Sinclair and Coulthard's IRF' Model in A One-To-One Classroom An Analysis.
Sinclair and Coulthard's IRF' Model in A One-To-One Classroom An Analysis.
Sinclair and Coulthard's IRF' Model in A One-To-One Classroom An Analysis.
1. INTRODUCTION
The perceived role of the EFL teacher is to prepare learners to use English outside the
classroom. Nunan (1987) argues that the style of language used in the classroom
environment may seriously effect a student’s ability to cope in the real world, although
this is contended by Seedhouse (1996). It seems therefore expedient for language teachers
to analyse the language of the classroom and assess its effectiveness.
Sinclair and Coulthard (1975) developed a model for analysing spoken language, which
was developed from classroom discourse in general secondary classrooms. It should
therefore be useful when applied to the language of the classroom. 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.
Class sizes vary considerably between institutions, but in my own case, I teach a high
percentage of one-to-one lessons. In this paper, I will look at an example of language used
in one-to-one English language classrooms. My main objective is however, to analyse,
evaluate and comment on the Sinclair and Coulthard (1992) model. I am interested in
seeing if the model will be effective when used to analyse the smallest class size and
whether there will be problems in assigning the data to the categories of the model.
In this paper, I will first review some of the literature available about the Sinclair and
Coulthard model. This will be followed by a description of methods and procedures used
in conducting the analysis and lastly I will address the objective stated above.
2. REVIEW OF LITERATURE
Halliday (1961) developed a description of grammar based on a rank scale. The different
ranks on the scale are related on a ‘consists of’ basis. This theory has been used by Sinclair
and Coulthard (1975, 1992) to create a model for spoken discourse analysis which is
commented on by Malouf (1995b: 1) 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 (1991: 22) who says it ‘is very
useful for analysing patterns of interaction where talk is relatively tightly structured.’ and
should be suitable for the one-to-one classroom.
The Sinclair and Coulthard 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. These are
2
‘lesson; transaction; exchange; move and act, and these are related to one another in a
‘"consists of" relationship.’ Willis (1992: 112). The ranks are hierarchical in nature with
lesson being the largest unit and act being the smallest. Sinclair and Coulthard identify
twenty-one different classes of act (see Appendix 1), 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.
Lesson
Transactions
Teaching Boundary
exchanges exchanges
Free Bound
exchanges exchanges
Re-initiation (1)
I R Ib R F
The Sinclair and Coulthard model, sometimes called discourse analysis (DA), is not
without critics. Malouf (1995a: 1) argues that ‘DA has only been applied to two-party
discourse and would seem to fall short of the full range of linguistic communication.’
DA has however been modified by a number of scholars to account for less structured
discourse patterns (see Brazil and Coulthard 1992,Coulthard 1992; Farooq 1999a,
Francis and Hunston 1992; and Tsui 1992) such as telephone and casual conversations. I
see no reason at this stage of analysis to deviate from the original model, as this would
defeat the purpose of my investigation.
Francis and Hunston (1992) point out one drawback that I have been careful to address
in my data. Para-linguistic features such as gestures and eye-gaze may also be part of the
discourse in face-to-face communication. I felt that direct observation or recording the
lesson on video might have had a negative effect on the naturalness of the discourse in
such a small class. To avoid any negative effect, the lesson was recorded on cassette and
para-linguistic items were looked at with the lesson teacher after the recording.
3.2 Transcription
I had never transcribed any data before, but I had studied a number of transcriptions and
felt ready to start. Nunan (1992) suggests that the ‘presence of [the] machine [can be]
off[-]putting’ (1992: 153). I therefore chose to start transcribing the data from about
halfway through the lesson as I thought the cassette recorder would be having less effect
4
I decided to follow the procedure outlined in Brazil (1995: 29-46) as it seemed very
logical. I first went through the data and separated it into moves. This meant first
identifying framing and focussing moves as well as opening, answering and follow-up
moves. At this stage the analysis was quite straight forward, although at a later time a
number of the moves were re-assessed and re-allocated in part or totality, due to
problems which emerged with the next stage.
It was then necessary to divide the moves into acts and assign them all act labels. This
was the most difficult and time-consuming stage of the analysis, but also the most
enlightening. After a great deal of time, most of the act labels had been assigned, but
some still remained problematic. I shall explain the problems and difficulties I faced in
the next section of the report. Due to the length of the finished analysis, I have attached
it as Appendix 3.
Sinclair and Coulthard (1992: 4-5), provide a definition of the discourse act: ‘Discourse
acts are typically one free clause, plus any subordinate clauses, but there are certain
closed classes where we can specify almost all the possible realizations which consist of
single words or groups.’ (ibid.). They continue to discuss the form and categorisation of
acts (ibid.: 8-21). Francis and Hunston (1992: 133) in Farooq (1999a: 7) provide more
guidance stating that ‘An act must always begin with a new tone unit’. This implies that
pauses will signpost the start and finish of an act. Based on Brazil (1994) it is my
understanding that tone unit boundaries are only a possibility for a pause and are not
always realised with a pause. Brazil (1997: 5-7) suggests the level of difficulty in
5
separating tone units and says there is ‘uncertainty about the observable fact that
continuous speech is broken up into perceptible blocks or units to be interpreted’ (ibid.:
5). He continues to say more on the difficulty of separating tone units, ‘there are cases
where the phonetic evidence for segmentation is less straightforward, and problems of
other kinds arise once we move outside a carefully selected corpus.’ (ibid.: 6). I am
aware of the guidance given by Brazil (1994 and 1997) and this has been taken into
consideration in my analysis.
the student’s attention towards the area of study. I eventually decided on the
classification below (see Appendix 1 for a classification of act categories).
Initiation
Opening move (Elicit)
Mm, questions. Number two. Find
questions with have got and has got
in the conversation.[Read from
textbook] So, can you find any
questions? (s)
What questions use have got
and has got in this conversation?
(el) Can you read them? (cl)
4.2.2 Example 2
Sinclair and Coulthard (1992) are unclear about whether it is possible for a student (or a
teacher) to self-elicit. The first section of the utterance below can be seen from a number
of perspectives.
016-S She’s got brown eyes. Mm. Has she got brown eyes?
‘She’s got brown eyes.’ Appears to be a repeated elicit from the teacher’s previous elicit in
so much as the student is reading aloud the question she must answer. I would have
categorised it as an elicit if it had been said by the teacher, but instead labelled it a
(student) aside. Again, I am not sure if this is a possibility, but Sinclair and Coulthard
(1992: 21) suggest that this category is for items they ‘have trouble dealing with. It is
really instances of the teacher talking to himself.’ This appears to be a case of the student
talking to herself and I found the model lacking in guidance.
4.2.3 Example 3
The next two examples, I have treated as one item as they share many similar features.
They are both part of bound exchanges, in that they follow previous elicits.
(1) 068-S Vera and Dona . . .
069-I Donatella.
070-S Donatella.
071-I Mm hmm.
Utterance lines 068 and 084 are intended as responses but the student is unable to
vocalise the name ‘Donatella’ and falters without finishing. In utterance 069, the teacher
provides a model of how the word is pronounced, which I eventually assigned as an
elicit. In the case of 069, the student repeats the modelled word, and I believe this is
7
intended to be so by the teacher as she accepts the repeat. The case of utterance 085 is
noticeably different. The teacher in this case, helps the student to finish her response,
which is characteristic of a clue. The student seems aware of this function, as she
continues with her response.
4.2.4 Example 4
The next example is a simple looking one but does not fit neatly into one of Sinclair and
Coulthard’s (1992) categories. Anyone who has taught in Japan will be familiar with
‘Janken’ [the stone-paper-scissors game]. It is something done when faced with the
problem of who is to take the first turn in many different situations. It is particularly
popular with school children although not exclusively so. It is similar in a sense to
flipping a coin. It is realised by a non-linguistic action which is characteristic of a react,
as well as a linguistic accompaniment, which is characteristic of a response.
103-I So . . . Mm, ‘Janken’. ‘Janken poi. Ai ko de sho’ [stone, paper, scissors game]. Mm OK.
Have you got a pet?
I feel that the non-linguistic part of the game is most important to its performance and
have therefore assigned the “Janken” said by the teacher to the category of direct. The
following acts have been labelled react, as they are more importantly physical. Two
similar situations are referred to by Willis (1992: 120) who offers:
The non-verbal action in each case provides a react, which shows that the initiation has been
treated as a direct. In these cases, the react is the head of the move with an accompanying verbal
act as . . . post-head.
Coulthard and Brazil (1992: 77) are unsure about the status of the directing move and
say:
[I]t is not clear whether it is better to regard directing moves as a separate primary class of move,
or whether to regard them as a subclass of informing moves concerned with what the speaker
wants B to do . . . Thus the final decision must depend on the significance attached to the non-
verbal action.
In the case of ‘Janken’ the non-verbal action is of primary importance and thus is a
directive. The reason this caused problems of fit is that it is unlikely to occur outside of
Japan, in the kind of traditional classrooms that Sinclair and Coulthard(1975) used to
develop the ‘IRF’ model.
4.2.5 Example 5
The final part of the transcription that I had great difficulty in analysing is not as simple
as the preceding example. It contains numerous acts that I spent time fitting into the
model in numerous different configurations. The original transcription is below:
8
It was very difficult to even understand quite what was happening in these utterances,
especially those which contain only one word.
The main problems start at utterance 133, the teacher informs the student mainly with
the use of gesture what she means by the word ‘laptop’. To this move, I have assigned
the label informative. The next move is made by the student, and as Sinclair and
Coulthard (1992: 19) state that ‘The only response [to an informative] is an
acknowledgement of attention and understanding’, the next move must be an
acknowledgement or part of a new exchange. In utterance 134, the student states the
Japanese word for a laptop computer, which is ‘notebook’, I believe that the student is
showing she understands what the teacher has said and confirming it to herself, this
makes it an acknowledgement. The teacher repeats the student’s utterance, which I have
taken to mean that she accepts the student’s acknowledgement. The teacher continues in
the same utterance with ‘Ah, notebook’. I have assigned this as an elicit, because the
student responds with ‘Mm’ which has the intonation of an acknowledgement. This
appears to be similar to what Tsui (1992: 104-106) calls an ‘Elicit: confirm’ exchange as
the teacher is asking the student to confirm her understanding.
Utterances 137-143 are very confusing to the inexperienced analyst. The teacher starts
by explaining why we use the word ‘laptop’ for a portable computer (utterance 137).
This is, I think, acknowledged by the student with a repeat of the word laptop (utterance
138), however she continues her turn at speech by saying ‘Lap, lap.’ This appears to be a
request for the end of the word, but I am cautious that it may just be an attempt to try and
remember the word. However it is understood by the teacher to be a request for the end
of the word and she says ‘Top’. The student repeats ‘Top’, and I suspect makes some kind
of gesture. I have taken this to be the opening move in an exchange. This is again
understood by the teacher to be an elicit to confirm her understanding and the teacher
confirms that she is correct with ‘Mm’, which appears to be an answering move. I am still
9
undecided what the next move is (utterance 143). I have labelled it an answering move
although I am unsure what it is answering, though it does sound like an
acknowledge/accept, which means it may be a follow-up move. The next move
(utterance 143) sounds like an evaluative follow-up move and I have allocated it to this
category, but it may be perceived by some to be a kind of boundary move. Some
confusion seems to present in the classroom at this time as there is a 28-second pause in
the proceedings.
4.3 Commentary
The Sinclair and Coulthard (1975,1992) IRF model was developed in a more traditional
style of classroom than the one I have examined above, and I believe many of the
problems I have encountered are due to the different styles of discourse which occur in
each situation. The model has been developed by other researchers to account for
different styles of discourse and believe that it would be appropriate to develop a model
n
for one-to-one classes. The structure is now believed to take the form I (I/R) R (F )(see
Francis and Hunston 1992: 123-125 and Coulthard and Brazil 1992: 50-78 for a
description), which would account for a number of the problems I have encountered,
especially those connected with problems of exchange allocation, e.g. Examples 3 and 5.
Coulthard and Brazil (1992: 75) have also developed the categorisation of acts so that
reply is considered part of informative and can occur in both opening move and
answering move slots. This would have overcome some of the problems I encountered
in example 5.
One-to-one classes appear, after this preliminary analysis, to have some features of
discourse which differ form the traditional larger class, and this is to be expected. Act
categories such as, cue, bid and nomination are redundant as they are used to manage
and control the order of the large classroom. These acts are omitted by Francis and
Hunston(1992) in their analysis of telephone conversations.
The high frequency of the utterance ‘Mm’ to show that one party is listening to the other
may in fact differ from the acknowledgement act given by Sinclair and Coulthard (1992)
in that it may not actually be a true move/act as it can occur in the middle of the other
person’s eliciting or informing move. This function would probably be realised in a
larger class by paralinguistic features such as eye contact and nodding of the head. This
could be labelled as an interlocutory sub-move, perhaps.
10
5. CONCLUSION
I have given in the previous section an outline of some of the problems/difficulties I
have encountered in trying to apply this model and suggested briefly some adaptations
which could be made to account for the unique nature of the discourse which occurs in
the one-to-one classroom. Not withstanding the problems I encountered the majority of
the data does seem to fit the given categories, although it takes careful assignment of
labels. It really is no good just looking at the grammatical structures of each act, instead
the acts preceding and following must be used to assess the function of each act. The
intentions of the speaker and the intentions of the speaker as perceived by the listener are
the key to determining the appropriate category.
I believe that the exercise of analysing this model is a very valuable activity for teachers
who wish to gain a greater understanding of the classrooms they teach in. However, the
enormous amount of time it takes to complete, may deter all but the most tenacious.
This preliminary analysis provides opportunity for further research in the field of
discourse analysis and it may be prudent for the writer to further study the small
classroom environment and the form of discourse which occurs there in order to refine
and improve an adapted version of the original model.
11
Appendices
Appendix 1
Sinclair and Coulthard’s (1992) Speech act categories:
16 accept acc Realized by a closed class of items – ‘yes’, ‘no’, ‘good’, ‘fine’, and
repetition of pupil’s reply, all with neutral low fall intonation. Its
function is to indicate that the teacher has heard or seen and that
the informative, reply or react was appropriate.
17 evaluate e Realized by statements and tag questions, including words and
phrases such as ‘good’, ‘interesting’, ‘team point’, commenting on
the quality of the reply, react or initiation, also by ‘yes’, ‘no’,
‘good’, ‘fine’, with a high-fall intonation, and repetition of the
pupil’s reply with either high-fall(positive), or a rise of any
kind(negative evaluation).
Appendix 2
Lesson transcript
001-I Mm. Questions^. Number two. Find questions with have got and has got in the conversation. So,
can you find any questions? What questions use have got and has got in this conversation?
Can you read them?
002-S I can’t make questions.
003-I Mm. No. Just in the conversation you look for the questions.
004-S Mm. [pause14] Mm. Have you got any brothers or sisters? 005-
I Mm.
006-S Has your sister got dark hair, too?
007-I Mm.
008-S What colour eyes has she got?
009-I Good. Now, B, make these statements into questions.
010-S Mm.
011-I You’ve got a brother.
012-S Mm.
013-I What’s the question?
014-S Have you . . . have, have you got a brother?
015-I Good.
016-S She’s got brown eyes. Mm. Has she got brown eyes?
017-I Good. So in a sentence we have the subject.
018-S Mm.
019-I Then the verb, then the second, two parts to the verb have and got. Second part of the verb and
then the object. And then in the question we, the verb is split and the subject goes in the middle.
Have you got. So, you is forward to make the question.
020-S Mm.
021-I OK, so, number three. Look at the conversation again.
022-S Mm.
023-I Complete these questions and answers. A.
024-S Have you got a sister?
025-I Good.
026-S Yes, I have.
027-I Good.
028-S Has he got fair hair? No, she, no, he hasn’t. 029-I
Good. OK. B. How do we make short answers? 030-S
Mm.
031-I Short answers. Yes, I have. No, she hasn’t.
032-S Mm. Yes, I’ve . . .
033-I Mm. Do we use, in a short answer, for yes and no, Do we use short form?
034-S Mm.
035-I Yes, I’ve.
036-S Yes, I have.
037-I Mm. Good. So, how do we make short answers? No short form. For example, Yes, I have. Yes
she has. Not, yes, I’ve. Not, yes, she’s. Do we use got? Yes, I’ve got.
038-S No, we don’t.
039-I OK.
040-S No, she haven’t.
041-I Hasn’t. So we use the first part of the verb, but no short form. And we don’t need got.
042-S Mm. Yeah.
043-I OK. Grammar reference three point one, gives um the rule.
044-S Mm.
045-I For have got. But looks very similar to this one. OK let’s practice. Practice, OK, number one for
practice, complete the description of Andrew and his sister.
046-S Mm.
047-I So Andrew.
048-S Andrew has got a sister. Andrew has got brown hair.
049-I Good.
050-S Andrew and his sister have got blue eyes.
14
051-I Good.[Pause 15] Right, number two, has got. Complete the sentences with have got, has got,
haven’t got or hasn’t got. This is Pilar.
052-S Pilar.
053-I Vera.
054-S Vera.
055-I Donatella.
056-S Donatella.
057-I Mm hmm. I think um Italian.
058-S [Laughter]
059-I English.
060-S Mm.
061-I French.
062-S Mm.
063-I OK. So, what about number one, Pilar? What hair, what kind of hair has Pilar got?
064-S Mm Long hair.
065-I OK.
066-S Long. Pilar has got long hair.
067-I Good.
068-S Vera and Dona . . .
069-I Donatella.
070-S Donatella.
071-I Mm hmm.
072-S . . . hasn’t, haven’t long hair.
073-I haven’t [rising tone]
074-S haven’t got
075-I Mm hmm.
076-S long hair.
077-I Good. Number four, three.
078-S Pilar has, hasn’t got brown eyes. [Pause8]
079-I Number four.
080-S Pilar and Vera has got dark hair.
081-I Ooh.
082-S have got
083-I Good.
084-S dark hair. [pause5] Vera and Dona . . .
085-I tella
086-S Donatella haven’t, haven’t got blue eyes.
087-I Good.
088-S [pause6] Pilar has got blue eyes.
089-I Mm hmm.
090-S Vera is got, Vera has got brown eyes. [pause9] Donatella . . . hasn’t got brown . . . hair, hair.
091-I Good. OK. Number three. Work with a partner.
092-S Mm.
093-I Ask questions and write down the answers. Try to find out some details. For example, have you
got a pet?
094-S Mm.
095 Yes I have.
096-S Mm.
097-I or, no I haven’t.
098-S Mm.
099-I Mm. What have you got? What’s its name? How old is it? What colour is it? OK.
100-S Mm.
101-I Extra questions.
102-S Mm.
103-I So . . . Mm, ‘Janken’. ‘Janken poi. Ai ko de sho’ [stone, paper, scissors game]. Mm OK. Have
you got a pet?
104-S Yes, I have.
105-I What colour is it?
106-S Mm, it is black.
107-I Uuuh. Is it a boy or a girl?
15
108-S It is girl.
109-I What’s her name?
110-S Her, her name is Marron.
111-I Marron. What . . . mm, what have you got? What animal have you got?
112-S What animal . . .
113-I Yeah, what kind of pet have you got?
114-S Mm . . . I, I have got a dog.
115-I How old is she?
116-S Um . . . she is three years old.
117-I OK, thank you. So you can ask any question.
118-S Mm . . . Mm, have you got, have you got a computer?
119-I Oh, yes, I have.
120-S Mm. [Pause7]What colour is it?
121-I Mm, it’s dark grey.
122-S Dark grey.[Pause11] Mm, How, how do you spell grey?
123-I OK, Mm, English, uh British English or American English? (laughs)
124-S Um, British English.
125-I G-R-E-Y. American English is A-Y.
126-S A-Y.
127-I Mm. Ok. Mm.
128-S Mm. What have you got?
129-I It’s a laptop.
130-S Laptop?
131-I Mm, um. Portable. I can put it in a bag.
132-S Uh.
133-I Laptop.
134-S Notebook.
135-I Notebook? Ah, notebook.
136-S Mm.
137-I We say laptop, because um, this is the, when you sit down this is your lap.
138-S Lap, top? Lap, lap.
139-I Top.
140-S Top.
141-I Mm.
142-S Mm.
143-I Okay. [Pause28] What are you thinking? [laughter]
144-S [laughter] What shall we do? (?)
145-I Oh, it’s OK. New question. Yeah, yeah, OK, yeah, no problem. Mm, OK, have you got a car?
146-S Car? Yes, I have.
Appendix 3
Classroom Analysis
Initiation Response Feedback Ex
Opening move (Elicit) Answering Follow-up 1.
Mm, questions. Number two. Find I can’t make questions.(rep) Mm, (acc)
questions with have got and has got No. (e)
in the conversation.[Read from
textbook]So, can you find any
questions? (s)
What questions use have got and
has got in this conversation? (el)
Can you read them? (cl)
sisters? (rep)
Answering Follow-up 3.
Has your sister got dark Mm. (acc)
hair, too? (rep)
Answering Follow-up 4.
What colour eyes has she Good. (e)
got? (rep)
Framing move 5.
Now^, (m)
Focusing move Answering 6.
B. Make these statements in to Mm. (ack)
questions.[Read from textbook] (s)