Silence: A duoethnography

This article explores the phenomenon of silence from the perspectives of two English language teachers, language teacher educators, and applied linguists. Utilizing duoethnography, which allows for people to become the sites of their own inquiry and investigate issues and phenomena of importance to them (Sawyer & Norris, 2013), we engaged in critical dialogues on the topic of silence from our varied perspectives, cultures, and backgrounds as a way to challenge broader normative assumptions, bring meaning to our experiences, and find solutions for some of our immediate teaching issues. Data were gathered from online conversations, reflective notes, and memos between the two authors with emerging themes including 1) Shifting perspectives on silence, 2) Silence in the online classroom, and 3) Turn-taking and student/teacher discourse expectations. While specific to the authors, it is hoped that this duoethnography may resonate with readers and serve as a catalyst for further reflection on silence. This is an open access article under the CC–BY-SA license.


Introduction
In this article, we explore silence through duoethnography. As English language teachers, language teacher educators, and applied linguists, silence plays a significant role in our professional lives. The language classroom is often a communicative environment and students are expected to speak in order to enhance their language acquisition and practice communication. However, silence also serves important functions for communication and language acquisition as there are varying types of silence (Bao, 2014;Harumi, 2011). Non-verbal activity can be interpreted in many different ways and have different meanings for different people depending on their experiences and background. This is potentially exacerbated in the field English language teaching (ELT) as ELT is a border-crossing field with students and teachers travelling to various contexts to teach/learn English. With these considerations and with the formation of this new journal focused on silence, we conducted an exploratory duoethnography into our experiences with silence from our varied perspectives and roles as English language teachers, language teacher educators, and applied linguists to reflect on our teaching practice and generate new understandings, for ourselves and potentially others, about silence in English language teaching (ELT).
In the following paragraphs, we present our duoethnography on silence. However, it is important to note that duoethnography eschews many 'traditional' research practices and norms (Norris, 2008). Because our duoethnography was exploratory in nature, we did not have research

Duoethnography
Duoethnography allows participants to be the sites of their own research and explore topics of interest to them in a dialogic manner. Duoethnography can serve as both research methodology and tool for reflective practice . As a research methodology, it can be used to counter normative assumptions and challenge grand narratives, while as a method of reflective practice, it allows practitioners to focus on an area of immediate need in their teaching . However, the distinguishment between duoethnography as methodology and as reflective practice tool is not always a clear delineation and duoethnographies can present both individualized counter-narratives as well as explore practical teaching issues (Karas et al., Forthcoming;. While there is not necessarily a 'wrong' way to conduct a duoethnography, dialogue is a crucial element and participants are encouraged to push one another to generate new understandings about the phenomenon under question. Duoethnographers can have research questions that they wish to explore (e.g., Banegas & Gerlach, 2021), or duoethnographers can eschew traditional research questions and explore their topic under question in a more fluid manner, allowing the investigation to develop naturally as the participants see fit.
In applied linguistics/English language teaching, numerous duoethnographies have been enacted across a variety of topics. Ahmed and Morgan (2021) explore the impact of postmemory on multilingual identity negotiation. Banegas and Gerlach (2021) investigate teacher identity and agency as the two educators engage in critical language teaching via comprehensive sexuality education in their English teaching. Native-speakerism (Lowe & Kisckowiak, 2016), English as an international language (Rose & Montikantiwong, 2018), native-speakerism and 'hidden curricula' in ELT training (Lowe & Lawrence, 2018), the Certificate in English Language Teaching to Adults (CELTA) (Huang & Karas, 2020), and other topics, have been explored by authors using the duoethnographic methodology. Other authors have used duoethnography explicitly as a reflective practice tool to address their various teaching issues (e.g., Schaefer & Brereton, 2020;Smart & Cook, 2020). The place of duoethnography as part of language teacher education programs has also been investigated (e.g., Karas et al., Forthcoming;Tjandra et al., 2020). These studies, and others, share duoethnography as the methodology/method utilized, but at the same time, display the flexibility with duoethnography allowing participants to personalize their investigation and carry out their research/reflections in the manner most beneficial to them.

Methodology
We began our duoethnography on silence by having recorded synchronous video discussions on Zoom. Our discussions broadly began by discussing our general impressions of silence and our experiences with silence as teachers/language learners. The dialogues were intentionally flexible and allowed us to discuss silence in ways that were most meaningful for us. These video discussions resulted in 4 hours of recorded conversation. While we were free to discuss silence as we wanted, we supplemented our discussions by reading academic articles/book chapters which served as artifacts for further reflection (Huang & Karas, 2020). In duoethnography, literature can serve a dual role as it can help situate the duoethnography in relation to other literature on the topic, which is common with other uses of literature in research, but literature can also take on a role of 'co-participant' and function as an artifact for further reflection (Huang & Karas, 2020). When functioning as an artifact, duoethnographers 'interact' with the literature and can bring the literature into discussions to enhance reflection. Our duoethnography began as an exploratory project into the phenomenon of silence. However, because Michael, the first author, had conducted research on silence and was familiar with literature, he proposed a book chapter (e.g., Bao, 2014, chapter 1) and journal articles (e.g., Ellwood & Nakane, 2009;King, 2013) that both authors read and discussed. These were selected because the first author felt they would help generate discussion and also because they focused on Japanese students. Our reactions and reflections on our conversations, the literature, and each other's reactions/reflections were shared in a Google Doc where we made notes as part of the main document and also used 'memos' to pass notes to one another and to note our reactions to each other's comments. Our conversations, reflections, notes on notes, and memos all served as 'data' for our study.
Duoethnography often does not follow a linear research pathway, meaning the borders between data collection, analysis, etc. are often not clear (Sawyer & Norris, 2013). Thus, as we 'analyzed' and reflected on our discussions and our notes, new 'data ' (i.e., interpretations, reflections etc.) were added as we further engaged with the topic. This allowed a more dynamic approach that could accommodate our constantly shifting perspectives. This flexibility and non-linearity continued into the writing up stage as we continued to reflect on silence and our experiences with silence.
Writing of the duoethnography was completed individually. After analyzing our discussions and noting prominent themes, the manuscript was passed back and forth between the two authors as writing occurred. As discussed, this allowed us further space to reflect and to add/clarify our statements as needed as we continued to engage with the topic of silence. There are many ways a duoethnography can be written and presented. Often, reconstructed dialogues are used because they are easier to read and also show the dialogic nature that lies at the core of duoethnography Sawyer & Norris, 2013). However, it should be emphasized that while duoethnographic articles are presented as completed manuscripts with 'clean' conversations, our duoethnographic process and our conversations were not as organized as we present them. Like much research, duoethnography can be 'messy'. However, we present them as dialogues because they more closely capture our thinking about silence and our experiences and allow the reader a window to our conversations. While these dialogues are not direct quotes, they are emblematic of the conversations we had and show how our current understandings have been impacted by our duoethnography.
Furthermore, we present reconstructed dialogues because it is important to think how a duoethnography may resonate with readers. This is a key tenet of duoethnography (Sawyer & Norris, 2013). A duoethnography does not end when it is published. Rather, readers can generate new understandings as they connect their own histories and experiences with the duoethnography they are engaged with (Ahmed & Morgan, 2021). Thus, while the dialogues can appear informal in comparison with traditional research norms, we hope they allow the reader a more authentic window into our conversations so they can generate their own understandings and interpretations, and perhaps use our duethnographic data a 'jumping off point' for their own reflections. As writers, we expect stakeholders and prospective readers to potentially inform our inquiry, and we acknowledge we prepared our duoethnography on silence with potential stakeholders and readers in mind. This allows us to acknowledge the meaning-making potential of stakeholders and consider how our own experiences and reflections may (or may not) resonate with members of our community of practice (Ahmed & Morgan, 2021).

Participants in Duoethnography
A key tenet of duoethnography is that participants need to have a level of trust and respect and can act as 'critical friends' . Without trust, duothnographers may be reluctant to share their true experiences or thoughts . Because of this, often, duoethnographies outline how participants know each other and offer information on participant backgrounds to further understand duoethnographers' positioning. In this spirit, we present brief biographies of ourselves below.
We met at Western University in Ontario, Canada while we were both completing our PhDs in applied linguistics. Michael, born and raised in Canada, spent 6 years teaching English/studying abroad in various places before commencing his doctoral studies. Takumi, born and raised in Japan, started his doctoral studies a year later after years of study abroad in the UK and United States. During our doctoral studies, we enrolled in different classes together and completed a meta-analysis together as well. Our varied backgrounds aided us in developing a trusting relationship together as we assisted each other throughout our doctoral studies. Takumi offered statistical assistance to Michael as he learned quantitative research methods, while Michael supported Takumi in his role as an upper year PhD student. Both of us have completed our doctoral studies and now work as lecturers, Michael in Canada and Takumi in Japan, but we continue to work on research projects focusing on high variability phonetic training (HVPT) and have weekly synchronous video meetings to discuss this work.
Michael has experience with silence as a researcher. When approached about the formation of this new journal focusing on silence, Michael sought out Takumi to complete a duoethnography on silence. This is partially because of the trusting relationship noted above, but also because it allowed him to learn about a new context, Japan, that he knows little about. For Takumi, it served as opportunity to expand on his quantitative research record and engage with a new qualitative methodology on a topic that would help his immediate English language teaching needs. Takumi was in a favorable position to contribute to discussions with regards to the roles of silence in a Japanese context, since he completed primary, secondary, and undergraduate university education in Japan and is familiar with Japanese educational contexts. Takumi also studied at an American university as an exchange undergraduate student (1 year), completed his master's degree in TESOL in the UK (1 year), and completed his doctoral degree in applied linguistics in Canada (4 years). His time overseas helped him discuss his experience with silence in Japan in tandem with (or in contrast to) his reflections about what roles silence had to play in Western contexts (i.e., USA, UK, and Canada). Takumi had never researched silence in educational contexts prior to the current study.

Findings
Duoethnographies can be presented in various ways. As discussed, due to the dialogic nature of duoethnography, many duoethnographies are presented with reconstructed dialogues (Sawyer & Norris, 2013). However, this can be done in various ways. For example, Rose and Montakantiwong (2018) present their duoethnography using extended individual narratives. Other authors present reconstructed dialogues that extend throughout the paper with subheadings, but with no introductions or conclusions to the conversations (e.g., Ahmed & Morgan, 2021;Tjandra et al., 2018). Another method is to present excerpts of the conversations with introductions and conclusions to situate the conversation and offer further explanation (e.g., Lawrence & Nagashima, 2020;Lowe & Kiczkowiak, 2016). This is the approach used for this article. Below, readers will find a brief introduction to our reconstructed dialogues, followed by the dialogue, then a reflective conclusion to situate the discussion.

Theme 1 -Shifting perspectives on silence
The first theme to emerge from our discussions focused on Takumi's shifting perspectives on silence. As discussed, Michael had conducted research studies on silence, but as a research topic, silence was not something Takumi had considered before this duoethnography. Thus, Takumi's views on silence shifted profoundly and quickly.
Michael: For me, silence is a topic I have done research on and reflected on. However, I'm not sure how much thought other people give to silence. What do you think when you reflect on silence?
Takumi: As an SLA (second language acquisition) researcher, silence for me has been a measure or indicator of fluency. Until we started this, I seldom thought about other elements of silence. When teaching before, I would try to reduce the silent time and transition time for students when they were speaking because unintended pauses can cause communication breakdowns. I always thought it's important to reduce the processing time, which is often done in silence. For me when teaching, the emphasis was always to get students to speak. Indeed, I implemented fluency training activities in the previous semester with the aim of improving students' ability to speak L2 English faster with fewer silent pauses and repairs.
Michael: How have your views changed?
Takumi: I've realized the importance of silence now. It's important to consider the place of silence in language education, especially for teachers. Teachers should be explicit about how they perceive silence in classrooms. Teachers need to be clear about the role of silence at the beginning and explain that silence is a necessary part of L2 speaking practice and it should not be seen as a marker of disfluency. In order to improve L2 speaking proficiency, it is natural to be silent and at times silence can be a marker of effort and an attempt to improve speaking skills. I feel strongly so especially after reading Ellwood and Nakane (2009), finding that there was a gap between students and teachers in how they perceived silence. It is important to clarify the way teachers perceive silence such that students feel assured that they are in a safe environment where they feel comfortable to speak.
Michael: That's interesting to hear your views have changed. It was similar for me when I began to read articles/books on silence. It's something that is so prominent and important for communication but something we think about so little.
Takumi: I think my previous views were connected to my experiences as a language learner. I'm impressed when other people who are not native speakers of English can speak without hesitancy and they don't seem to need long pauses of silence to speak. I still do sometimes. As a student learning English, I was very focused on speaking and I spoke a lot. I think that's a big part of my English language success. Because of this, I try to have my students do the same. However, I now appreciate the complexity of silence much more. It's important to push students to speak, but it can't be the only consideration.
Michael: Was speaking emphasized in your English language classes? Some of the readings we have done (e.g., Bao, 2014;Ellwood & Nakane, 2009) discuss the 'silent east' and a more 'talkative west' and note that there is often a dichotomizing between the two. I think we were guilty of this dichotomizing as well at first when we broadly discussed Japan and Canada. However, we don't really fit this stereotype. We have been together in many social and academic settings, and I find you usually speak much more often than I do and without much hesitation.
Takumi: I guess that is true. Bao's (2014) discussion about 'us' vs 'other' reminded me of real-life situations in Japanese classrooms. Especially in middle school and high school, students who are talkative stand out and can be perceived as 'other' because most students remain silent and do not speak until called upon. When I was preparing to go overseas to study, my teachers in Japan told me that I should try to speak a lot, so I really pushed myself to speak as much as I could. I don't like being silent in English. It can make me feel bad, like I don't have proficiency. In Japanese, it's different, but with English, I always try to speak as much as I can. I don't think I have a high aptitude for languages, but as I mentioned, I think this push to speak was a big reason for my success with English. Now though, I still believe it is important to push students to speak, but I'm more aware of the importance of silence as well.
This discussion highlight's Takumi's shifting perspective on silence as a measure of fluency to a deeper consideration of the complexities of silence, especially in the classroom environment. Part of this transition was related to our discussions and noting that despite some stereotypes that place Japanese English speakers as reticent and silent, our personal experiences were counter to this as Takumi often spoke more than Michael in their academic and social encounters. Furthermore, the reading of Ellwood and Nakane (2009), which highlights incongruent discourse expectations between Australian instructors and Japanese students in Australia, was highly impactful. Ellwood and Nakanes's (2009) study emerges again in our third theme and proved to be one of the key artifacts for this duoethnography.

Theme 2 -Silence in the online environment
As our conversations progressed, we moved into discussions about our current teaching situations. As mentioned, we both now teach English to university students, Takumi in Japan, and Michael in Canada. Due to the Covid-19 pandemic, our classes are all online, and silence in the online environment emerged as a key theme in or our discussions.
Michael: In the online environment, silence has taken on a whole new meaning for me. After re-reading chapter 1 of Bao (2014) and his discussion on the different types of silence, with our discussions and teaching online now, I started to notice all these different types of silence that I had not considered before. The easiest one is 'silence of muted microphone', when someone begins to speak but their microphone is muted. This is one I think most of us have experienced recently at some point in the last few months! Teaching pronunciation online I had students listen and repeat but with their microphones muted. This was another version of this silence for me. As the teacher, I would be watching all these screens with moving mouths but no sounds. It didn't allow me to hear students, but it did give them a chance to practice pronunciation and listen and repeat as a group but still be able to hear how they actually sound individually without being self-conscious. It was 'silent listen and repeat', as bizarre as that sounds.
Takumi: It is interesting to think about silence in the online setting versus the face-to-face (F2F) classroom. I tried to have a discussion test with my students in our online class, but it didn't go very well. They struggled to enter the conversation and most groups just nominated a leader to facilitate the discussion. It was filled with awkward silence, and I could see many of the students were at a loss when to break the silence and what to say. They needed more structure, which is common for many Japanese students.
Michael: I think that is common for many students and the online environment makes it worse. In face-to-face settings and online settings, I think there can be a tendency to emphasize inauthentic talk over authentic silence. This is one way they can be similar, but the inauthenticity seems more problematic online. For example, in the online class on SLA that I teach now, we have online chat forums to discuss course content. I have required students to post at least 5 times because if I didn't, I know they likely wouldn't contribute anything. I hate doing this but without it, they likely won't contribute to the classroom discussion, even with written posts.
Takumi: There can be that urge to quantify things. Students want to know what is required to get marks, and it can be more about the quantity instead of the quality. It's much harder in the online environment. Classes are recorded sometimes, and many students don't want to make mistakes, especially when there is a permanent record of it. Just saying 'speak more' doesn't really help. There has to be some type of structure. At the same time, interestingly, I found out that quite a few students prefer taking classes online. There are probably many reasons for this, but I think this partly relates to the way they like to learn things, such as listening to lectures rather than speaking and participating in discussions. So, it is unlikely that students will volunteer to articulate their opinions, and we have to push them to speak and ask questions.
Michael: It can feel very perfunctory, but it is necessary sometimes to get them to speak. For some students, without the structure or clear instructions, it can lead to 'silence of confusion', which King (2013) talks about. The structure can help avoid this type of silence I suppose. At times the talk feels genuine, but at other times, it feels less so. Silence of confusion is only one possibility though. They could be silent for many reasons, for example if they just don't like the topic, or are simply tired. It can be difficult to know what the cause of the silence/reticence is, and this is even more pronounced in the online setting. We can't see students' faces sometimes, can't see body language, it's much harder.
Takumi: Yes, silence is not monolithic. It's really important to create a safe environment where students feel comfortable to speak. This may partially depend on class size as well, but the online environment can make this much more complicated.
Michael: I found that with my classes when I put them into groups. For the pronunciation class I mentioned earlier, I would use breakout rooms and put them in to small groups to practice. Because it was a pronunciation class, all of the exercises were geared towards having them speak. In a F2F class, it can be easier to monitor multiple groups. I could often peripherally see when groups were struggling or not communicating and then try assist them. Silence can be revealing when students are put into groups for a speaking exercise, and no one talks. In the online setting with breakout rooms, it always surprised me going from room to room and finding one room filled with chatter, and another in total silence as the students opted to do their pronunciation exercise silently or with a muted microphone. Unlike the F2F class, there was no way of knowing how the group was doing until I was actually in the same breakout room.
Takumi: Yes, the 'unexpected breakout room silence'. When you put them into a speaking/pronunciation exercise, but they opt for silence. We have talked about how western methods/approaches can be improperly imported to different contexts. Task-based language teaching (TBLT), communicative language teaching (CLT), and then add in prominent theories like interaction hypothesis and output hypothesis, we see a real push to have students talk. It's interesting in our recent work with pronunciation instruction and specifically high variability phonetic training (see Thomson, 2018 for review), we see one area of research where there is still strong consideration of silence as actually better than talk sometimes. With pronunciation and ear training, sometimes talk does not add much to learning (Lee, Plonsky & Saito, 2020). Doing it silently can sometimes be as effective as listening with production practice.
Michael: Good point. Looking back, I prioritized talk because it was a pronunciation class, but I did give them silent time. I suppose I was startled when I saw how different tasks/activities can be done in such different ways. I imagine them doing it with a lot of talk, and I explain the instructions with talk as I want them to produce the sounds we were working on that day, but for some groups, they still did them silently. It was just more startling than in the F2F environment where it is all in front of you.
In this excerpt, different types of silence, especially in the online environment are discussed. This theme emerged and was influenced by our reading of Bao (2014) and King (2013) who discuss various types of silence. For Michael, when he was teaching English pronunciation online, new types of silence began to emerge for him and the complexity of silence in the online environment became even more apparent. His experience teaching pronunciation displayed the creativity students can use when completing tasks, as students at times opted to do exercises in silence, despite being encouraged to speak. This aligns well with Bao's (2014) proposal that it may be worthwhile for teachers to consider both speaking and silent options for some activities. For both of us, it also showed how we began to consider silence in relation to our work on pronunciation training/instruction, and how despite many areas of applied linguistics that push for talk, pronunciation remains an area of research (there are certainly others) where there is still strong consideration for silence and listening-focused practice.

Theme 3 -Turn-Taking and Student/Teacher Discourse Expectations
The final theme we present relates to turn-taking in the language classroom and the issues that can arise when there are different discourse expectations with teachers and students. Ellwood and Nakane's (2009) article about Japanese students studying in Australia again was impactful for this discussion.
Takumi: The Ellwood and Nakane (2009) article really resonated with me. The findings of this study are truly a reflection of my own experience being educated in Japan, and what I went through in English speaking countries for 6 years. Silence may occur because of different expectations on how the classroom environment is supposed to be. The difficulties between the Japanese students and the Australian teachers seemed to show this. Any teacher who is thinking of coming to Japan to teach English should read this article.
Michael: That's high praise! This article reminded me a bit about my teaching experience in China. It took a bit of time for me to better understand the communication habits of students. At times, this caused some misunderstandings. It was an EAP class of over 35 students and I wanted to have discussions. We would use pair work and small group work, but when it came time to discuss as a class, I was often met with silence. It was partially this experience that made me interested in silence and how turn-taking in the classroom is related to silence. Vol. 1, No. 1, November 2021, pp. 64-75 Karas, M. and Uchinara, T. (Silence: A duoethnography) Takumi: Turn-taking is an important skill that doesn't get enough attention in my view. However, this needs to be combined with more awareness on the part of teachers. The students in Ellwood and Nakane (2009) show this by noting some of the reasons they stayed silent in their classes. They preferred teacher-nomination, had difficulty with turn-taking, needed time to reflect before speaking, had a fear of making mistakes, and also did not want to be the centre of attention and stand out. I think the second and third of these, turn-taking and reflecting before speaking, can be practiced and improved.
Michael: I read this and put myself in the teachers' shoes. Nominating students is such a challenge for me still. I never want to embarrass a student. When I read the students' selfperceptions of silence in Ellwood and Nakane (2009), it made me think 'what's a teacher to do?'. The students preferred to be nominated by the teacher, but at the same time, they had a fear of making mistakes and did not want to be the centre of attention. If I am a teacher and I nominate a student, this puts them at risk of making mistakes and certainly puts them as the centre of attention. How can a teacher interpret all this?
Takumi: You're misunderstanding I believe. If a student is nominated by the teacher to speak, it is an external force. If the student makes a mistake or is embarrassed, it is the teacher's fault, not the student's. Hence the importance of nomination. When the teacher nominates the student, it takes control from the student. It may still be embarrassing if the student gives an incorrect answer, but at least the student can't be accused of trying to show off. The discussion here comes back to Bao's (2014) discussion about 'us' vs 'other' again. If a student nominates themselves and articulates their opinions a lot, they would run the risk of being seen as 'other' by their classmates. The student who stands out too much is considered 'smart,' 'diligent' or 'different', and not necessarily with positive connotations. 'Difference' has not been much appreciated in Japanese contexts where many schools still require students to wear the same uniform and carry the same bag, contrasting with Western contexts.
Michael: That's interesting. I didn't think of it like that.
In this excerpt, Ellwood and Nakane's (2009) article that highlights the incongruence between Australian teachers and Japanese students' views on proper discourse behaviour served as a main discussion point. The article strongly resonated with Takumi and his experience studying overseas in the U.K., United States, and Canada. It also resonated with Michael, as he compared some of his experiences with students in China. However, it resonated in different ways as the article sparked reflections about Takumi's past experiences as a student, but for Michael, it generated reflections about his teaching in China and his experiences as an English teacher today. The discussion highlights how discourse expectations can often be very different, and these differing expectations can cause interaction problems in the classroom (Poole, 2005). Michael's (mis)interpretation of the Japanese students' expectations is clarified by Takumi and the importance of teacher nomination is explained.

Discussion
Our duoethnography began as an exploratory project investigating silence through the lens of our shared roles as English language teachers, language teacher educators, and applied linguists, but also from our varying perspectives and backgrounds. Three prominent themes emerged from our duoethnography: 1. Shifting perspectives on silence, 2. Silence in the online environment, and 3. Turntaking and student/teacher discourse expectations.
The first reconstructed dialogue focused on Takumi's shifting perspectives on silence as this was the first time he had considered silence in depth from a research/pedagogical perspective. Takumi's perspectives on silence progressed and he began to view silence as a complex phenomenon, as opposed to only a measure of disfluency in speaking. This interaction highlights the importance of duoethnography as a methodology/reflective practice method. One of the strengths of duoethnography is that it allows participants to engage with a topic together, as opposed to more isolating forms of research (e.g., autoethnography) or reflective practice methods (e.g., narrative writing). Working with Michael, who has conducted research on silence, allowed Takumi to discuss the topic with someone who could assist with the reflective process. Because Michael had some knowledge about silence as a topic, he selected readings for further reflection and these, especially Nakane and Ellwood (2009), resonated with Takumi. While Michael's selection of the articles is biased to his preferences, it enhanced the duoethnography and allowed for more informed reflection.
The first interaction also highlights the importance of addressing silence, and teachers' beliefs about silence, in language teacher education programs. Takumi discussed his experiences as an English language learner and the strong push to speak, both from some teachers and himself, when interacting in the classroom. The dialogue demonstrates the impact our prior learning experiences can have on us as teachers, which Lortie (1975) described as the 'apprenticeship of observation'. As an English language learner, Takumi was pushed to speak and viewed this as success. However, because silence was not addressed from a pedagogical perspective in his teacher education, he had not considered the pedagogical elements of silence until this duoethnography. Michael, on the other hand, had more experience with silence as part of his LTE. Teachers' beliefs/knowledge (i.e., teacher cognition) are acknowledged as potentially impactful on teachers' practice (Borg, 2015); thus, it is important for silence to have a place in LTE programs as teachers' beliefs about silence may impact their classroom behaviour. Teachers' pedagogical beliefs, like other beliefs, can be difficult to impact and it should not be assumed that teachers' beliefs need to be changed (Li, 2010), but small developments can certainly occur (Borg, 2011). With a topic such as silence in the English language classroom, as was seen in the first dialogue, this could occur even more profoundly as it may be a topic, like it was for Takumi, that simply did not receive attention in comparison to other areas. The call to include silence as part of LTE programs has been made previously (e.g., Bao, 2014;Karas & Faez, 2020). Reflecting on silence as part of LTE programs could help prospective teachers further understand the complexity of silence as a phenomenon on its own, as opposed to just in relation to speaking, as was found in our initial discussions.
The second reconstructed dialogue looked at our discussions about silence in the online environment. Our discussion focused on how different types of silence seem to emerge in the online environment and how the push to have students speak in our online speaking/pronunciation classes emphasized 'inauthentic' talk where students are participating simply because of course requirements and not because they wish to contribute. Duran (2020) noted a similar issue as students felt 'pushed' to contribute with discussion posts. Students sought to "interact with course content" (p. 90) without having to interact with other classmates, but minimum speaking requirements, similar to our classrooms, pushed students to participate in ways they perhaps would not have normally. While our discussion noted that silence in the online environment can further exacerbate communication issues, it can also provide different opportunities. For example, Choi (2015) notes the difficulty that two Korean graduate students had speaking in their F2F classrooms in the United States, but Choi also emphasizes that both participants were far more active in online discussions where they could prepare and edit their statements. Bao (2014) and King (2013) note different types of silence, highlighting the complex nature of silence and many motivations/reasons why people do not talk sometimes. In the online environment, other types of silence emerge, as we noted in our discussions, and as other authors have noted long ago (e.g., Kalman, 2000). Further attention to silence in the online classroom, and the different types of silence in the online environment, will be useful.
The discussions about silence and pronunciation were also interesting. Pronunciation instruction/training became a key part of our dialogues because Michael taught pronunciation online and also because both of us are engaged with pronunciation as a research topic. Silent learning and active listening can be hugely important for improving perception and even spoken production (Lee et al., 2020). Our research on high variability phonetic training (HVPT) is another area where silent listening can be helpful to improve learner perceptions and productions of sounds. We don't necessarily advocate for only perception-based training, but the role of silence in pronunciation instruction is again an area we feel could be fruitful for further research.
The final topic focused on turn-taking and teacher/student discourse expectations in the classroom. Again, Ellwood and Nakane (2009) had a profound impact on our discussions. In our recreated dialogue, we discuss the difficulties teachers and students can have in the language classroom when they attempt to have interactions but have differing discourse expectations about how to engage in conversation. Ellwood and Nakane's (2009) discussion about Australian teachers and their misunderstanding of Japanese students resonated with both authors, but especially Takumi, who related many of his experiences studying in the U.K., USA and Canada. Other studies have noted this issue when teachers and students have different discourse expectations and the impact it may have on classroom discussion (e.g., Poole, 2005;Waring, 2013). The discourse issues in Ellwood and Nakane (2009) were partially attributed to the differing expectations of the Australian teachers and the Japanese students, but even with teachers from similar backgrounds, difficulties can occur. Looking at native-English speaking teachers (NESTs) and local Japanese English language teachers, Harumi (2011) notes that the NESTs had awareness of the "culturally oriented use of silence" (p. 265) with their Japanese students, and even had strategies to elicit student speaking, but did not have full awareness as to why this silence occurred. Contrastingly, the Japanese teachers often interpreted students silence as boredom, while this was not always the case, and they had fewer strategies to elicit student speaking. However, the Japanese teachers were more aware of students' preferred discourse patterns and their turn-taking preferences (e.g., nomination from the teacher) (Harumi, 2011). As these studies show, teachers need to be aware of their students' classroom discourse expectations and preferences. At the beginning of term, explicit conversations about classroom discourse expectations may be useful for both students and teachers alike to avoid misunderstandings and create a more a comfortable atmosphere.
Turn-taking is an important, but often underappreciated, skill that both teachers and learners must be aware of. Ellwood and Nakane (2009) noted students preferred to be nominated but also argue that is important for students to learn how to enter conversations themselves as this is an important skill to learn. The students in Harumi (2011) also expressed the desire for nomination, but a minority also expressed they did not want to be "singled out" (p. 264). This need for nomination expressed by many Japanese students combined with the fear of embarrassing students featured prominently in our discussions as we reflected on our experiences, Michael as a teacher, and Takumi as a student. Our discussions, as well has Harumi (2011) and Ellwood and Nakane (2009), highlight how difficult turntaking can be for students and the need for teachers to be aware of this. Beyond enhanced teacher awareness, it may be useful to perhaps 'teach' turn-taking and emphasize the importance of students learning how to enter the conversation on their own terms when they want (Karas, 2017). Furthermore, teachers can adopt many different strategies to make classroom discussions less stressful and more inviting for students to participate (e.g., give topics in advance) (Peng, 2015).

Conclusion
Our duoethnography on silence is specific to us; thus, results cannot be generalized. However, we hope our project can serve as a catalyst for further discussion, reflection, and research as readers perhaps connect their own experiences to ours and seek to generate new understandings. As a methodology and reflective practice tool, duoethnography allowed us to explore silence together and learn from each other. Indeed, duoethnography was effective as a reflective practice tool as we both found heightened awareness of silence and the needs of our students and have begun to incorporate this knowledge into our teaching. While the results are specific to us, we hope others will engage with duoethnography about topics of interest to them. Exploring topics through dialogue, even silence, can allow participants to be the 'sites' of their own research and present their understandings in an approachable manner accessible to both researchers and practitioners.