The Interactional Features of Billingual Teacher Talk in the Classroom of Arabic as a Foreign Language

_______________________ *Correspondence Address: ahmad.basyori@radenintan.ac.id Abstract: Communication is essential in studying Arabic as a foreign language to make the lesson interactive. This research aimed to identify and assess three aspects. First, the Teacher Talk (TT) interaction feature appeared in the Arabic Education Study Program, Faculty of Tarbiyah and Teacher Training, Raden Intan State Islamic University of Lampung. Second, each lecturer's primary type of TT interaction element. Third, the utilization of code-switching as a component of bilingual learning interactions. This research was a case study of three Arabic lecturers (T1, T2, and T3) who used Google Meets to conduct online learning. Each lecturer was videotaped three times and then examined using Miles and Huberman's qualitative approach. The findings of the L2 investigation demonstrate that 1) T1 produced twelve different types of interaction features. Furthermore, T2 produced eleven different types of interaction features. Finally, T3 produced twelve different types of interaction features. 2) The most dominant TT interaction features in the classrooms by T1 was Scaffolding (18.2 percent), T2 was Display Questions (24.6 percent), and T3 was Confirmation Check (23.1 percent). 3) In the context of bilingual classes, this study also found that L2 lecturers used other interaction features to facilitate interaction, namely code-switching from Arabic (L2) to Indonesian (L1) and vice versa. The total occurrence of codeswitching by T1 was 9.1 percent (the least), code-switching by T2 was 27.3 percent, and code-switching by T3 was 63.6 percent (the most dominant). Code-switching can be an excellent alternate approach or interaction tool for facilitating communication in the classroom when learning Arabic; nevertheless, it must be used appropriately and proportionally.


INTRODUCTION
The researchers were initially interested in developing L2 learning from the standpoint of the learning method. However, because the learning method was not a solution in the learning process, the researchers shifted their focus on Teacher Talk (TT) in the interaction process in L2 classes. According to some research findings, TT appears in nearly 70% of classroom language because its function is not only for the class organization but also for language acquisition (McCarthy, 1991). TT on students is widely accepted as a source of input for language acquisition (Cullen, 1998). TT is an important tool in classroom activities for implementing lesson plans and meeting teaching objectives (Khusnaini, 2019). Students will learn more effectively if their teacher performs well in their role. According to Mercer (Mercer, 1995) in (Wasi'ah, 2016), teachers use their speech to do three things: first, to obtain relevant knowledge from students, so they can see what students already know and understand; second, to respond to what students said; and third, to describe the classroom experiences they shared with students in such a way that the teaching significance of the shared experiences was revealed and emphasized.
Language serves a dual purpose in language classes, including Arabic classes. It is a subject to be taught and an introduction to teaching. The language used in the classroom is a tool for learning new things. It is also the goal of research (Walsh, 2011).
The purpose of this research is to identify and analyze class interactions between lecturers and students in Arabic classes that are related to a variety of topics: 1) the types of interaction features in the TT that appear during the learning process, 2) the types of interaction features in the TT that are most commonly used during the learning process, and 3) the use of code-switching on TT in learning.

THEORETICAL SUPPORT Classroom Interaction
Because interaction is a key component of communicative language learning, communication plays an important role in making the classroom interactive (H. D. Brown, 2001). In other words, interaction is an important factor in the teaching and learning process because it involves the exchange of thoughts, feelings, or ideas with one another (Huriyah & Agustiani, 2018). The teacher's role in the classroom is to encourage interaction and guide students to become more interactive (H. D. Brown, 2001). Class interaction aims to exchange experiences, knowledge, experiences, and attitudes or to communicate and transfer them to influence the recipient's behavior (Kusuma & Nawawi, 2018).
Teacher Talk (TT) refers to verbal communication used by teachers (Lynch, 1996). TT is a distinct language that L2 teachers use to explain subjects in class (Sahlim, 2018). The TT grows due to its centrality in the learning process, which values the organization of relationships and meanings (Edwards & Westgate, 2005).
TT is critical to the success of L2 learning activities. The TT method and the interaction between the teacher and the students are critical to learning. A teacher who gives the speech should know how much and what type of speech should be given to help students develop the target language (Siti Astri et al., 2020). The primary goal of teaching is to change each student (Cazden, 2001).
It is common in the context of L2 learning for interactions in the classroom to move quickly, and understanding what happens during the process can be difficult (Tajeddin & Ghanbar, 2016). As a result, a teacher must understand how to increase students' roles and contributions to the learning process because the teacher serves as a role model for their students. Lei (2009) states that TT in the L2 class can sometimes be difficult for the language teacher. On the one hand, if the teacher speaks too much, students will be deprived of speaking in class. However, TT can effectively facilitate learning in the classroom and promote communicative interaction. Talk  L2 teachers often face several  challenges, including creating classroom  interactions  and  implementing  pedagogical objectives effectively (Seedhouse, 1997). These challenges can be overcome by understanding the types of features that occur in the classroom and facilitating classroom interactions. Understanding the context in the classroom through discourse analysis can explain classroom interaction practices (Walsh, 2013).

Interactional Features of Teacher
Generally, teachers still dominate the talk time in class (Menegale & Coonan, 2008). Teacher talk time, TT interactional features, and class modes are interrelated and influence. The frequent use of certain interaction features by the teacher will determine the amount of talk time and shape the class mode simultaneously (Murekson, 2017).
An important step for teachers is to maximize interaction features in the teaching process to improve classroom interaction (Walsh, 2013). Maximizing interaction in the classroom is an important part of the teacher's role, where interaction is part of planning the learning process. The current TT still follows the IRF pattern, which is in line with previous research. However, one thing that should be underlined is that the feedback part is sometimes omitted in the first round of IRF followed by the second round or when feedback is not needed (Jing & Jing, 2018). Walsh argues that there are four features of classroom interaction that represent context: control of communication patterns; elicitation techniques; improvement strategy; and delivery modification (Walsh, 2006) Researchers chose the types of interaction features that exist in the Self-Evaluation of Teacher Talk (SETT) framework developed by Steve Walsh (2011) to reflect TT in the learning process in the classroom. The SETT framework has 14 interaction features, namely: 1) scaffolding, 2) direct repair, 3) content feedback, 4) extended wait time, 5) referential questions, 6) seeking clarification, 7) confirmation checks,8) extended learner turn, 9 ) teacher echo, 10) teacher interruption, 11) extended teacher turn, 12) turn completion and 13) display question, and 14) form-focused feedback. Next, the researchers will explain code-switching as an alternative interaction feature in foreign language learning.

Bilingual Classroom
According to Weinreich (2010), bilingualism refers to using two languages interchangeably. This alternating use of languages can be in using a foreign language to a local/first language or vice versa, such as between Arabic (L2) and Indonesian (L1). L1 is commonly used in L2 classrooms in varying amounts and for a variety of functions, such as promoting classroom discipline, developing interpersonal relationships, increasing student confidence in the target language and comfort in the classroom, explaining grammar and vocabulary, and implementing and managing activities (A. Brown, 2021) The choice of language of instruction is a determining factor in-class interaction because the transmission of information between teachers and students is determined by the language chosen. In the context of learning Arabic as L2 in Indonesia, the use of L2 and L1 as the language of instruction should consider students' needs, conditions, and characteristics (Aulia & Kuzairi, 2020).
Bilingualism and code-switching are closely related. When the teacher performs code-switching, the bilingualism process occurs. When teachers see that their students do not understand what is being said, the process of bilingualism occurs through the use of code-switching (Mandang, 2019). Suwito in (Indrastuti, 1997), mengklasifikasikan alih kode menjadi classifies code switching into two types. First, internal code-switching occurs between languages in the country where the language is spoken, such as from Indonesian to Malay or vice versa. Second, external code-switching occurs between local languages and L2, such as switching from L1 to Arabic or vice versa.
There are many opinions regarding the use of L1 and L2 in learning. Among them is the opinion, which states that when the teacher speaks with L2, students will motivate themselves to use the language because they imitate their teacher (Raihani, 2011). The more students hear L2, the more they use it in their interactions and the greater their linguistic growth. This shows how important it is for a teacher to use the target language in classroom interactions (Met & Rhodes, 1990). However, Brown (2001) states that the use of L1 in L2 class situations is an option as long as the needs of L1 itself limit it.

METHOD
This study employs case studies to investigate real-world situations in the classroom. The goal is to investigate and comprehend social phenomena (Creswell, 2017), in this case, the TT phenomenon in Arabic language learning. Because the object under study is limited and unique in terms of using a combination of Arabic and Indonesian as the language of communication in the classroom, this study can reveal the phenomenon of bilingualism in greater depth.
Three Arabic language lecturers (T1, T2, and T3) who taught Arabic courses in class A, semester 5 of the PBA study program at UIN Raden Intan Lampung in 2020 participated in this study. The lesson was delivered online through Google Meets. The three lecturers were chosen because they were qualified and certified Arabic language teachers.
The researchers recorded three online meetings of each lecturer to capture interaction features from realworld Arabic classroom interactions using the SETT framework. After recording, the videos were transcribed and analyzed for features before compiled using the Miles and Huberman (1994) steps. The steps consisted of four activities: 1) data collection, 2) data reduction, 3) data presentation and 4) conclusion drawing/verification.

RESULT AND DISCUSSION
After recording three Arabic lecturers in 3 meetings, the recordings were transcribed and analyzed. The researchers found the use of interaction features by each lecturer in L2 learning as follows:

T1's Classroom Interaction Pattern
In detail, the TT interaction features by T1 in the L2 class can be seen in the Table 1.  Table 1 shows that T1 used 12 interaction features in the Arabic learning process in the classroom in 3 meetings. These features were 1) scaffolding, 2) direct repair, 3) content feedback, 4) extended wait-time, 5) seeking clarification, 6) confirmation checks, 7) extended learner turn, 8) teacher echo, 9) extended teacher turn, 10) turn completion, 11) display question and 12) Form-focused feedback. Meanwhile, the other two interaction features (referential questions and teacher interruptions) did not appear.
Of the 12 interactions, the most dominant interaction feature that appeared was scaffolding (58 times or 18.2%). Meanwhile, the feature that appears the least was teacher interruption (9 times or 2.8%).

T2's Classroom Interaction Pattern
The TT interaction features by T2 in the L2 class can be seen in the Table 2.  Table 2 shows that T2 used 11 interaction features in the Arabic learning process in class in 3 meetings, namely: 1) scaffolding, 2) content feedback, 3) extended wait-time, 4) Referential questions, 5) seeking clarification, 6) confirmation checks, 7) teacher echo, 8) extended teacher turn, 9) turn completion, 10) display question, and 11) formfocused feedback. Meanwhile, the other three interaction features (direct repair, extended learner turn, and teacher interruption) did not appear.
Of the 11 Interaction Features, the most dominant interaction feature that appears is the Display Question (84 times or 24.6%). Meanwhile, the least features are referential questions (6 times or 1.8%).

T3's Classroom Interaction Pattern
The TT interaction features by T3 in the L2 class can be seen in the Table 3.  Table 3 shows that T3 has used 12 interaction features in 3 meetings during the Arabic learning process in class. These features are 1) scaffolding, 2) content feedback, 3) extended wait-time, 4) Referential questions, 5) seeking clarification, 6) confirmation checks, 7) extended learner turn, 8) teacher interruption, 9) turn completion, 10) extended teacher turn, 11) display question, and 12) form-focused feedback. Meanwhile, two other interaction features (direct repair and teacher echo) did not appear.
Of the 12 Interaction Features, the most dominant interaction feature is Confirmation Check (43 times or 23.1%). Meanwhile, the least one that appears is Form-focused feedback (3 times or 1.6%).
The interaction features in the SETT framework developed by Walsh are based on English lessons in England (Korkut & Ertaş, 2017) which use monolingualism. Therefore, code-switching does not exist in the SETT framework. The context is different from learning Arabic in Indonesia, which is L2 learning, which allows code-switching from L1 to L2 or vice versa. The use of L1 will be adjusted to the conditions of the students.
Code-switching in L2 learning from the three Arabic language lecturers can be seen in the Table 4.  Table 4 displays that the lecturers did not fully use the target language in learning. The lecturers several times used code-switching as an interaction feature to facilitate interaction in the classroom. Of the total occurrences of code-switching in L2 learning, code-switching in T1 speech was 9.1% (the least). Code-switching in T2 speech was 27.3% and in T3 speech was 63.6% (the most frequent).
Furthermore, the comparison of the interaction features by Arabic lecturers based on the SETT framework and the code-switching feature in the bilingual class can be seen in Table 5.  The excerpt shows that lecturers directly used L2 in giving examples or pronunciation models for students to follow. This result is different from Prastiwi's (2015) findings, where the teacher, when assisting, usually catches the idea in L1, then he reformulates the speech model with L2.

2) Direct Repair
This interaction feature only appeared in the TT conducted by T1 in L2 learning in the classroom, as can be seen in the following excerpt:

Excerpt 2
From the excerpt above, T1 chose to directly justify the student's answers so that the interaction could continue. This finding is in line with research by Yu (2008), who states that the use of direct repair can advocate for students' self and cognitive development.

3) Content Feedback
This interaction feature appeared in L2 learning carried out by T1, T2, T3 in class, as can be seen in the following excerpt: The excerpt above shows that the lecturer gave input on the content of the L2 material after a student had previously expressed his opinion. These results support the statement of Aniwijayanti (2016) that teacher feedback is very important to determine a successful learning process.

4) Extended Wait-Time
This interaction feature only appeared in the TT conducted by T2 and T3 in L2 learning in the classroom, as can be seen in the following excerpt: The excerpt above shows that the lecturer gave students more pause to take turns. This result is in line with Shamsipour & Allami's ( 2012) findings that extended wait-time can increase student responses, leading to longer answers and more student contributions.

5) Referential Questions
This interaction feature only appeared in the TT conducted by T2 and T3 in L2 learning in the classroom, as can be seen in the following excerpt: The excerpt above shows that T2 asked students about things they didn't know yet, and students responded well to T2's questions. This result aligns with Rohmah's (2006) findings that referential questions effectively motivate students to respond to the teacher's conversation.
The excerpt above shows that the lecturer clarified the previous student's answer. This feature can provoke students to give reasons for his previous statements. This is in line with Khamwan's (2007) opinion, which states that requests for clarification can develop students' understanding and enrich their knowledge of L2.

7) Confirmation Checks
This interaction feature appeared in the TT conducted by T1, T2, and T3 in L2 learning in the classroom, as can be seen in the following excerpt: The excerpt above shows that the lecturer has checked whether the students hear his voice. Trisanti (2004) also states the same thing, that in L2 teaching and learning, the use of confirmation checks is very important to determine students' understanding and achievement.

8) Extended Learner Turn
This interaction feature only appeared in the TT conducted by T1 and T3 in L2 learning in the classroom, as can be seen in the following excerpt: In the excerpt above, a student ran a fairly long turn after being invited by T1. It aimed to train students' speaking skills in L2 (Valentika & Yulia, 2020).

9) Teacher Echo
This interaction feature appeared in the TT conducted by T1 and T2 in L2 learning in the classroom, as can be seen in the following excerpt: The extract above showed that T1 repeated his words to clarify the material presented. This repetition is done on many occasions by T1. However, this fact is different from the opinion of Walsh (2006) that teacher echo is used only in a few teaching opportunities.

10) Teacher Interruption
This interaction feature only appeared in the TT conducted by T3 in L2 learning in the classroom, as can be seen in the following excerpt: Excerpt 10 85 S1 : Fahashoni al-thobib, wa qala li:: anta bikhoir, (The doctor checked me up and said, "You are alright.") 86 T1 : Aina al-shout? Karrir! (Lauder? Repeat!) 87 S1 : Fahashoni al-thobib, wa qala li:: anta bikhoir, Ma al-musykilah? (The doctor checked me up and said, "You are alright.") The excerpt above shows that T1 interrupted the students' turn when they were practicing conversation in L2 because their voice was inaudible, and the lecturer asked to repeat the practice.

11) Extended Teacher Turn
This interaction feature appeared in the TT conducted by T1, T2, and T3 in L2 learning in the classroom, as can be seen in the following excerpt: The excerpt above shows that T3 explained instructions to one of the students to read the text. This result aligns with Rohmah's (2006) findings that teachers need to explain and provide appropriate explanations related to descriptive monologues before students get the opportunity.

12) Turn Completion
This interaction feature appeared in the TT conducted by T1, T2, and T3 in L2 learning in the classroom, as can be seen in the following excerpt:

Excerpt 12
187 S1 : Ughmiya 'alaihi wahuwa fi al-syarikah (She fainted when she was in the office) 188 T3 : Na'am, Tsumma! (Yes, and then?) 189 S1 Wa 'indama afaqa wajada nafsahu fi al-musytasfa (When she came to, she found herself in the hospital) From the excerpt above, T3 asked students to continue their turn. Turn completion occurs in L2 class when the teacher fills in the gaps and advances the discussion (Walsh, 2011).

13) Display Question
This interaction feature appeared in the TT conducted by T1, T2, and T3 in L2 learning in the classroom, as can be seen in the following excerpt: The excerpt above shows that T3 asked about someone discussed in the material, and students responded to the lecturer's questions. Display Questions can keep students participating in class discourse. This result is supported by Pujiastuti's (2013) opinion that the number of display questions posed by the teacher motivates students to respond.

14) Form-Focused Feedback
This interaction feature appeared in the TT conducted by T1, T2, and T3 in L2 learning in the classroom, as can be seen in the following excerpt: The excerpt above shows that T1 provided feedback regarding Arabic grammar. This result refutes the opinion of Suryati (2015), who states that Formfocused feedback offered by teachers tends to be accompanied by explanations that are generally given in L1.

15) Codeswitching
This interaction feature appeared in the TT conducted by T1, T2, and T3 in L2 learning in the classroom, as can be seen in the following excerpt: The excerpt above shows that T3 asks students to show a paragraph containing the correct answer using L2. Because students did not respond immediately, T3 switched to using L1 so that students immediately responded to their questions. This way, the interaction process can continue. These results follow the views of Mandang et al. (2018) that the use of code-switching in bilingualism must pay attention to students' abilities. If students do not respond or do not understand what the teacher is saying, the teacher can choose to use language easily understood by students. Code-switching is a choice, and its use is done consciously (Chaer & Agustina, 2004).

CONCLUSION
Based on the results and discussion, the findings of the L2 investigation demonstrate that 1) T1 produced twelve different types of interaction features. Furthermore, T2 produced eleven different types of interaction features. Finally, T3 produced twelve different types of interaction features. 2) The most dominant TT interaction features in the classrooms by T1 was Scaffolding (18.2 %), T2 was Display Questions (24.6% %), and T3 was Confirmation Check (23.1 %). 3) In the context of bilingual classes, this study also found that L2 lecturers used other interaction features to facilitate interaction, namely codeswitching from Arabic (L2) to Indonesian (L1) and vice versa. The total occurrence of code-switching by T1 was 9.1% (the least), code-switching by T2 was 27.3%, and code-switching by T3 was 63.6% (the most dominant). Code-switching can be an excellent alternate approach or interaction tool for facilitating communication in the classroom when learning Arabic; nevertheless, it must be used appropriately and proportionally. With code-switching, students can easily understand and quickly respond to questions or commands from the teacher. So that communication runs smoothly, and interactive goals can be achieved. This study has limitations, namely only revealing phenomena in the lecturer's speech. For further research, the researcher suggests revealing the use of code-switching or code-mixing with local or regional languages (L1) in student talk in learning Arabic (L2) in the context of bilingual or multilingual classes in Indonesia.