Print   |   Save

The Role of Repetition in CLIL Teacher Discourse:
A Comparative Study at Secondary and Tertiary Levels

Emma Dafouz Milne
Universidad Complutense de Madrid

Ana Llinares García
Universidad Autónoma de Madrid


The present study is based on the analysis of teacher discourse in two different CLIL educational contexts: secondary and tertiary settings. Specifically, this research deals with the subject of teacher repetitions since it is generally believed that repetition is a key feature of CLIL classrooms, given the added complexity of learning concepts through another language. Using a Conversation Analysis framework (Tannen, 1989) this paper draws on classroom video-recordings and transcripts from four sessions, with different teacher profiles across the content and language continuum. The preliminary results show that repetitions are indeed present in both settings, especially with a pedagogic function. However, the study also reveals that there are differences in the types of repetitions and in their functions depending on the teacher profile and classroom methodology.


Keywords: Classroom discourse analysis; teacher repetition; secondary education; tertiary education; CLIL

1. Introduction

1. 1. CLIL in secondary and tertiary education in Spanish contexts

In Spain, there are a growing number of Primary and Secondary state schools involved in CLIL projects. One of the pioneering projects was the result of the agreement between the British Council and the Spanish Ministry of Education for the teaching of an integrated curriculum (Spanish/English) in a number of schools ranging from nursery, through primary to secondary level, in a pilot scheme. This project started in 1996 at the pre-school level, and has recently reached secondary education. At this level, all the participating schools teach social sciences (geography and history) in English, while the other subjects selected for the projects depend on the availability of specialists willing to teach their subject in English. At secondary level, CLIL teachers need to be content specialists with a high level of English since the selected subjects are entirely taught in the target language. These schools are even offering the IGCSE (International General Certificate of Secondary Education) to their students, thus providing them with equal opportunities to students attending elite private English schools 1.

Page 50
Page 51

At the tertiary level, however, the situation is much more heterogeneous, given the autonomy that Spanish universities enjoy in order to design their undergraduate and graduate degrees and to offer courses or entire programmes in a foreign language (usually English). In brief, high education centres are aware of the importance of attracting students, now seen as potential “customers” (Wilkinson, 2004). Thus, to do so, teaching through English is viewed by educational authorities as a differentiating force and an added value in any forward looking university (Dafouz and Núñez, in press). Hence, over 30 institutions in Spain are currently offering their “bilingual degrees” at the undergraduate level in fields ranging from Business Administration to Tourism, International Law, Telecommunications or Engineering. By and large, these bilingual degrees are taught by content specialists with a command of English; however, unlike the official requirements in secondary education, university professors do not need to certify their level of English nor do high education institutions need to follow a common syllabus.

1.2. Discourse analysis in the CLIL classroom

In the last ten years, both in Europe and America, a significant number of research projects have dealt with the implementation of new CLIL programmes and evaluation of already established ones (i.e. Snow and Brinton, 1997; Marsh and Langé, 1999). However, research at the micro level with a clear focus on CLIL participants (teachers and students) has not yet reached such a high level of interest (Dalton-Puffer and Smit, 2007). As Leung (2005: 250) points out, we should pay attention to “…how teachers and students use their languages in teaching and learning activities so that we have a better understanding of what goes on in bilingual education classrooms in different world locations.”

The structure and features of classroom discourse vary according to different parameters such as the subject being taught, the age of the students, the language of instruction or the teacher profile, to name a few. Some general features of classroom discourse are related to teacher discourse/talk, since it is a well-known fact that teacher talk dominates approximately 2/3 of classroom time (Chaudron, 1988; Mehan, 1985). In addition, research has also shown that the most common exchange pattern in classrooms is IRF, (i.e. Initiate, Response, Feedback), following Sinclair and Coulthard (1975) and that questions are usually formulated by teachers rather than students. Within the question taxonomy, Long and Sato (1983) refer to knowledge-checking questions as display questions whereas those to which the teacher does not have the answer are known as referential questions. According to Tsui (1995), display questions generate interactions that are typical of pedagogic or didactic discourse, while referential questions generate interactions typical of social communication (Creese, 2006).

Focusing specifically on CLIL contexts, some recent studies have analysed the role of IRF exchanges. Dalton-Puffer, for example, refers to this pattern as activator of “the students’ existing reservoir of knowledge” (2007: 18) which leads to trigger learning through forging connection between old and new knowledge. Nikula (2007) compares IRF exchange patterns in CLIL and EFL classes concluding that in CLIL contexts this structure is less rigid and that teachers and learners have the opportunity of engaging in a different type of dialogue, giving learners more space for interaction.

Page 51
Page 52

1.3. The role of repetition in classroom discourse

The role of repetition in interaction has been one of the research interests of conversational analysts in their study of social patterns of interaction in conversations. The most common distinction made in terms of types of repetitions is the one proposed by Tannen (1989) and Johnstone et al., (1994), which include self-repetition (i.e. repeating what is said by oneself), and allo-/other-repetition (i.e. repeating what is uttered by another speaker). Tannen also identifies exact repetition, repetition with variation, and paraphrase. Exact repetition, as its name suggests, occurs when the speaker repeats a string ipsis litteris. Repetition with variation may be a question that is repeated as a declarative (or vice-versa), a sentence with one word modified. Paraphrase, again as expected, occurs when the speaker expresses the same point with different words. It is also interesting to distinguish between immediate and displaced repetition (Johnstone, 1994), given the importance of the latter in the establishment of textual and social cohesiveness.

In the analysis of second/foreign language classroom discourse, the role of repetitions has also been the focus of a good number of studies. As far as teacher repetitions are concerned, these usually have the dual function of enhancing comprehension, while concurrently providing the learners with more opportunities of becoming aware of L2 features (Pica, 1994; Richards and Lockhart, 1994). In the acquisition of a second language, Tomlin (1994) argues that repetition is a social act with cognitive consequences. With repetition, the teacher helps the pupil to understand the sentence produced, and this, in turn, has the cognitive consequence of helping the learner to transform “input” in the L2 into “intake”.

As far as bilingual contexts are concerned, Llinares’s (2003) study shows that self-repetition is one of the three most common functions in the language of the teacher, especially in low-immersion contexts, where instructors feel the need to reinforce the message to make sure that the learners understand.

2. Research questions and goals of the paper

Some comparative studies in CLIL classroom discourse have focused on EFL/CLIL contexts (Nikula, 2007) or native/non-native teachers (Llinares and Romero, in press). However, to our knowledge, there is no research that contrasts CLIL teaching practices at secondary school and university levels. In this particular paper, thus, we are interested in comparing the functions of repetitions in two different educational contexts (secondary and tertiary education) in order to identify differences or similarities in teaching practices, as well as the conceptualisation of teacher and learner roles.

The research questions put forward are the following:

  1. What is the role of repetition in the secondary and tertiary CLIL classrooms analysed?
  2. What types of repetitions are used by the teachers in these contexts?
  3. In what type of classroom register do these repetitions occur?
Page 52
Page 53

3. Data and instrument of analysis

The set of data that form this paper are a subset of a larger pool which has been collected since 2006 under two research projects 2.

Table 1 summarises the characteristics, participants and subject topics of each context.

  Classroom A Classroom B Classroom C Classroom D
Teacher Content and EFL teacher.
Content teacher with good command of English.
Content teacher with EFL training.
L1= Spanish
Content teacher.
L1= Spanish
Students (n) 24
80 approx.
90 approx.
Topic Feudal Europe Feudal Europe Literature course Engineering course
Subject Geography and history Geography and history North-American Literature Thermodynamic efficiency
Educational Level Secondary Education Secondary Education Tertiary education Tertiary education
Session length 50 minutes 50 minutes 1 h. 15 min. 1 h. 25 min.

The secondary school data consist of videotaped classroom sessions with second year students of history with Spanish as their native language. The data presented here comes from two groups in different schools working on an end-of-topic discussion of the same history topic from the curriculum (“Feudal Europe”). Both teachers, native speakers of Spanish, are history specialists but one of them is also an English specialist who has always worked as an EFL teacher before becoming a CLIL history teacher (classroom A). The teacher in classroom B is a content specialist with high competence in the target language. The subjects from the university context were also videotaped and belong to two different settings: a) second year English Philology students with Spanish as their native language, and b) second/third year Aeronautical Engineering students with English as a Lingua Franca 3. Both university teachers are native speakers of Spanish, the main difference between them being that teacher in classroom C is a content specialist with working experience as an EFL instructor, whereas teacher in classroom D is a content specialist with no experience in teaching through a foreign language.

Page 53
Page 54

The instrument of analysis for this study draws on Tannen’s (1989) model of self- and allo-repetition (including also repetitions with variation and paraphrases).e combine this taxonomy with Llinares (2003) in her distinction between allo-repetitions that function as pedagogic feedback and allo-repetitions that function as interactional feedback. While pedagogic feedback refers to the use of positive or negative evaluation in teacher repetition, interactional feedback is used to encourage learners’ participation and turn-keeping. Examples of both types are included in section 4. Finally, we bring in Christie’s (2002) distinction of two main registers in classroom discourse: regulative and instructional. This register difference will enable us to identify whether repetition appears more in the organizational part of the classroom, when the teacher is giving out instructions on how to proceed (i.e. regulative register), or whether, by contrast, repetition occurs more frequently when the teacher wants to focus on the content, establish a dialogue with students, and encourage their contributions (i.e. instructional).

Page 54
Page 55

4. Results

4.1 Results from the secondary school context

The analysis of the secondary school data shows that teacher repetition is more frequent in the instructional register. In other words, the teachers use more repetitions when they are focusing on the subject content or the language needed to learn this content. Self-repetition is similarly used by both teachers, mainly with the aim of enhancing comprehension. They also coincide in the function of their allo-repetitions: the most frequent is that of pedagogic feedback, as can be expected from any classroom context. However, the two teachers also show interesting differences in the way they use feedback. In the case of the group with the content/language expert teacher (classroom A) more student turns are followed by feedback. On the other hand, the content expert teacher (classroom B) uses feedback in a much lower degree. This finding matches Nikula’s study (2007) when she claims that the IRF structure in EFL classes is more rigid than in CLIL classes. The vast experience of the content/language expert as an EFL teacher might explain her more frequent interest in teacher-initiated IRF sequences (see examples 1 and 2 below with the teacher’s use of pedagogic feedback on content and form, respectively):

Classroom A-Example 1:

TCH: Who were around the roads? No? I’ll give you a clue: Robin Hood
STU1: The rebels
TCH: The rebels. Excellent!

Example 2:

TCH: Yes, but the question was they did not have the same good things in rural areas as they had in cities. Can you explain that?
STU1: They have less salary in a… rural area than in the city.
TCH: They “had”, past. Right?

Another interesting difference between the two secondary school teachers is the use of interactional feedback. Although the most frequent type of feedback used by both teachers is pedagogic, the content/language expert shows fewer instances of interactional feedback 4. As shown in example 3 below, the use of interactional feedback by the teacher in classroom B seemed to encourage the students to adopt a more active role in the interaction, allowing for student-initiated turns (see STU2 below) and thus facilitating the learners’ active participation in the construction of knowledge through the foreign language:

Classroom B-Example 3:

STU1: Eh … also that … eh … Middle Ages … eh … there were a lot of invasions and now no.
TCH: There were a lot of invasionsinteresting…
STU1: We can stay in our homes without … without being afraid.
STU2: We can … we can … travel.

Page 55
Page 56

4.2 Results from the university context

As evidenced from the data analysed, there is more teacher repetition in the instructional register than in the regulative register. In other words, the two university teachers observed use repetitions more frequently when they are focusing on the subject content rather than when they are dealing with the dynamics of the session, or regulative register. Specifically in classroom C (content and EFL teacher) there is a considerable presence of allo-repetitions, since the instructor is constructing the session on student participation. The teacher is presenting the content in an inductive way, by prompting from the students the characteristics of “Fiction” as example 4 shows:

Classroom C-Example 4:

TCH: So, let’s see. What do you think are the elements of fiction? You remembered when we started the course and we started with the poetry we were discussing the most important elements of poetry; so, what are the elements of fiction? For example, just tell me one. Basic elements of fiction.
STU1: The characters.
TCH. The characters. The characters. Another element. Almudena, another element.
STU2: The narrator.
TCH: The narrator, good. The characters, the narrator, more.
STU: The prose?
TCH: The prose, the fiction itself. Eeh, what do you mean, the style?
STU3: XXX (inaudible)
TCH: Ok. All right, the style, like fiction by definition is written in prose, right. Helena?
STU: The th themees
TCH: The themes, all right.

Regarding feedback in allo-repetitions, the data shows that both pedagogic and interactional feedback is used, with pedagogic feedback being more frequent than interactional. This finding reveals that by repeating students’ contributions, the instructor aims to make the subject content clear and accessible while at the same time confirms the validity of students’ input. This type of feedback is sometimes accompanied, although to a lesser extent, by evaluative comments (e.g. good, all right).  

By contrast, in Classroom D (content teacher), allo-repetitions are less frequent since the teacher does not elaborate the session on students’ contributions but rather adopts a lecturing style. Overall, teacher D uses self-repetitions to hold the floor while concurrently thinking what to say next. As example 5 displays, the teacher-student exchange is triggered by the student’s question of some part of the explanation. The teacher uses self-repetition in the first instance (and your question was, what was your question) to encourage the student’s participation. The examples of repetition that follow alternate allo-repetition (the velocity … the velocity of…?) with self-repetition. While allo-repetition serves the teacher to clarify problems of content comprehension, by echoing the student’s difficulty and providing an answer, self-repetitions, which are very frequent in teacher D, seem to act as cohesive devices that help the speaker to organise his speech and the hearer to follow better. In other words, self-repetitions, as in case classroom C, enable the teacher to underline key ideas and ensure that the students understand the main concepts.

Page 56
Page 57

Classroom D-Example 5:

TCH: And your question was, what was your question?
STU1: XXX (Inaudible)
TCH: For the area? Or to increase the area perhaps?
STU1: uummmm
TCH: Or to increase the..
STU1:.the velocity
TCH: The velocity of…?
STU1: The velocity of the valve
TCH: Yes, the characteristic velocity of in the valve, ok? Yes we want to increase this value. To increase this value, okay, perhaps, it’s not correct all because we want to increase this value, we want to increase that but we can increase only to the half of the speed of the sound, has an upper limit.
STU1: (…)
TCH: Yes, we have to increase the area…to reduce… for reduce, perhaps, the area of the piston, but the area of the piston is fixed by the maximum…because the area of the piston is the work in the piston…the work is more or less the pressure multiplied by the area.

4.3 Secondary and tertiary data compared

On the whole, there are similar uses of teacher repetitions in both contexts as regards register. More specifically, in both levels repetitions are more frequently used in the instructional register than in the regulative register. Another similarity lies in the fact that CLIL teachers, irrespective of their background and teaching context (secondary and tertiary), seem to favour pedagogic feedback over interactional feedback. The use of pedagogic feedback does not seem to be connected to L2 practice. In other words, teachers C and D in the university context as well as teacher B in secondary do not incorporate repetitions in their discourse to exemplify, correct or practice any particular linguistic item, but rather to ensure that subject content is rightly understood. It is only the secondary teacher with EFL experience (teacher A) who occasionally uses pedagogic feedback with a linguistic focus (see example 2).

As far as interactional feedback is concerned, there seem to be certain differences depending on the teacher profile. Drawing on our data, interactional feedback is less represented in the case of teachers from an EFL tradition (as teacher A in the secondary context) or those with less teaching experience in the target language (as teacher D in the tertiary context). The dichotomy can be then presented as follows: it is important for the CLIL teacher to be aware of the language needs of his/her specific subject (Llinares, Dafouz and Whittaker, 2007; Llinares and Whittaker, 2007) and language awareness is something that content teachers sometimes lack. However, implementing traditional FL classroom interaction practices (based on the aforementioned IRF model) in CLIL contexts may not provide sufficient opportunities for more authentic communication in the foreign language.

Page 57
Page 58

5. Conclusions and implications

Given the qualitative nature of this study and the limited sample, findings need to be interpreted with caution. Nevertheless, this paper has yielded some interesting results regarding teacher interactional practices in CLIL contexts, specifically in the use of teacher repetitions. In secondary and tertiary CLIL classrooms, repetitions are used by teachers as a strategy to reinforce the understanding of subject content (instructional register). In addition, this qualitative analysis has served to identify a higher number of allo-repetitions over self-repetitions in both educational levels, and the predominance of pedagogic feedback over interactional feedback.

Interestingly, there are differences in the presence of interactional feedback, however, these are more related to the teacher profile and experience than to the educational context. The use of this type of feedback in the classrooms observed has shown a more active involvement of the students’ role in interaction and seems to offer more opportunities for language learning (Llinares, 2003). We believe this type of interactional practice might also have positive effects in the CLIL students’ construction of content knowledge.


Chaudron, C.: 1988, Second Language Classrooms. Cambridge: Cambridge University Press.

Christie, F.: 2002, Classroom Discourse Analysis. London: Continuum

Creese, A.: 2006, Supporting Talk? Partnership Teachers in Classroom Interaction, in Journal of Bilingualism and Bilingual Education (9) 4: 434-453.

Dafouz, E. and Núñez, B. (in press) CLIL in higher education: devising a new learning landscape, in Dafouz, E. and M. Guerrini (eds.) CLIL across education levels: opportunities for all. Madrid: Richmond Publishing.

Dalton-Puffer, C.: 2007, Discourse in Content and Language Integrated Learning (CLIL). Amsterdam: John Benjamins Publishing.

Dalton-Puffer, C. and Smit U.: 2007, Introduction. In Dalton-Puffer, C. and U. Smit (eds.) Empirical perspectives on CLIL classroom discourse. Frankfurt: Peter Lang. pp. 1-13.

Johnstone, B. et al.: 1994, Repetition in Discourse: A Dialogue. In Johnstone, B. (ed.) Repetition in Discourse: Interdisciplinary Perspectives. Norwood, NJ: Ablex, 1994. 1-20.

Leung, C.: 2005, Language and content in bilingual education. Linguistics and Education 16: 238–252.

Llinares García, A.: 2003, Repetition and young learners´ initiations in the L2: a corpus driven analysis. In Archer, D., Rayson, P., Wilson, A. and T. McEnery (eds.) Proceedings of the Corpus Linguistics 2003 Conference (237-245)

Page 58
Page 59

Llinares García, A. and Romero Trillo, J.: (in press) Discourse markers and the pragmatics of native and non-native teachers in a CLIL corpus. In Romero-Trillo, J. (ed) Pragmatics and Corpus Linguistics. A mutualistic entente. Berlin: Mouton de Gruyter

Long, M. and Sato, C.: 1983,  Classroom foreign talk discourse: Forms and functions of teachersユ questions. In H. W. Seliger and M. H. Long (eds.) Classroom oriented research in language learning. Rowley, MA: Newbury House. (268-285)

Marsh, D and Langé, G. (eds.): 1999, Implementing content and language integrated learning. A research-driven TIE-CLIL foundation course reader. Jyväskylä: University of Jyväskylä.

Mehan, H.: 1985, The structure of classroom discourse. In van Dijk, T. (ed.) Handbook of discourse analysis vol.3. Discourse and dialogue. London: Academic Press. (pp. 119-131)

Nikula, T.: 2007, The IRF pattern and space for interaction: comparing CLIL and EFL classrooms. In Dalton-Puffer, C. and Smit, U. (eds.) Empirical Perspectives on CLIL Classroom Discourse. Frankfurt: Peter Lang (179-204)

Pica T.: 1994, Research on negotiation: what does it reveal about second-language learning conditions, processes, and outcomes? Language Learning 44 (3): 493-527.

Richards J.C. and Lockhart, C.: 1994, Reflective teaching in second language classrooms. Cambridge: Cambridge University Press.

Sinclair, J. and Coulthard, M.: 1975, Towards an Analysis of Discourse. Oxford:
Oxford University Press.

Snow, M.A. and Brinton, D.M. (eds.).: 1997, The content based classroom. Perspectives on integrating language and content. White Plains: Longman

Tannen, D.:1989, Talking Voices: Repetition, Dialogue, and Imagery in Conversational Discourse. Cambridge/New York: Cambridge University Press.

Tomlin R.S.: 1994, Repetition in second language acquisition. In Johnstone B (ed.) Repetition in discourse interdisciplinary perspectives. Norwwod, NJ: Ablex.

Tsui. A. B. M.: 1995, Introducing Classroom Interaction. London: Penguin.

Wilkinson, R. (ed.):  2004,  Integrating Content and Language: Meeting the challenge of a Multilingual Higher Education. Netherlands: Universitaire Pers Maastricht.


1 There are also a significant number of CLIL programmes run by different regional governments.
return to text

2 Data from the secondary school context comes from a research project financed by the Comunidad Autónoma de Madrid and the Universidad Autónoma de Madrid (09/SHD/017105; CCG06-UAM/HUM-0544; CCG07-UAM/HUM-1790). Data from the university context belongs to a project funded by the Comunidad de Madrid and the Universidad Complutense de Madrid (CCG07-UCM/HUM.2602).
return to text

3 There were 13 different nationalities represented in the course (Polish, Portuguese, Romanian, Spanish, German, etc). Thus, English is viewed as a lingua franca in this context rather than a foreign language.
return to text

4 We have also collected data from two classes of history on the same topic with students working in the L1 (Spanish). It was interesting to see that the Spanish L1 history teachers used interactional feedback more frequently than the CLIL teachers.
return to text