripts, a reason which resonates with our approach to our primary objects of study: the recordings of poetry performances. Poetry research has long been dominated by the written word; our project is an intervention into this practice, aiming to create literary history that attends to poetry as an oral medium. In our performance analyses, we focus on the audiotext and body communication, asserting the importance of the produced sound and the orality and aurality of poetry in performance. In our interview work, we have chosen to work with and make available the oral sources themselves instead of a written transcript. In both cases – poetry performance and oral interviews – transcripts can never capture all the nuances of the originals (Good 459; Novak 125–26). Therefore, we encourage engagement with the oral sources in all their particularities, as we believe that the experience of listening to voices is essential to poetry performance research.
***
CLAIRE PALZER is a PhD researcher at the University of Vienna in the Poetry Off the Page project. Her doctoral work focuses on spoken word poetry in Ireland from the 1990s to the present day and the particularities of this performative and situated art form. She has conducted over thirty interviews to date with participants in the Irish poetry performance scene. Other research interests include Irish Studies more broadly, historical fiction, cultural memory, feminist, and queer studies. She is also a published poet.
***
To cite this blog post:
Palzer, Claire. “Implementing the Poetry Off the Page Interviews.” Poetry Off the Page, PUBLICATION DATE, LINK.
***
References
Duncombe, Jean, and Julie Jessop. “‘Doing Rapport’ and the Ethics of ‘Faking Friendship.’” Ethics in Qualitative Research, edited by Melanie Mauthner et al., Sage, 2002, pp. 108–21, https://study.sagepub.com/sites/default/files/Duncombe%20&%20Jessop.pdf.
Good, Francis. “Voice, Ear and Text: Words, Meaning and Transcription.” The Oral History Reader, edited by Robert Perks and Alistair Thomson, Routledge, 2015, pp. 458–69. www-taylorfrancis-com.uaccess.univie.ac.at, https://doi.org/10.4324/9781315671833-34.
Kvale, Steinar. Doing Interviews. SAGE, 2010.
Novak, Julia. Live Poetry: An Integrated Approach to Poetry in Performance. Rodpoi, 2011.
Noy, Chaim. “Sampling Knowledge: The Hermeneutics of Snowball Sampling in Qualitative Research.” International Journal of Social Research Methodology, vol. 11, no. 4, Oct. 2008, pp. 327–44. DOI.org (Crossref), https://doi.org/10.1080/13645570701401305.
“Regulation (EU) 2016/ 679 of the European Parliament and of the Council of 27 April 2016 on the Protection of Natural Persons with Regard to the Processing of Personal Data and on the Free Movement of Such Data, and Repealing Directive 95/46/EC (General Data Protection Regulation).” Official Journal of the European Union, no. L 119, May 2016, pp. 1–88.
Ritchie, Donald A. Doing Oral History. Oxford University Press, 2014, https://web-p-ebscohost-com.uaccess.univie.ac.at/ehost/ebookviewer/ebook/bmxlYmtfXzgzOTEwOV9fQU41?sid=cf712377-4122-49cb-8765-f3f688ea71ba@redis&vid=0&format=EB&lpid=lp_xi&rid=0.
Shopes, Linda. “Editing Oral History for Publication.” The Oral History Reader, edited by Robert Perks and Alistair Thomson, Routledge, 2015, pp. 470–89. www-taylorfrancis-com.uaccess.univie.ac.at, https://doi.org/10.4324/9781315671833-35.
[1] Shefali Banerji, Rachel Bolle (who was not part of the team at the time), Julia Lajta-Novak, Martina Pfeiler, Shalini Sengupta, Helen Thomas (who also joined PoP later on), and Emily K. Timms.
[2] After consulting Jörg Mühlhans from the University of Vienna’s Faculty of Philological and Cultural Studies’ MediaLab, the PoP project purchased Zoom Hn2 recorders for the in-person interviews. Mühlhans also did workshops with the team on how to use these audio recorders and how to create the best possible recording conditions.