An Analysis of College English Classroom Discourse

English classroom teaching and learning is an important aspect of English teaching and learning. The analysis of classroom discourse is a very important form which classroom process research has taken. The present study focuses on college English classroom discourse. Through a detailed description and analysis of the collected data by referring to Sinclair and Coulthard’s classroom discourse analysis model, the problem of patterns of the classroom discourse is made clear and on the basis of which a few strategies for college English teachers are put forward by the author in order to improve college English teaching and learning.


Introduction
Discourse is defined as "the language in use" (Cook, 1989:6) and discourse analysis is concerned with the "the analysis of language in use" (Brown and Yule, 1983:1).Discourse analysis refers to the study of the relationship between language and the contexts in which it is used.It involves looking at both language form and language function and includes the study of both spoken interaction and written texts.It identifies linguistic features that characterize different genres as well as social and cultural factors that aid in our interpretation and understanding of different texts and types of talk.
Discourse analysis helps us in understanding how real people use real language, as opposed to studying artificially created sentences.It is therefore of immediate interest to language teachers because we need to consider how people use language when we design teaching materials, or when we engage learners in exercises and activities aimed at making them proficient users of their target language, or when we evaluate a piece of commercially published material before deciding to use it (Michael McCarthy, 2002) Since spoken language is "the medium by which much teaching takes place and in which students demonstrate to teachers much of what they have learned" (Cazden,1987, cited from Wittrock,1988),the application of discourse analysis to second language teaching and learning can reveal much about how teachers can improve their teaching practices by investigating actual language use in the classroom, and how students can learn language through exposure to different types of discourse .Classroom discourse analysis is an aspect of classroom process research, which is one way for teachers to monitor both the quantity and quality of students' output.Nunan has pointed that "If we want to enrich our understanding of language learning and teaching, we need to spend time looking in classroom" (Nunan, 1989:76).According to researchers of communicative teaching, "failed communication is a joint responsibility and not the fault of speaker or listeners.Similarly, successful communication is an accomplishment jointly achieved and acknowledged" (Jack C. Richards and Theodore S. Rodgers, 2000: 77); "It is the teachers' responsibility to organize the classroom as a setting for communication and communicative activities" (Jack C. Richards and Theodore S. Rodgers, 2000:78 ).In recent years, a much greater role has been attributed to interactive features of classroom behaviors, such as "turn-taking, questioning and answering, negotiation of meaning, and feedback" (Chaudron, 1988:10).The background of this lies in the fact that "second language learning is a highly interactive process" (Richard and Lockhart, 1996:138) and "the quality of this interaction is thought to have a considerable influence on learning" (Ellis, 1985, cited from Richard andLockhart, 1996:138).In second language classrooms, "learners often do not have a great number of tools…, teachers' questions provide necessary stepping stones to communication" (Brown, 1994a: 165).Questioning is reported as one of the commonly used strategies, as the success of a class largely depends on questioning and feedback.One reason, as Mercer (Candlin & Mercer, 2001:245) states, is that they form the most frequent model of teacher-student talk in the classroom, in terms of the model described by Sinclair and Coulthard as Initiation-Response-Feedback (IRF) exchanges.
More fundamentally, however, it is because they are a teacher's best instruments to regulate the quantity and quality of language used in the classroom.
The problem which is going to be mainly investigated here is patterns of classroom discourse.Because "two of the most common ways in which L2 teachers engage in interaction with learners is by way of asking questions and providing feedback, and these deserve some consideration" (Holland and Shortall, 1997:104), focusing on them can be expected to show useful findings which will contribute to deeper insights about the ways to improve L2 teaching and learning.
The context specified here is a college English classroom of non-English major.College English refers to English learned by Chinese college students whose major is other than English.In China, with the development of society and international communication, more and more importance has been attached to college English teaching and learning.College graduates with good command of English language knowledge and competence are urgently needed in our society.As the fundamental aim of college English education is to foster communication skills and international understanding, it is of great interest to analyze their college English classroom discourse in order to improve teaching and to make learners proficient users of English.
The data are collected from three college English classrooms of non-English major at Qingdao University of Science and Technology.As many classroom-oriented researchers have pointed out, it is only through a better understanding of classroom interaction processes that teachers can render their teaching more profitable for learners (Jamila Boulima, 1999:15).This study is significant in that it reveals the characteristics of classroom interaction that is most favorable for promoting learners' English and has insightful implications for English teaching and learning.The aim of the present study is to develop teacher's reflective thinking about what goes on in their own classrooms and to provide information for improving teaching and learning in college English classrooms.

Literature Review
The discourse analysis tradition in language classroom research grew out of the contributions of various disciplines.It provided a foundation for research in applied linguistics and language pedagogy.The first language classroom research of Bellack et al (1966) is traditionally considered as a pioneering study within this tradition.The study offered a simple description of classroom discourse involving a four-part framework: 1) structure, 2) solicit, 3) respond, 4) react (Dick Allwright & Kathleen M.Bailey, 1991: 98).
Historically, the British work has principally followed structural-linguistic criteria, on the basis of the isolation of units, and sets of rules defining well-formed sequences of discourse (Michael McCarthy,2002:6).One important study was carried out at the University of Birmingham by Sinclair and Coulthard(1975),who developed a model for the description of teacher-pupil talk based on a hierarchically structured system of ranks by analogy with Halliday's (Chaudron, 1988) ranked scale approach.They published their work "Towards an Analysis of Discourse: The English Used by Teachers and Pupils" in 1975, in which their study results were made public.Coulthard published "An introduction to Discourse Analysis" in 1977, summarizing their theory.They suggested that pedagogical discourse at the analytical level be considered in terms of the linguistic level of sentential analysis and the social/pedagogical level of programs and courses.The discourse level involves five ranks: lesson, transaction, exchanges, move, act, each of which builds up the elements of the higher rank, in accordance with the hierarchical structure.They found in the language of traditional native-speaker school classrooms a pattern of three-part exchanges: 1) teacher elicitation, 2) student response, 3) teacher feedback.One aspect of the present study is to find out discourse patterns in college English classrooms for the purpose of which Sinclair and Coulthard's model will be consulted and discussed in Chapter three.Mehan (1979) offered the three structural components of a pedagogic discourse: 1) an opening phase where the participants inform each other that they are in fact going to conduct a lesson as opposed to some other activities; 2) an instructional phase where information is exchanged between teacher and students; 3) a closing phase where participants are reminded of what went on in the core of a lesson (Ellis, 1990).However, Mehan notes that not all language lessons follow the same structure in that language classes may consist entirely of an instructional phase.
McTear (Ellis, 1994:577) observed four types of language use in EFL classroom discourse: 1) mechanical (i.e.no exchange of meaning is involved); 2) meaningful (i.e.meaning is contextualized but there is still no new information to be conveyed); 3)pseudo-communication(i.e.new information is conveyed but in a manner that would be unlikely to occur outside the classroom); 4)real communication (i.e.spontaneous speech resulting from the exchange of opinions, jokes, classroom management etc).Pedagogic discourse is believed to result from real communication type of language use, pseudo-communication is between these two aspects (Ellis, 1990).
The concept of language classroom discourse has undergone various interpretations.Nunan (Nunan, 1993) views classroom discourse as the distinctive type of discourse that occurs in classrooms.In his eyes, special features of classroom discourse include unequal power relationships which are marked by unequal power opportunities for teachers and pupils to nominate topics, take turns at speaking etc.
Ellis (Ellis, 1990) views classroom discourse in terms of two dimensions: 1) the interactive goal, 2) address.The interactive goal is divided into three types as follows: a) core goals are revealed in the explicit pedagogical intentions of the teacher.These goals can be medium-centered, message-centered or activity-centered.b) Framework goals are defined as the interactive goals related with the organization of classroom activity.c) social goals occur when the participants interact on daily social matters.The second dimension of classroom discourse identified by Ellis is the "address" and its related types: address types are associated with one of four identities which a classroom participant can have: teacher, pupil, class member, or group member and with the interactive role he possesses: speaker, addressee and hearer.Van Lier (1988) describes classroom discourse in terms of two dimensions.He classifies the discourse of classroom interaction according to whether the teacher controls the topic (i.e.what is being talked about) and activity (i.e. the way the topic is talked about).Based on this classification, four basic types of classroom interaction are identified: the first type of classroom interaction takes place when neither the topic nor the activity is controlled by the teacher.The second type of interaction occurs when the teacher controls the topic but not the activity.This type of interaction requires teacher transmitting some information or explaining an issue.Type three interactions involve teacher control of both the topic and the activity.Type four occurs when the teacher controls activity but not the topic.This type of interaction involves teacher setting up small group work prescribing the rules but giving freedom of choice of topic (Ellis, 1990).
In China, the research is comparatively few.The materials I can get are very limited.Except "English Conversation" by Amy B.M.Tsui, others are only articles on this subject.For example, Yang min ( 2001) wrote an article "Foreign Language Classroom Research" in which she introduced history and method of classroom research; Xu Erqing and Ying huilan (2002) wrote "New College English Classroom Discourse Research", analyzing the discourse pattern and communicative features.They concluded that classroom discourse they observed was quite near to natural discourse, which is opposite to my assumption.Other works include "features and function of teacher talk" by Hu Xuewen ( 2003), "Investigation and Analysis of Teacher Talk in College English Classroom" by Zhouxing & Zhouyun (2002), "Classroom dialogue and Questioning Strategies" by Mu Fengliang ( 2000) and so on.

Data Collection
The data were collected in three college English intensive reading classes of non-English majors in Qingdao University of Science and Technology where the college English teaching and learning has always been the center of attention because of college English test Band 4.Here "college English" as mentioned above refers to the English learned by non-English majors in the university.On account of the low passing rate of CET-4 in this university in recent years and the reform of CET-4, much emphasis and attention has been attached to college English teaching and learning.As a teacher of college English, the study of college English classroom discourse is of great interest to me in that I want to know what really goes on in our own classrooms and what I can do to improve college English teaching and learning.
The three classes which were observed and recorded from September to November were band 1 college English classes from different departments with about sixty students in each one.The students were about 18 years old and were attending the first year of their college.As freshmen they were conscious that without a good level of English their possibilities on the job market are greatly reduced.The majority of the students have studied English for 6 years although there is considerable difference in level because they came from different places with different level of English teaching.The three teachers in these classes had 5-6 years of teaching experience.The text book used was "New Horizon College English (I)" which is designed to train students' listening, speaking, reading, writing, translating ability with the reading ability as a priority.So the classes chosen here are all intensive reading classes.
Nunan (Nunan, 1989:76) said " there is no substitute for direct observation as a way of finding out about language classrooms", therefore in this study ,the author came into the classroom personally to observe.The observation was conducted in six classes; about 5 hours (50 minutes for each class) were observed and recorded, 3 of which were transcribed and used for analysis, then allowing as many patterns of behavior and inconsistencies as possible to emerge.After class, the author had an interview with the students, knowing more about their ideas and feeling about their teacher and class.Consequently, in order to collect quantitative and qualitative data needed for the analysis, the method employed was that of ethnographic research, and audio recording and field notes were applied as well.Sinclair and Coulthard's "IRF" model Sinclair and Coulthard (1975) developed a model for the description of teacher-pupil talk based on a hierarchy of discourse units.The language of the classroom differs from many forms of spoken discourse in that it is formally structured and controlled by one dominant party, i.e. the teacher.

Introduction of
Halliday (Allwright and Bailey, 1991) developed a description of grammar based on a rank scale.This theory has been used by Sinclair and Coulthard (Andrew Atkins, 2001) to create a model for spoken discourse analysis which is commented on by Malouf (Andrew Atkins,2001) who says "the strongest effort to actually implement Halliday's ideas in a well-grounded, descriptively adequate theory of discourse has been made by Sinclair and Coulthard (1992)…developed as a tool for systematic study of classroom discourse, concentrating mainly on interactions between the teacher and individual students".This is echoed by McCarthy (Michael McCarthy, 2002) who says "it is very useful for analyzing patterns of interaction where talk is relatively tightly structured".So it should be suitable for analyzing the college English classroom discourse patterns.
The Sinclair and Coulthard's model was devised in 1975 and slightly revised in 1992.It, like Halliday's model, is also a rank scale model and consists of five ranks: lesson, transaction, exchange, move and act, and they are related to one another in a" 'consists of' relationship"(Willis, 1992,cited from Andrew Atkins,2001).The ranks are hierarchical in nature with lesson being the largest unit and act being the smallest.Sinclair and Coulthard identify twenty-two different classes of act (Appendix A), which combine to make the five classes of move.These are framing and focusing moves, which combine to make boundary exchanges and opening, responding and follow-up moves, which combine to make teaching exchanges.A number of these exchanges combine to make transactions, which combine to make the lesson (Figure 1).
There are eleven subcategories of teaching exchanges, one of which is eliciting exchange.To put it in detail, that is, the teacher raises a question, then the students answer it, and the teacher gives an evaluative follow-up before raising another question.The three moves that constitute an eliciting exchange are referred to as "Initiation", "Response" and "Follow-up".The three-move structure of an exchange (IRF) is commonly cited, and will also be the basis of my data analysis.When a student replies the teacher's question, the other students may not hear clearly sometimes.So the teacher needs to repeat the students' words so as to make it clear to all the students.What's more important is that the teacher should give "feedback" to the student's answer, to show whether the answer is right or whether it is the answer expected by the teacher.For example: T: (elicit) What does the food give you?P: (reply) strength.
T: (feedback) Not only strength, we have another word for it.P: (reply) energy T: (feedback) Good, energy, yes.
In this kind of three-move structure if the third move doesn't appear, that usually is a hint that the student's reply is not correct.For example: T: (elicit) Can you think why I changed "mat" to "rug"?P: (reply) Mat's got two vowels in it.(Wang Dexing, 1998: 202-209) In this case, pupil's words are not only a response to teacher's question, but also an initiation.In college English classrooms where the students have low proficiency, discourse acts such as "loop", "nomination", "prompt", and "clue" are expected to appear, because when a teacher does not get a response or gets a wrong answer to an elicitation, he can start again by repeating or rephrasing the question, or move on to another student.A discourse element for these teacher's acts is called "bound initiation" (Ib), and it may be bound in ways of "re-initiation", "listing", "reinforce", or "loop".An exchange which reactivates an element in another exchange instead of repeating it or rephrasing it is called "bound exchange", containing single or a few Ib slots (Coulthard, 1992).
In a word, Sinclair and Coulthard's "IRF" model provides us with a set of description category and analysis procedures, on the basis of which data collected can be analyzed.

Results
The results of discourse structure analysis using Sinclair and Coulthard's Initiation-Response-Feedback (IRF) model are shown in Table 1.

Discussion
The discourse analysis by Sinclair and Coulthard (1975) is for classroom situations where the teacher exerts the maximum amount of control over the structure of the discourse (Michael McCarthy, 2002).The above table shows us that several discourse patterns appear simultaneously in one class, with IRF pattern accounting for 35.42% on average, IR pattern 21.88%,IIbRF 16.67% and I pattern 13.54%.These four patterns were observed at relatively high rates, which agree with Sinclair and Coulthard's findings on discourse structure.This is probably because (a) the classes examined in this study were all large classes of about 60 students where the students were in formal or restrictive circumstances.The teachers exerted the maximum amount of control over the classes and the patterns of the discourse.(b) Chinese students have been disciplined not to speak in classes without a teacher's direction, and most of them are unwilling to speak English in front of their classmates because they fear making mistakes and causing laugh from others.(c) The students simply can't speak freely because of their poor spoken English, so they only speak when they were asked to in most of cases.
It is noteworthy that patterns with a bound initiation accounted for about 29.17% of all exchanges on average.It is observed that the students sometimes just kept silence or gave a wrong answer to the teacher's elicitation so the teacher would start again by repeating or rephrasing the question, or move on to another student as in the following examples: 1) I T: For example, "The patient should listen to the doctor's advice and act on it."Who can translate?Qianyang, could you please?R P1: Sorry, I can't.2) I T: And also there is the word "remark".What is its noun form?Ib It's noun form (in Chinese)…… R PP: the original form ( in Chinese).
F T: Good,the same word, meaning speaking negatively( in Chinese) R/I in Table 3 means that response and initiation elements exist in one Utterance, making up only 1.05% of the total exchanges, which also shows that teacher-initiated exchanges are dominant.The following is an example of R/I: In example 4, after the teacher's initiation, students kept silent, so the teacher answered his own question several seconds later, and then started another exchange.
In example 5, when students kept silent, the teacher nominated a girl student.Failing to get correct answer from her, the teacher went on to another student, then the second bound initiation was formed, followed by Response and Feedback.
As a result, these two patterns are also as reasonable as others although they are inconsistent with Coulthard's research.Their existence also suggests students' low English proficiency and teachers' dominance over the class in college English classroom, which has been mentioned above.

Conclusion
In sum, Sinclair and Coulthard's model can be applied to college English classroom discourse analysis in that there were quite a lot patterns found agreeing with their model although there also existed some discrepancy.The patterns of discourse in college English classrooms were that: teacher-initiated exchanges dominated; "IRF" patterns did exist and accounted for the most proportion of all the discourse patterns with an average percentage of 35.42%.Several different patterns found in the data showed that teachers of college English were attempting to motivate students' participation in classroom discourse, and complicated discourse patterns did exist in these classrooms but most of the exchanges were still limited in the pattern of "IRF".

Inspiration from the present study
The present study showed that college English teaching and learning in the three classrooms observed aimed to apply new teaching methods and concepts and teachers made a great effort in motivating the students in participating in classroom activities and communication.Communication did exist.However, there is a long way to go before it can be said that the classes observed were student-centered and had communicative approach as the main teaching method to train students' communicative competence.The study result is significant in improving college English teaching and learning and promoting students' English proficiency in that it outlined the patterns of the classroom discourse.From this study we can know how teachers used their spoken language in conducting a college English lesson and how their way of speaking influences students' response and participation.We can also know the good points and weak points in terms of discourse usage in college English classrooms so as to overcome the shortcomings in college English teaching and learning.
Based on the study above, the following strategies are suggested for the college English teachers to refer to for the purpose of sustaining student engagement and communicative interaction in classrooms: 1) Involving more "negotiated interaction" in classroom discourse From the discourse patterns found in this study, it is clear that "IRF" still dominate the classroom exchange pattern.This means that teacher and students lack skill in keeping the classroom discourse in a sustained and smooth flow.Negotiated interaction is a solution to this problem.Negotiated interaction involves skills in "helping one another to communicate, for example, by jointly expressing messages, filling in lapses in the conversation, indicating gaps in understanding, and repairing communication breakdown" (Boulima, 1999:4).In non-negotiated interaction, the moves "initiation", "response" and "feedback" usually succeed one another, and have a clear-cut function."Initiation" opens the exchange, "response" constitutes a reply to the preceding "initiation", and "feedback" evaluates the preceding "response" and closes the exchange.In "negotiated interaction", however, these moves do not necessarily succeed one another, and their functions seem to overlap.Negotiated interaction can be initiated for two main purposes: to resolve interactive problems or to sustain the conversation.Obviously, negotiated interaction will be useful and helpful in sustaining classroom discourse and make it more tactful.

2) Maximize opportunities for students' participation
The above data analysis tells us it was the teacher who was dominating the classroom.The teacher did the most talk and controlled the topic.In this way the students were passive---they answered questions and got information passively from the teacher.The teacher should bear in mind that it is students who are learning language; therefore maximizing opportunities for students' participation to let them dominate the classroom is very necessary and important.Such target can be reached by such activities as role-play, story telling, debating, holding seminars and making presentations etc.In a word, enlarging students' portion in classroom and letting the learners dominate their own classroom, with the teachers only playing the role of a guide and an organizer is a challenging but beneficial aim, which is also the direction to which teachers in college English classrooms are working.
As a college English teacher, you can refer to these strategies to activate more students into communicative teaching activities in classrooms for the purpose of improving college English teaching and learning.
We should also keep in mind that classroom discourse is a cooperative event in which the teacher and the students cooperate and negotiate with each other in achieving certain instructional goals in the classroom.Language classroom teachers can consider classroom discourses being composed of pedagogic and natural discourse at two poles of an instructional discourse continuum rather than as alternatives.However, we can encourage natural discourse in the language classroom by motivating more equal participation in the negotiation of meaning through various tasks and thus shifting the focus from on the interactive process itself to on promoting students' virtual language proficiency.

T
, an exchange is a structure made up of five moves: I(R/I) R (F)(F).An exchange is formed by at least two moves (initiation and response), and at most by five moves.Look at the following example: T: Can anyone tell me what this means?P: Does it mean danger men at work? T: Yes… (The above three examples are quoted from Ib T: Sorry, I can't.Liupei, what about you?R P1: "The patient should listen to the doctor's advice and act on it."F T: Very good.Sit down please. discourse patterns observed also had some discrepancy from Sinclair and Coulthard's model, that is, patterns "I" and "IIbRIbRF".Just as mentioned above, according to Coulthard, an exchange is formed by at least two moves (initiation and response),and at most by five moves.But the pattern "I" is made up of only one move, while patterns "IIbRIbRF" with six moves as shown in the following examples: 4) I T: Good.Could you give me example?PP: (silence) I T: For example, "The patient should listen to the doctor's advice and act on it."Who can repeat the sentence?(in Chinese) Qianyang, could you please?5) I T: Very good.For example, "He expressed himself,taking advantage of the opportunity" ( in Chinese).How to translate?[pause] Who would like to try? took advantage of the meeting to express his opinion.F T: Yes, very good.Sit down please.Well, so much for what I want to explain this class, next, try your best to remember as many new words as possible.You may start.