Collaborative Learning in a Japanese Language Course

This article describes ethnographic action research that explored experiences of the first author and her undergraduate students as they engaged in collaborative learning (CL) activities in a university Japanese language course. The purpose of the study was to generate new practical knowledge of CL for her, so that she might subsequently improve her teaching practice. A thematic analysis of the interview and descriptive data revealed that the incorporation of CL helped promote a comfortable environment and reduce the effects of the hierarchical authority of the instructor. While facing new challenges, the class also co-constructed its own knowledge about the reading content, language concepts, and cultural matters by working as a collaborative group. These findings are represented in the form of a performative text that invites readers to actively engage with the study’s findings.


CL
CL permeates the discourse of teaching and learning in higher education and is widely accepted as a legitimate pedagogy, especially by those who value peer-to-peer learning. For example, Dillenbourg (1999) defines CL as "a situation in which two or more people learn or attempt to learn something together" (p. 1). Bruffee (1984Bruffee ( , 1995Bruffee ( , 1999Bruffee ( , 2003, perhaps the best known proponent of CL in the higher education classroom, views CL as a kind of non-foundational learning process in which students work in groups or with members of knowledge communities that are larger and more complex than their own. The purpose of CL in this case is re-acculturation; that is, students change their membership from one learning culture to another through negotiating and modifying the language and values established from the original community and becoming fluent in the language of the new community. For example, a college-bound, high school graduate may have learned how to learn in a K-12 situation but may not have anticipated different ways of learning required in college, especially how knowledge is created. The aim of CL in this case would be for the student to transition successfully from his or her K-12 education into the broader academic community and the ways of knowing characteristic of the discipline he or she chooses to study. According to Bruffee, students in higher education should be expected to examine critically their taken-for-granted beliefs based on earlier, foundational learning experiences. They should also inquire with others who may hold different assumptions, beliefs, and values, all in the interest of broadening and growing their own knowledge. For Bruffee, this process is best supported by "abnormal discourse" (Rorty, Williams, & Bromwich, 1980) that enables students to establish knowledge or justify beliefs collaboratively by challenging each other's biases and presuppositions; by negotiating collaboratively toward new paradigms of perception, thought, feeling, and expression; and by joining larger, more experienced communities of knowledgeable peers through assenting to those communities' interests, values, language, and paradigms of perception and thought. (Bruffee, 1984, p. 646) Bruffee and other widely cited writers about CL (e.g., Barnes, 2004;Panitz, 1999;Rochelle & Teasley, 1995) are especially concerned that students come to value their own, collaboratively developed knowledge and rely less automatically on their assumptions about the authority of knowledge. This extends to discipline-specific knowledge, including knowledge held by an instructor considered a representative of his or her discipline. However, nearly all CL scholars position the instructor apart from their students in this regard and none identifies the instructor as a purposeful co-constructor of knowledge alongside the students (Johnson, Johnson, & Smith, 2007). Nevertheless, as Dillenbourg (1999) points out, this positioning by the instructor can profoundly affect the nature of students' interaction with them and other students. This brings into question how free students really are in terms of their ability to co-construct knowledge outside limits set by the instructor. Peters and Armstrong (1998) proposed a typology of teaching and learning that positions the instructor's relationship to students differently according to type of teaching and learning involved. Their typology consists of three types of teaching and learning. Type I (T-I) involves "teaching by transmission, learning by reception"; Type II (T-II) is called "teaching by transmission, learning by sharing"; and Type III (T-III) is "collaborative learning" (Peters & Armstrong, 1998, pp. 78-79).
Closely resembling Freire's (1970) banking concept, T-I represents a traditional type of teaching and learning. The focus of T-I is on the subject matter that reflects philosophy and expertise of the teacher and members of his or her discipline. The teacher is the primary source of information, and the flow of information primarily goes from teacher to student. Thus, the primary mode of discourse is monologue and the focus is on individual student learning. Examples of T-I include lecture accompanied by demonstration, drill, and repetition (Peters & Armstrong, 1998).
T-II involves more interaction among students who are seen as already having knowledge and experiences that they may use to help them develop new knowledge. As in T-I, the focus is on individual learning and the teacher still serves as the primary source of information. Information flows from teacher to student and student to teacher, and student to student as they learn by sharing within small groups. The primary mode of discourse is discussion. Lecture followed by discussions exemplifies this type (Peters & Armstrong, 1998).
In T-III, the teacher and students work together to co-construct new knowledge. The teacher becomes a member of the group, and communication flows from member to member, member to group as a whole, and group to member. Thus, individual members learn, and the group as a whole learns. This conception of CL is rooted in a social constructionist view of knowledge (Gergen, 1999). In this perspective, knowledge is generated within relationships from which humans derive their conceptions of what is real, rational, and good.
The primary mode of discourse in T-III is dialogue. Although they may have specific knowledge of the topic, the instructor is not presumed sole expert in the classroom. Students are expert in their own lives, including, in some cases, prior experience with the subject matter. Whereas in T-II the instructor facilitates works from the outside, in T-III he or she facilitates from the inside of the group. Group and individual members learn through critical reflection on past experience as well as what Schön (1983) calls reflection in action. Because of its interactive nature and inclusiveness, T-III appears strongly related to what Garrison, Anderson, and Archer (2010) call a community of inquiry (cf. Skinner & Peters, 2014).
Even though each type of teaching and learning is distinctive in terms of its purpose and other dynamics, the typology should be viewed as a set or grouping of types, not as a hierarchy, continuum, or stages of teaching and learning. One type is no more or less important than the others at any given time in the classroom experience. An instructor could use all three types across multiple class sessions or one, two, or three types in one session. The goal is to achieve a balance in types, with each serving in harmony with the other two (Skinner & Peters, 2012, 2014. Students and instructors already know how to act and what to expect of one another in T-I classrooms and, in some instances, T-II settings. However, studies show that T-III can be particularly difficult for students to learn how to negotiate successfully, especially in the beginning of a T-III experience (Armstrong, 1999;Burress, 2013;Creekmore, 2011;Crosse, 2001). It appears that students need time to adjust to different working relationships with each other and the instructor after a lifetime of mostly T-I and T-II classroom experiences (Burress, 2013). Instructors may also struggle with T-III in terms of their own goals and interests, such as the need to share responsibility for the learning process with students more so than in T-I and T-II learning (Cross, 1999).
According to a review of literature by Slavich and Zimbardo (2012), T-III helps students "restructure their own knowledge and understanding of concepts . . . recognize gaps in their understanding, synthesize, communicate, and discuss ideas in ways that advance conceptual understanding, (and engage in) modeling of effective problem solving strategies" (p. 574). T-III extends these benefits to instructors, yet little is said in the literature about the instructor's role as purposeful co-learner. However, it is clear that instructors have opportunities to reveal their ways of thinking and learning about the subject matter, including ways of knowing prevalent in their discipline. These opportunities allow instructors to model the processes of critical thinking, joint construction of knowledge, and dialogical communication, while facilitating students' participation in the process.
Other studies report positive learning experiences and outcomes in classroom settings across several disciplines. For example, Merrill (2003) examined her own and her student's experiences with CL in a community college information technology course. Her ethnographic analysis indicated that CL was helpful in creating a comfortable learning environment, building relationships between members, and developing an understanding of differing perspectives among participants. Similarly, Gray (2008) engaged her freshman composition class in CL alongside both T-I and T-II. A thematic analysis of her data revealed that engagement in CL helped her students gain confidence as writers, learn their writing strengths and weaknesses as well as those of others, and improve their writing skills. CL also provided increased opportunities for the creation of an open, comfortable, and safe classroom environment characterized by critical reflection and idea generation.
Finally, Li (2011) investigated how a group of her Chinese university students made meaning of their CL experiences as they engaged in an Intensive English reading course. The students were 19 to 20 years of age, and their English proficiency was intermediate to high. Based on a phenomenological analysis of students' interviews and class reflections and her field notes, Li concluded that the incorporation of CL helped promote positive interpersonal relationships among students, increase their confidence in speaking English, improve their engagement in learning, and change their thinking about teacher and student roles. Li's study is of particular interest because her course shares some features of the course that was the focus of the present study: for example, both courses involved learning of a second language, the introduction of an unfamiliar way of teaching and learning, and students who were at the intermediate level of language proficiency.
To our knowledge, no other research has investigated foreign language classroom settings, including Japanese, to examine student and teacher engagement in CL as defined in this article. Nevertheless, the aforementioned research and the first author's graduate studies in CL gave impetus to her decision to incorporate CL in the intermediate Japanese language course she facilitated in 2009 and to study how she and the students experienced the process. Her research question was "How do students and instructor experience the inclusion of CL (T-III) as one type of teaching in the intermediate Japanese language course?" The following is a discussion of how the study unfolded.

Context of the Study
The intermediate Japanese language course consisted of 26 undergraduate students, ages 19 and 20. The majority were native English speakers and were traditional students who enrolled in the university immediately after high school. The group met for a 50-min session 4 times a week for one 14-week academic semester. The instructor decided to add CL to her usual T-I and T-II repertoire. The latter types included lectures or presentations used to introduce new vocabulary and grammar patterns, as well as discussions of these grammar patterns and their usage and conjugations (lecture and drill sessions). The instructor reasoned that CL would not be appropriate pedagogy for these aspects of the course as they typically involved right or wrong answers and that T-I and T-II were better suited. On the other hand, the instructor reasoned that CL would be particularly applicable to sessions in which assigned readings, such as biographic literature, were discussed. Knowing that CL is a process that engages students and instructor in joint construction of new knowledge, it seemed especially suited to readings involving controversial issues and cultural matters that generally did not call for right or wrong answers. In addition, as her students had already attained the intermediate level of language proficiency, the instructor believed that she and her students could go beyond the development of basic language skills to explore and construct knowledge through readings and dialogue about the Japanese language and culture.
In the reading sessions of the course, students and instructor sat in a large circle and engaged in dialogue (Bohm, 1996;Isaacs, 1999), in both Japanese and English, about assigned readings that they previously read in small groups. To engage in dialogue means to be sensitive to what is happening in the moment of discourse, develop shared meanings, and act in ways that nurture participants' ways of being together, such as listening, reflecting, and valuing (Isaacs, 1999). Dialogue also involves mindfully attending to and cultivating knowledge that is being constructed by individual participants and jointly with others. Therefore, the instructor positioned herself as part of the group, and she and students assumed mutual responsibility for joint construction of new knowledge. The process of dialogue began with the instructor's open-ended questions to students, such as "What stood out to you about this reading?" "Which part of the reading did you find particularly striking or resonating?" and "Which part did you continue to struggle to understand?" Students were encouraged to ask their own open-ended questions to other students and the instructor. The group then considered these questions in terms of all aspects of the reading materials, including language, cultural matters, and specific contents of the readings. The questions asked by students and the instructor encouraged various responses and these responses led to additional questions, responses, and reflections. Instead of looking for predetermined answers or relying on the instructors' expertise, participants allowed discourse to emerge spontaneously from their moment-to-moment experiences and diverse voices. The instructor encouraged students to communicate in Japanese as much as possible to further develop their language skills.

Method
The instructor decided to study her own and her students' experiences with CL in the course for three reasons: (a) the experience was new to all participants, including the instructor, and she assumed that the most direct route to understanding participants' experience was to seek their first-person reports of the experience; (b) the action research by Gray (2008), Li (2011), andMerrill (2003) yielded data that helped improve their respective teaching practices, and it seemed reasonable to assume that the instructor in the present instance would benefit from a similar undertaking; and (c) her findings could contribute to related literature.
To examine the students' and her own experiences in the CL reading sessions, the instructor selected a qualitative research method known as performance ethnography (Conquergood, 1998). The purpose of utilizing this method was not only to capture how students experienced CL but also to represent the findings in a style more accessible than conventional approaches that rely on summary displays of findings.
Data sources consisted of the instructor's field notes and individual phenomenological interviews with volunteer students. Following Bogdan and Biklen's (2007) suggestion, the field notes consisted of two kinds of notations: descriptive and reflective. The instructor's field notes were used to portray the participants and the research setting, describe particular events and activities during the CL reading sessions, as well as capture the instructor's thoughts and feelings about the method, related ethical dilemmas and conflicts (e.g., relational concerns), and her frame of mind (e.g., her own assumptions and beliefs) while conducting the study.
All students were invited to participate in the study, and seven of them volunteered. This number of participants fell within the acceptable range of participants (n = 6-12) in phenomenological studies as established by Thomas and Pollio (2002). None of these participants had engaged in CL in classroom settings prior to this study. For the interview data, each of the seven volunteer study participants and instructor met for an approximately 1-hour, audio-recorded interview about each student's experience with the CL reading sessions. The interview followed a phenomenological procedure described by Thomas and Pollio (2002). It began with the question, "What was your experience with our CL reading sessions like?" This question was designed to make possible a broad range of descriptive responses while also focusing the students on the topic of their CL experiences. This was followed by some probing questions that were derived from the participants' responses to the first question and subsequent responses. Probing questions or requests (e.g., "Please say more about that . . . ") were designed to increase the richness of a participant's description of his or her experience. Instead of being in control of the direction of the interview, the instructor/interviewer approached the process from the "humble stance of perpetual learner" (Thomas, 2005, p. 73). Overall, the goal of the data collection was to obtain from the students a rich description of the CL experiences from their individual perspectives.
After the interviews were completed, the instructor transcribed interview recordings into a Word document for analysis. She then iteratively engaged with the data, reading and re-reading the transcripts for salient themes to the point of redundancy. Seeking to identify key themes of experience that surpassed the variations presented in individual accounts, she attended to broad patterns-first within each individual's data and then across the data set. In this process, in vivo codes (the actual words used by the informant) were utilized to demarcate shifts in meaning and describe areas of significance to the experience within the data, as well as sociologically constructed codes (this summary glosses over what the informant seemed to be describing) to identify recurring patterns and organize the data set (Coffey & Atkinson, 1996). Theme redundancy occurred after analysis of six transcripts; therefore, the seventh transcript was not analyzed. The instructor also read her field notes to gain additional insights into the contextual background of the themes. The second author assisted with the thematic analysis, providing an additional perspective and interpretation of the data.

Findings
The back-and-forth interpretation and dialogue resulted in identification of four themes: (a) learning environment, (b) process and outcomes of learning, (c) responsibilities for teaching and learning, and (d) resistance to engaging in dialogue. Below is a brief description of each of these themes.
1. Learning environment: The physical surroundings, psychological or emotional conditions, and social or cultural influences affecting the teaching and learning of a course participant and the group.

Process and outcomes of learning: Actions taken
(what and how) individually and collectively in an effort to achieve the goal of CL, that is, co-construction of new knowledge. What was gained, improved, constructed, and achieved as a result of the CL experience. 3. Responsibilities for teaching and learning: The authority and power one has to manage and direct the course of his or her learning. The relationship of power between the group members, both the students and the teacher, as a classroom community. 4. Resistance to engaging in dialogue: Hesitation, reluctance, or refusal of a participant to engage in class dialogue either in English or Japanese.
To afford greater access to the meaning of these themes, they were organized and represented in the form of a performative text (Madison, 2008). In this research context, performance refers to "an interpretive event involving actors, purposes, scripts, stories, stages, and interactions" (Denzin, 2003, p. 8). The text was constructed based on the thematic structure of the findings as well as the manner that students and instructor engaged in a dialogical conversation in the Japanese classroom. The text was edited to make it more readable than would be the case if verbatim quotes taken from transcripts and notes were used instead. However, the original voices of the participant interviews and in the instructor's field notes were maintained as much as possible. The purpose was to make the verbal exchanges as authentic as possible as the participants were imaginarily situated in a circle where they interwove their utterances into a dialogical conversation. Upon construction of the text, the relationship was ensured between the text and the findings by relating all parts of the text back to the themes to which they corresponded.
As the data were collected and analyzed, the authors practiced recursive reflexivity (Pillow, 2003). This proved especially helpful to the first author/instructor. Even though she attempted to suspend her assumptions and biases during the analysis, she also recognized that her relationship with the students might have affected various aspects of the study. For example, she ascribed grades in the course during the same time as the interviews. This is one instance in which the power differential between students and instructor had to be acknowledged. During the interviews, the instructor worked to minimize this effect by emphasizing to the participants her promise of no harm or penalties based on any aspect of their participation or lack of participation. Nevertheless, one cannot assume that this attempt completely freed the interviews from the power dynamic of the teacher-student relationship. Indeed, the study's findings and the performative text may represent only a "partial truth" (Clifford, 1986) as it is situated in this context.
The following script, based on the performative text, is intended to help readers experience CL in a way similar to what the students and instructor experienced in the classroom. We invite readers to engage actively with the text in multiple ways, such as by reading it silently or, preferably, by reading it aloud. The reader of this article might also enlist others to play various characters' parts while listening to the text. It may also help to imagine oneself playing the role of instructor, or as a director of the scenes presented in the text. The point is to find ways to engage in the experience represented by the performative text. Links to the participants' thematic experiences will be discussed later in this article.

A Performative Text
An introduction of the characters is provided below, followed by a script of their performance.

The Characters
Edward: A Caucasian male in the senior year of a Language and World Business program. He studied in Japan as an exchange student during his senior year of high school. Edward enjoys poems and writes them for self-expression, such as his determination and independence. Grace: An Asian American female in the junior year of a Studio Art program. She is also a member of the honors program. When leaving the classroom, Grace hopes to communicate better with her Japanese grandmother in Japanese. Kita: A Caucasian female in the senior year of a Linguistics program. She is the founder and organizer of the Japanese movie night group. Kita's effort to develop the language skillby speaking with her friends and writing diary entries in Japanese-is remarkable. Mike: A Caucasian male in the senior year of an Architecture program. He is also an honors student and is thrilled that his long-time dream, attending a graduate school in Tokyo, will finally become a reality in a few months. Rochelle: A Caucasian female in the junior year of a Language and World Business program. She is a close friend of Kita, and both of them are active members of the course. Rochelle finds time in her busy schedule to practice yoga and watch Asian dramas. Walter: A Caucasian male pursuing a double major in Theater and History. He attends pilot training to be certified to fly helicopters and looks forward to teaching English in China upon graduation. Meg: The teacher and a native speaker of Japanese.
As a doctoral student specializing in CL, she began practicing CL in her Japanese language courses and is learning to become a better facilitator of CL.

Act
Edward, Kita, Mike, and Walter are gathered in a classroom, talking about a TV show that was aired the prior evening as they await the arrival of others. They are here to have a dialogue about their experiences with the CL methodology in their Japanese course. Soon Meg enters the room, briefly greeting the students while also trying not to interrupt the ongoing conversation. After catching a breath, she glances at a watch on her wrist and asks the students to rearrange the tables and chairs to create a small circle conducive to dialogue. As they sit in a circle, Grace and Rochelle enter the room hastily and join the group. Meg then briefly describes the purpose of today's meeting and begins the dialogue by asking the group a question. 1 groups, instead of one large one? When we are in 251 small groups, maybe we have less pressure about making mistakes or looking stupid in front of 252 the whole class. And I think being in a small group can force everyone to engage; they have 253 more pressure to stay focused and contribute. 254 255 ROCHELLE: I think working in small groups has some advantages, but I really do feel having 256 you (Meg) with all of us throughout the processlike thinking and reflecting together-makes 257 CL different from other methods of teaching and learning. What do you think, Meg? 258 259 MEG: Yeah, you know, once we break into groups, I'm afraid I will be positioned outside of the 260 group, and this way, I will be participating mostly as teacher or facilitator rather than co-learner. first started out. I think our first reading was 287 Doraemon, and that was just telling us about something that we maybe didn't know about, and 288 that was about Doraemon, his TV show and manga, and how it started out. And after that, we did 289 the Yoko Ono piece. That was really nice because it brought together a lot of different things, like 290 abstract concepts that are common in poetry and literature. And how controversial all the things 291 Yoko did were at that time, not to mention the context of that time, which was the Vietnam war 292 and the hippie movement. 293 294 EDWARD: Yeah, I think the reading needs to be interpretive as well as informative. It was only 295 a few lines, but I really liked the poem in Yoko Ono's, the Cloud piece. I stopped and read this 296 poem just because it spawned millions of other ideas. I got a different meaning out of it every 297 time I read it. 298 299 MEG: Right, and I, too, have thought about the relationship between the reading content and our 300 engagement levels. And I agree that some changes need to be made about the choice of our 301 readings as they might be more or less suitable for class dialogue. 302 303 WALTER: It's great to know there are many things we can do to improve our CL experiences, 304 isn't it? 305 306 MEG: Yes, Walter. And thank you for your positive outlook. I also would like to thank you all 307 for participating and sharing today. I really enjoyed hearing, talking about, and learning from our 308 experiences. We addressed many important points, and these will be helpful in improving the 309 situations in our CL sessions. Our next CL reading session is scheduled for a week from now. So 310 let's practice again then with some changes and see what happens!

Discussion
The authors return to the themes and discuss them from the students' perspective based on their interview data and from the instructor's perspective based on her field notes and further reflections. Examples of themes are found in links to numbered pages in the performance text.

Learning Environment
Students' perspective. The learning environment stood out as an aspect of students' experience with CL (e.g., 3-17). Engaging in CL helped them create a sense of comfort and safety that in turn enabled them to learn about one another and build relationships over time. This finding corresponds to that of Merrill (2003), Gray (2008), and Li (2011), all of whom reported that the creation of an open, comfortable, and safe classroom environment and the promotion of positive relationships among their study participants contributed to meaning making by students and instructor. In this study, the comfortable environment, especially the positive reinforcements students received from one another and the instructor while learning to engage in CL, stood in sharp comparison with their former T-I and T-II classroom experiences, which tended to be competitive and focused on getting correct answers and grades.
Instructor's perspective. Closely related to creation of a comfortable learning environment was the instructor's attempt to reduce the effect of power differences that often exist between teacher and students in a classroom setting (e.g., 15-39). The instructor noted that with time, students became more relaxed about these differences. This finding is consistent with Peters and Armstrong's (1998) concept of CL and the role that dialogue plays in reducing the effects of hierarchical authority associated with T-I and T-II teaching and learning. This finding is also in line with Burress' (2013) discovery of the critical role that time plays in the formation of a strong group dynamic in the CL classroom.

Process and Outcomes of Learning
Students' perspective. A few student participants described their experiences with the CL process (e.g., 88-93) by employing metaphors, such as likening dialogue to playing music in a band (e.g., 95-98). On the whole, these metaphors represented the students' experience with the contributions of other group members as well as the emphasis the instructor placed on the learning process, for example, valuing of the "yet-to-be-created (knowledge)" (Peters & Gray, 2007), instead of seeking a specific, previously known answer to a question. As a result, instructor and students jointly constructed their own new knowledge about the reading content, language concepts, and cultural matters (e.g., 170-190). For example, as they engaged in dialogue and interwove their existing knowledge, they learned that phonetic reasoning behind yakudoshi may account for honyaku (great calamity) at the ages of 42 for men and 33 for women. These numbers are phonetically unlucky as 42 can be pronounced shi-ni, which is homophonous with the words "to die" or "death," and 33, when pronounced as sanzan, means "terrible" or "disastrous." This knowledge was not a simple reproduction of what participants knew in advance from the text or other related resources; instead, it was jointly constructed as they dialogued, taught, and learned with one another in the comfortable and safe environment that they had developed together. Dialoging on these issues further contributed to improved understanding of the content of the readings as it reinforced participants' need to review the materials and consider the issues multiple times, both as individuals and with the group as a whole (e.g., 129-131). In short, participants' experience with the CL process and the way they worked together enabled them to think and learn not only individually but also as a group.
Teacher's perspective. Working as a collaborative group, the participants engaged with one another as co-learners and coconstructors of knowledge by utilizing and interweaving their experiences, skills, and relationships and by assuming mutual responsibility for their constructive process (e.g., 100-128; 192-194). Excerpts from the instructor's reflective field notes illustrated the differences in this process and her previous experience with T-I and T-II.
I was sometimes astonished to see and experience how our engagement in dialogue could lead us to talking about the readings, language, and culture in a way that I did not expect or predict. We often created a new path of our teaching and learning, bringing in our own knowledge and skills and weaving them together into shared knowledge that was broad and nuanced. For example, the incorporation of CL helped me gain greater understanding about the readings, such as Yoko Ono's poem about counting and naming clouds, in a way that I could not have achieved alone . . . Each of the students and I brought in our own knowledge, experience, and talent, as well as our unique ways of approaching and understanding the poem and Yoko's life.
These same notes indicated that the instructor's experience helped her to learn more about the students. She came to realize that until she engaged in dialogue with her students that she did not know who they were-what they were interested in, what they already knew, and how thoughtful they were. There were times when she felt that she learned more about the students in one session than during an entire semester the previous year when she taught the same group without engaging with them in CL.

Responsibilities for Teaching and Learning
Students' perspective. Although the incorporation of CL produced positive outcomes, students faced some challenges associated with CL. They had to learn how to deal with responsibilities of helping to create and maintain conditions that support dialogue and co-construction of knowledge in the classroom, such as by taking an active role in the process (e.g., 41-80; see McNamee & Gergen, 1999). However, unlike the shift that occurred for Li's (2011) study participants in terms of their understanding of students' responsibilities, some students in the current study continued to expect to be taught, or hear correct answers, or learn "expert or native cultural perspectives" from the instructor only, and not from their classmates (e.g., 133-159).
Teacher's perspective. Similarly, the instructor often struggled to facilitate the group in a way that would engage all participants, students, and instructor, in dialogue. She had difficulty letting go of her control as a teacher and occasionally slipped out of her co-learner role and answered the questions that other members raised without searching with them for a newly constructed answer (e.g., 161-164). In addition, there were times when the teacher and students merely exchanged their stories and experiences and neglected to work together to generate new knowledge. This became a challenge for the instructor-a surprise, given her training in CL-and a new sense of respect for the complexities of T-III teaching and learning.

Resistance to Engaging in Dialogue
Students' perspective. As Li (2011) reported, the level of students' overall engagement increased with the incorporation of CL; however, in this study, some students continued to resist engaging in dialogue in English as well as in Japanese (e.g., 202-246). According to the students, their resistance stemmed from various sources, such as a fear of making mistakes and appearing stupid, a desire to avoid conflicts, and a concern with others' perceived disrespectful behaviors. To increase the level of all students' engagement in the process, students and the instructor continuously reminded themselves of the importance of developing and maintaining a dialogical space (Isaacs, 1999;Shotter, 2008) where all members are respected and respectful (e.g., 217-218).
Teacher's perspective. Students' resistance to playing new roles and increasing their engagement in learning was a concern not only to students but also to the instructor. Even when she called on students to speak, they often expressed that they did not have anything to say or did not know what to say in order to proceed. In response, the instructor engaged in frequent conversations with the students about how to encourage and help students to voluntarily and actively participate in the process (e.g., 248-301). However, students were more actively engaged at some times and less active at other times. This uneven engagement seemed to depend on the strength of appeal that particular readings had to students. Readings were more or less relevant to students' experiences, and some readings naturally elicited multiple interpretations or perhaps addressed more controversial and thus interesting issues than did others. Overall, the more relevant, controversial, and varied the readings, the more likely students were to engage in dialogue (e.g., 283-301).

Conclusion
One of the most widely discussed topics in contemporary higher education is how to get students more engaged in teaching and learning activities (Kahu, 2013). This usually means that instructors prefer students to be active learners, instead of passive learners. Active learning can be seen in the use of problem-based learning experiences, various other forms of experiential learning, and even question and answer episodes in the classroom. In the present study, it is clear that students and instructor were actively engaged through their discussion and dialogue about readings, even though the extent of their engagement seemed to vary with their interest in particular readings. Engagement in this case means thinking together (Isaacs, 1999), asking and responding to questions posed by peers and instructor, openly reflecting on the content of the readings as well as on reading assignments and engaging in dialogue about the readings. The very act of coconstructing new knowledge implies engagement, as working jointly to make meaning is in itself a form of engagement. This feature of T-III may be an obvious but overlooked area of student-to-student, student-to-teacher, and student-togroup engagement in higher education classrooms, as most literature on the subject has focused on T-I and T-II teaching and learning environments.
Although students in the present study were ages 19 and 20, some adult education scholars would identify them as adult learners about which much theorizing has been done (e.g., Brockett & Hiemstra, 1991;Cross, 1999;Knowles, 1980;Merriam, Caffarella, & Baumgartner, 2007;Mezirow, 1991;Mezirow & Associates, 2000). Knowles' work, in particular, set the stage for subsequent discourse about the need for instructors to involve students as active participants in the teaching and learning process. It is easy to see how Knowles' principles are applicable to T-II experiences. In fact, the latter in its lecture-discussion form may be the sine-qua-non of a Knowles teaching and learning experience. However, the experience that students get in a T-III teaching and learning setting takes the notion of active adult learning a step higher, as students and instructor labor together to construct new knowledge.
Even though their concept of CL seems bounded by what students can learn among themselves with input from the outside by an instructor, Bruffee and others who position the instructor on the outside nevertheless cite the potential of student-to-student engagement and active learning as catalysts for critical thinking. Peters and Armstrong's extension of this idea to include the instructor's direct involvement as co-learner places the whole of classroom participants in a position to think critically together about course content and process, moment-to-moment, during their classroom engagement. This attention to process and content has to do with metacognition, or thinking about thinking (Flavell, 1979;Hennessey, 1999;Kaplan, Silver, LaVaque-Manty, & Meizlish, 2013;Martinez, 2006), as much as it is to do with what they are thinking about. The students and instructor in this study were constantly thinking as critically about the topic of their discourse as they were about how they were engaging in the discourse. The instructor also reflected critically on her own facilitating moves and on students' responses to her moves.
So, how did the instructor and her students experience the inclusion of CL (T-III) as one type of teaching among others in the intermediate Japanese language course? Overall, their CL experience contributed to their individual and joint understanding of readings related to the Japanese language. However, CL was in many ways different, unfamiliar, and uncomfortable for students and instructor alike. This experience may have been due to the new way of learning that the particular process of CL, or T-III, presented for study participants. Students and instructors in general tend to take for granted how to act and what to expect of one another in T-I and T-II settings, and this was the case for participants in this study, at least in the beginning of their course. When it comes to T-III, students and instructors may be faced with learning a whole new way of engaging in teaching and learning (Peters & Armstrong, 1998;Skinner & Peters, 2012, 2014. For example, engaging in CL necessitates that students and instructor relate in multiple and complex ways; for example, as teacher, facilitator, co-learners, and co-constructors of knowledge. In this study, achieving a balance in these multiple ways of relating was a challenge. In spite of the instructor's attempts to demonstrate the value of all three types of teaching and learning, some student participants continued to expect to be taught in a more conventional manner and resisted aspects of T-III, especially participating in dialogue with the teacher and their peers. This resistance was exacerbated when the instructor also struggled to let go of her T-I and T-II pedagogies, in spite of having extensive training in CL. Finally, this research was undertaken in part because no other studies of CL in the Japanese language classroom exist. Based on this study, the authors feel comfortable in saying that CL (T-III) holds promise as a third approach to teaching and learning. This adds to the usual emphasis on practical communication and development of language skills (T-I and T-II) in a Japanese language course. The study also contributes to the limited research on Peters and Armstrong's typology of teaching and learning by virtue of its unique application in a Japanese language classroom. By and large, this study is consistent with other, differently situated studies of CL, thus adding to the level of support for the typology. However, because this study reports on only one course in Japanese on one university campus, the results may or may not be the same in other Japanese language classrooms or for other modern foreign languages. The authors suggest that similar research be undertaken in other modern foreign language settings, so that researchers may better understand what can be achieved by utilizing multiple types of teaching and learning in this area of undergraduate education.

Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding
The author(s) received no financial support for the research and/or authorship of this article. Note 1. Lines in the script are enumerated and keyed to the discussion that follows.