Willis J. Edmondson Subjective Parameters describing Teaching Roles Towards a theory of tertiary foreign language instruction Abstract. lt is argued that teachers' perceptions of their role(s) in foreign language teaching have a .dedsive influence on teaching effectiveness. The nature of these perceptions is investigated via an applied discourse-analytic treatment of .a recorded group discussion held by four university teachers of English in Germany. Two major parameters are proposed to capture performance variables. They are the role of theory in teaching practice, and the influence of the context of communication. Three perceptually distinct, but in practice co-present teaching roles are then developed according to these three perceptions, teaching is at the same time an institutional, a professional, and a personal activity. Differences in the weightings given these three roles lead to radically different views regarding appropriate teacher-pupil role relationships. 1. lntroduction "There is a quaint, old-fashioned und ultimately highly damaging British view that teaching . is an art. Higher education carries much of the blame for this view. lt is responsible also for the second factor that prevents an applied science of teaching the low status of applied and practical work in educational research" (David Reynolds, reported in the Times Higher Educational Supplement, May 22, 1998) In this article I wish to investigate the teaching of English as a Foreign Language at tertiary level. In particular, the question as to the teacher' s role will be explored. Tue exploration will be based on subjective data, in two senses. Firstly, the question will be investigated via the perceptions, insights, claims, boasts, theories and reported practices of experts, namely a highly experienced group of language teachers. Secondly, such subjective data will be filtered through the author's own cognitive/ affective perceptions. Tue goal is to reveal what appear as critical parameters, along which individual experiences, beliefs or proclivities are presented and/ or perceived. Tue article is structured as follows: firstly, the focus taken in this investigation will be described and argued for. Secondly, the methodology adopted will be elaborated, reference being made thereby to a preliminary investigation into teacher attitudes and beliefs, which can be seen as a precursor to the present study. The core of the paper consists of the analysis of a group discussion between university teachers of English. The analyis will be conducted on two levels in sections 4 and 5, focussing respectively on parameters of performance, and parameters of perception. A brief summary concludes the paper. FLuL 27 (1998) 82 Willis J. Edmondson 2. The Role Relationship Teacher-Learner The general assumption in. the paper is that the nature of the working relationship between teacher and leamer, between Lektor 1 and student, is of central importance in terms of leaming outcomes, i.e. the success of the language course concemed. This assumption can be supported on theoretical, observational, and pedagogic grounds. In supporting it, I shall then in turn refer to foreign language classroom leaming theory (2.1), the motives and goals operating for the special case of univetsity courses for students of English (2.2), and views of language teaching currently promulgated inside pedagogic theory (2.3). 2.1 The Interaction Hypothesis for FL teaching Consider the view of classroom learning developed by Allwright, and in particular the 'interaction hypothesis' he proposes (Allwright 1984). The starting-point for this hypothesis is the observation that leamers do not necessarily leam what teachers teach, and that, moreover, different leamers leam different things from the same lessons. This leads to a view of the classroom as a social event offering different kinds of leaming opportunities, which will be exploited, ignored, expanded upon or rejected in different ways by different leamers. lt is inside such a view of classroom leaming that the central role of classroom interaction becomes crucial, as, in Allwright' s formulation: "the processes of classroom interaction determine what leaming opportunites become available to be learned from" (Allwright 1984). Allwright further claims that inside the foreign language classroom the interactional process constitutes the process of language leaming/ acquisition. 2 If this second claim is accepted, it follows that the term 'interaction' now covers not only observable interactional structures and discourse pattems ('extemal interaction'), but also the cognitive processes underpinning and triggered by such discourse ('intemal interaction'). In other words, attention-focussing, noticing, awareness, restructuring, and affective dimensions of behaviour, which influence energy investment, depth of processing, and so on, are also to be captured in the term 'interaction'. On this view, then, language acquisition is to be located in the interaction between extemal and internal factors inside the learner, and this interaction is extemally stimulated by the interactional processes by means of which lessons are realized. I have decided to write this paper in English because the data on which it is based is in English, and will be cited extensively. As, however, the context and provenance of both data and paper are totally German, I shall, on occasion, indulge in code-switching, especially when using terms relevant to the German educational system, for which a translation would be both inadequate and possibly misleading. 2 In what follows, I shall use the terms 'leaming' and 'acquisition' in free variation. FLuL 27 (] 998) Subjective Parameters describing Teaching Roles ... 83 The argument is, then, not only that student/ lecturer role relationships in large part determine patterns of interaction, and therefore the provision of learning opportunities, but that they will also impact on what use is made of such opportunities for purposes of learning. 2.2 The case of university language classes for students of English My concern will be the teaching of English to advanced, adult students of English, as opposed to learners of English (students of physics attending ESP-courses, for example, are for my purposes here learners, not students). For main-stream students of English, there is, I suggest, in the normal case no dear-cut professional purpose inherent in such a course of studies. This means inter alia that the goals of the language learning/ teaching component of such a course of studies cannot be readily defined externally, i.e. in terms of real-world knowledge, skills, domains, or purposes 3 ( on the distinction between external and internal language teaching goals, see for example Edmondson & House 1993, 278-281). External syllabus specifications tend, then, to be rather general, referring perhaps to such things as degree of grammatical control, near-native pronunciation, active and passive control of a variety of different types of text, and suchlike. There is, then, I suggest, considerable scope for interpretation concerning how such broad syllabus specifications are to be implemented in practice. One procedure is to re-define the syllabus specifications internally, referring maybe to some testing procedure such as a translation, or to some activity such as the stylistic analysis of a piece of English literature (either or both may further be enthroned in a Staatsexam, for example). The syllabus becomes, in consequence, arbitrary, in that it is determined by the internal goals of the curriculum. Backlash and vicious circling result. In practice, then, syllabus specifications for university English courses for students of English are likely to be either rather open to interpretation, or determined by chance and/ or tradition. This is not to say, of course, that individual lecturers might not well have their own totally clear notions as to what the goals are for the course or courses they teach. Similarly, it is totally plausible to assume that some students especially, perhaps, if English is studied as a minor and not a major study option also have quite clear goals underpinning their participation in English language classes, goals which, however, may have little or nothing to do with the study of English in a traditional philological sense. As regards the effectiveness. of professional encounters between teachers and students, the question now arises whether and to what 3 This is not true for students aspiring to become secondary school teachers inside the German education system (Staatsexarnkandidaten/ -innen). However, specific-purpose language courses focussing on L2 classroom discourse, and, for example, various modalities of assessing leamer interlanguage performance are not, to the best of my knowledge, seen as an inherent part of the education of students of English aspiring to teach, though it seems rather obvious that they should be. FLuL 27 (] 998) 84 Willis J. Edmondson extent there is common ground between the specified (albeit, I have suggested, somewhat nebulous) goals laid down for the course of studies, the teaching goals set up for specific courses, the goals set up by different teachers, possibly teaching the same course or courses, and the personal goals and/ or motives underlying the attendance of the individual students, which, again, are likely to differ from one student to another.. My argument is, then, that this issue of mapping such "personal agendas" (in the sense of Schumann/ Schumann 1977) onto group encounters is particularly problematical for university language courses for students of English. lf personal agendas differ, then clearly some interactional work is necessary, if the encounter is to offer relevant leaming/ teaching opportunities for afl participants. The thrust of the argumentation is therefore to support the contention that the working relationship between teacher and leamer, between Lektor and student, is of central importance in terms of leaming outcomes, i.e. the success of the language course concemed, as perceived role relations to a large extent determine the nature and role of interaction and negotiation in classroom encounters. 2.3 The View from foreign Ianguage teaching theory This view is further consistent with various trends inside foreign language pedagogy and in teacher training (inside Fremdsprachendidaktik), which are covered by terms such as 'experiential learning', 'leamer autonomy', 'process-teaching', and doubtless others. Thus Kokonen (1992) distinguishes between "traditional" and "experiential" models of education, invoking various dichotomies, including: the nature of the power relationship holding in the classroom (emphasis on teacher' s authority versus viewing the teacher as a 'leamer among leamers') the nature of the teacher's role (frontal instruction versus leaming facilitation) the concept of a syllabus (static versus dynamic/ ptedefined versus open) view of leaming control (teacher-structured versus self-directed) Motivation (extrinsic versus intrinsic). 4 There is some overlap between the distinctions set up in comparing these two 'models' and the paradigm shifts inside language education and foreign language teacher training postulated in Legutke 1993, 307. Legutke characterises the new paradigm shift via, for example: a focus · on tasks and meaningful activities, inside which the leamer is not a passive recipient of language form, but an active and creative language user 4 a shift from the leamer as individual to the leamer as member of the social group actively involved in co-managing the leaming process Adapted from Kohonen (1992: 31). FLuL 27 (1998) Subjective Parameters describing Teaching Roles ... 85 a view of the course syllabus as something to be negotiated, rather than something to 'get through' or tobe 'complied with' new definitions of the roles of teachers and learners a theoretical focus on the learning process itself (rather than teacher input and/ or learner output). This focus leads to the view that much of the responsibility for successful language learning rests with individual learners and with their ability to take füll advantage of opportunities to learn. Both Kohonen and Legutke claim tobe simply describing, though in both articles the distinction between description and advocacy becomes blurred. 5 However, it is not my purpose to evaluate the accuracy or appropriacy of these distinctions, nor is it to be assumed that that which is set up in opposition to 'traditional' practices is thereby preferable, and more effective. My purpose is not to use these categories as yardsticks, against which the sulJjective accounts in my data are to be measured. My purpose is simply to show that the focus on the role. relationship between teacher and student s.eems to be of critical moment inside some current pedagogic views on teacher training and teaching practice. In fact, it will become clear that issues raised by Kokonen and Legutke map directly onto the topics discussed in my data. 3. Data and Methodology: subjective theories Tue use of group discussions as a source of evidence for the construction of theories of social behaviour is well-established, and is fairly common in some fields of enquiry (Dreher/ Dreher 1982). To the best of my knowledge, however, this elicitation procedure has not been adopted inside research into foreign language teaching and learning. The resultant data is clearly subjective, in the technical sense that an extemal validation of the truth value of its propositional content can only be carried out, if at all, by highly indirect means, which allow, at best, a probabilistic deduction. In other words, only the subject producing such data is really in a position to evaluate its truth-value. This means that the researcher is faced with a dual problem, as it is necessary to distinguish between objective and subjective validation, given that it is possible to topicalise one's own belief systems. Thus, given an utterance of the kind "I don't think I'm an authoritative teacher at all actually", we can ask whether the speaker is in fact an 'authoritarian teacher', a proposition that is in principle empirically testable, despite problems regarding the operationalisation of the key term 'authorative'; but we can also ask whether or not the speaker really thinks this is the case, and this second issue is much more 5 Blurring is in fact often built into such oppositions via the terms used. Who, for example, can be against active language use in the classroom, against co-management, negotiaiton, or a focus on the learning process? FLuL 27 (1998) 86 Willis J. Edmondson difficult to assess, being a subjective claim. Note, too, that the subjective claim may be true, although the objective one is not, and vice versa. This issue of Fremdsprachen Lehren und Lernen is devoted to teachers' subjective theories. The question as to what conditions need to hold, before a set of subjective data (a teacher's report on personal practices and beliefs, for example) can be construed as a subjective theory (of language teaching and/ or of language leaming) is subject to debate (see for example Groeben [et al.] 1988, Grotjahn 1991, Edmondson 1996, and cf. the Introduction to the thematic section of this issue). The theory developed by Groeben [et al.] 1988 invokes different types of subjective theory (using this term in a broad, non-technical sense), which can be constructed for different explanatory purposes. lt is important to realise that the notion of 'subjective theory' involves more than simply recording what subjects report about themselves. Inside Groeben [et al.]'s theory, they recomrnend first of all investigating the rationality and completeness of subjects' theories via 'communicative validation'. Via talk, therefore, subjects may be encouraged to explore things like the internal consistency of the views they put forward. Tue procedure raises critical issues, it seems to me. Although it is clear that humans may be viewed as 'rational' agents, it is by no means clear that our belief-systems and/ or motives are logically consistent. lt seems to me that the claim that they are should be the outcome of empirical validation, and not its premiss. Further, of course, this first validatory phase communicative validation is open to the objection that what is said in conversational behaviour is often determined by the demands of the communicative situation, rather than the dictates of one's beliefs or goals (this issue is raised once more in section 4 following). This truism is built into Groeben et al' s research programme, in that it is accepted that such validatory procedures may actually affect not only what subjects report about their beliefs and motives, but thöse beliefs and motives themselves. Communicative validation, we might say, thus becomes a form of communicative action research. We are now faced with an infinite regress, however, since if a validation procedure may actually change that which it seeks to validate, one presumably needs to go on to communicatively validate the effects of the initial communicative validation. Groeben et al recognise this, too, and therefore go further than communicative validation, and require 'explanatory validation', i.e. empirical verification. Exactly how such procedures might be operationalised in a concrete case remains, however, open. A major problem for such empirical observation is the initial recognition that there often is in fact no one-to-one correspondence between subjective theory and practical action. 6 Tlie claim that there is such a correspondence is in danger of making 6 That discrepancies may arise between what teachers say they believe, what they say they do, and what actually happens in their lessons is well-documented (e.g. Dinsmore 1985, Winkler 1988, Zimmermann 1984; 1990, Savignon 1991). Cases in which teachers appear to operate with mutually-incompatible goals and/ or belief systems - 'irrational subjective theories', perhaps are also well-documented (see e.g. Wagner 1987). FLuL 27 (] 998) Subjective Parameters describing Teaching Roles ... 87 subjective theory construction an exercise in reconstructivist psycho-analysis, it seems to me. There are therefore, in my opinion, serious problems with the ·research programme, and I do not wish, inside this paper, to operate inside the programme set up by Groeben and co-workers, and laid out in Grotjahn (1991). lt will be preferable, therefore, to avoid the term 'subjective theory' in what follows. My purpose, then, is not to characterise subjective theories of teaching, but, more modestly, to set up perceptual parameters, along which belief-systems, goals and practices reported on by different individuals can be located. 3.1 Individual Scripted Interviews (Student/ Lecturer) Before detailing the methodology of the present study, I wish to report briefly on an earlier piece of work, carried out by a group of students as part of their M.A. course of studies in Sprachlehrforschung. This early study is relevant in terms of giving further substance to the 'Critical Role-Relationship-Hypothesis' developed in 2. above, and is also relevant to the following discussion of subjective data collection procedures. Inside· a seminar entitled "Fremdsprachenlernen an der Universität: Ziele, Probleme, Lösungen", a project was conceived and carried out involving interviewing members of the language teaching staff at .the University of Hamburg. The interviews were structured, a wide range of different issues having been decided upon by the dass. Each interview was conducted individually between student and member of teaching staff, the majority being tape-recorded. Tue students simply approached different members of staff, and interviewed those willing to cooperate. Some participating teachers were willing to be interviewed, but not willing to have their comments taped. In such cases, two students took part in the interview, and together reconstructed the content afterwards. Tue data was analysed on a topicbasis only. In other words, profiles of individual participating teachers were not established or compared. Of specific relevance in the context of the present study was the following sequence of questions: TOPIC 8. Inwiefern stimmt es Ihrer Meinung nach, daß die Studierenden selbst die Verantwortung für ihren Lernerfolg übernehmen müssen? TOPIC 9. Sehen Sie es als Ihre Aufgabe, die Studierenden zu motivieren? TOPIC 10. Selbstverständlich sollten Lehrlp-äfte eine gewisse Autorität haben. Was verstehen Sie darunter? Nineteen interviews were carried out and documented, covering a wide range of languages and several different types of courses, with both contract teachers and full-time members of staff. In documenting this project, the participating students classified the answers they received to different questions, thus enabling a simplistic statistical resumee, before going on to more qualitative interpretation. As regards the question as to who, ultimately, bears responsibility for learning FLuL 27 (1998) 88 Willis J. Edmondson success, or the lack thereof (topic 8), nine answers from the interviews could be classified as saying that this responsibility was shared between student and teacher, seven claimed that the student (but not the teacher) bare this responsibility, while just one person claimed that the inverse holds (the answers given in two interviews could not be slotted into these categories). As regards motivation (topic 9), thirteen answers viewed motivating the students as a major concem of the teacher, three responses focussed more on not diminishing the students' motivation, and just two responses claimed, essentially, that a concem with motivation on the part of the teacher was unnecessary. Interestingly, the answers to topic 7 and to topic 8 appear to be independent, in that a belief that it is the student who bears responsibility for leaming is totally consistent with a belief that it is the teacher' s job to motivate. The opinions on the issue of authority (topic 10) are both varied and cautious. The clearest views refer to the professional competence of the teachers (five responses), their responsibility for the teaching programme "die Verantwortung, die sie für den Fremdsprachenunterricht übernehmen"), or that they be shown respect ,as teachers. These three views did not overlap. A complete lack of interest in the notion of 'authority' was voiced three times, while the view that a non-intimate relationship between student and teacher "nicht-kumpelhafte Beziehung") is desirable was expressed by two interviewees. Of particular interest inside this configuration of views is that the teacher' s responsibility for the teaching programme is seen as a major source of his or her authority, while at the sarne time the responsibility for leaming was seen to lie either wholly or equally with the student. The students conducting the interviews did so in the context of a seminar on Sprachlehrforschung 7, and reported that six of the teachers interviewed were critical of research in general, and of Sprachlehrforschung in particular, when answering the last scheduled question, which was simply an invitation to add any points not raised so far in the interview. This figure assumes even more significance when one considers that the teachers interviewed were self 0 selected in terms of their willingness to cooperate with the students on a project inside Sprachlehrforschung, and when we further consider that at least four of the teachers interviewed would have had no contact with Sprachlehrforschung whatsoever, as teachers of languages such as Chinese are not engaged by the university department in which Sprachlehrforschung is also taught. This is interesting in that a distrust of or indeed open hostility towards - 'theory' and 'research', towards what is perceived as ivorytower academical posturing is attested in the literature. Such attitudes are apparently widespread among teachers, and are not restricted to the case of the foreign language teacher (cf. for example Schlotthaus/ Noelle 1984, who report on interviews 7 In Hamburg the language teaching programmes for the more commonly-taught languages and the programme for students of Sprachlehrforschung are organised and located inside one department. FLuL 27 (1998) Subjective Parameters describing Teaching Roles ... 89 with teachers of German regarding their opinion of and approach to the communicative teaching of German). Tue results of this student project raise relevant issues, and demonstrate that a wide range of different opinions appear to hold inside the group of teachers interviewed. Clearly, however, the general pröcedure is open to criticism, as pointed out by the students themselves in their documentation. For example: Tue interviews were conducted in German, and the questions posed by the students were often differently interpreted by different subjects, and, indeed, in some cases ignored. Over half the students reported some tension before and/ or during the interview, possibly resulting from the duality of the role relationship in the elicitation context (roughly: A investigates B for research purposes, and A attends classes given by B for educational purposes). 8 Tue elicitation technique meant that very many different topics. were addressed, many of which may have .been of no importance whatsoever to the interviewees. 3.2 Small Group Discussion (Four lecturers, one researcher) Such problems are to some extent inherent in any form of subjective data elicitation. lt seemed appropriate, however, when planning a follow-up study, to use a different data-elicitation technique. The data for the following exercise in applied discourse analysis consists of a minimally-steered recorded discussion held between four university teachers of English, in the presence of the author, who sought to focus, re-direct, or summarise the topic(s) raised on four occasions, and who also participated on the content level during the later stages of the recording. Each of the four teachers of English has a füll-time university teaching post and a minimum of fifteen years experience in teaching English inside the German university system. All four are further known to each other personally via professional and private contacts, being employed in the same institute. The author is known to them also, but does not teach English, and has no direct responsibility for the English language teaching programme. All five particpants were bom in English-speaking countries, and English was the language of discussion. 8 The students reported, fo rexample, as follows: "Andere fühlten sich in eine Prüfungssituation versetzt und versuchten, die 'richtige' Antwort zu geben. Bei vielen Interviews hatten wir das Gefühl, daß durch die Präsenz des Kassettenrecorders die Dozenten nicht ganz frei redeten. Mögliche imaginäre Repressionen am Arbeitsplatz durch zu offene Meinungskundgabe beeinträchtigten unserer Meinung nach die Spontanität und Unbefangenheit der Probanden, was sich auf die Klarheit der Daten auswirkte... Unsicherheit ist sicherlich ein Stichwort in diesem Zusammenhang. Ein klar formuliertes didaktisches Credo konnte man nur bei ganz wenigen Personen ausmachen. Einige schienen die Fragen als Angriff auf ihre Person zu verstehen und antworteten dementsprechend leicht aggressiv oder begaben sich in eine Verteidigungsposition." FLuL 27 (1998) 90 Willis J. Edmondson An invitation to join in such a discussion was issued by the author via a telephone call, during which the following points were made: - The invitation was issued on a private basis, there being therefore no obligation to participate. - The talk would be recorded, and the author made it clear that he intended to use the recording as the basis for a publication, which would then contain transcribed excerpts. - Assuming this took place, the identity of the individual speakers would not be revealed (unless individuals wished tobe named). - The general focus would be on the role of the teacher in tertiary foreign language teaching, embracing issues such as authority, responsibility, and motivation. A time-slot of ca. 90 minutes was suggested, agreed to by' the participants, and fully used. 4. Parameters of Performance This discussion of research methodology, and issues of content validation will now be made more concrete via an attempted systematic treatment of performance parameters in the recorded data. The question to be handled here on a theoretical level is where, so to speak, the expressed opinions have their source. Why do the speakers say what they say? Two parameters will be proposed. Tue first concems the relationship between 'theory' and 'practice'. At issue here is how far in discussion reference is made to pedagogic edict or theory, and teaching practice explicitly based thereon, and how far practice is simply presented as theory, or indeed as self-justificatory. How far, we might ask, is speakers' pedagogic practice presented as the consequence of pedagogic persuasion, and how far is pedagogic persuasion simply a reflex of teaching experience? Along this parameter, different positions can be distinguished. In Diagram 1, foHowing the top-down versus bottom-up metaphor, whereby concepts are imposed upon or derived from behaviours, we may represent this parameter on a vertical scale. The second parameter concerns one aspect of the issue of validation, as raised in 2 above in the discussion of subjective theory research program. This parameter may be presented horizontally. What emerges then is as in Diagram 1: . Context of Cornmunication Beliefs/ Principles ['theory'] Memories/ Episodes ['practice'] Diagram 1: Where does what is said come from? FLuL 27 (1998) Subjective Parameters describing Teaching Roles ... 91 The vertical scale concerns then different weightings, according as what is said refers more to what is done than to what is believed, is more derived from practice than from principle, while on the horizontal scale the issue is how far what is said is a function of the situation of saying, and how far it is a function of the preexisting representations of belief systems/ recorded behavioural practice represented inside the speaker. lt is important to realise that both parameters are descriptive, and represent clines. As Diagram 1 (page 90) attempts to show, there is always a mix. For example, only a fool or a bigot holds on to belief systems that are contradicted in experience, and all communicative behaviour is conditioned by contextual constraints to a greater or lesser extent. I wish now to illustrate these two parameters in operation. 4.1 Principles and Practice The total discussion is clearly weighted towards behavioural practice, as opposed to pedagogic principle, or language learning theory. In fact, leaming is not a central issue in the data, although presumably the main point of giving English classes is clearly that learning of some kind takes place, and whether this happens is the main yardstick for evaluating pedagogic practice. There are in fact six references to students' learning in the discussion, and four of these are very general (e.g. "it is part of our task to give them something / \ to set up a learning .. procedure and .. " [Davies 9 ], or "they are / \ very interested in trying to learn that, and improving .. " [Bailey]. In just two further cases (DI and D2)1°, teaching behaviour and/ or teaching content is related to aspects of learning: Dl: ... and I think it's quite important for language learning that they have tobe relaxed [Corder). D2: ... it had to do with aktives Lesen .. with learning strategies / \ reading strategies and so on .. [Bailey) However, there are clear individual differences on this parameter, in that, for example, Austin's contributions are practically confined to descriptions of this person's own classes. Tue following extracts illustrate this tendency: 9 The four participants will be named Austin, Bailey, Corder, and Davies, these names being culled from the bibliography in Ellis 1994, chosen in terms of comparability and familiarity, and assigned rather arbitrarily. To assure syllabic conformity, and alphabetic continuity, the author will be identified as Edmond. 10 Data excerpts will be numbered sequentially throughout (DI - Dn) for ease of reference. When single tums are cited, or extracts therefrom, the source will be given afterwards in square brackets: in dialogic extracts, however, the speakers will be named before their tums. All elements enclosed in square brackets are comments: pausology is indicated (minimally) by arrowheads, heavy stress by italics. FLuL 27 (1998) 92 Willis J. Edmondson D3: Edmond: No. I mean is it that the language goals are non-negotiable / \ but the content you use or work with in order to reach them may be / \/ \ is that the difference? [Two or three turn claims] Austin: .. weil for example erm I assign them oral presentations / \ we can call them oral reports whatever you wantA but they often have then the chance to choose the topic it has tobe approved by me but that's basically to see that there's no repetition in the class we ... (followed by a further 80 words of descriptive exposition). Here the answer offered to the elicitation by Edmond is a further exemplification, In another extensive sequence, focussed on the topic of authority, Bailey and Corder discuss individual cases in which students directly challenged them for different reasons. Austin' s first contribution to this discussion essentially leads to a change of focus, and is as follows: D4: I think the Eingang especially you might say in this regard is probably the most difficult / \ [AGREES] all these people come in from different places from different schools... [Austin]. Note that the issue is seen as one of practical difficulty, for which structural/ institutional reasons are given. This topic of homogeneous learner groups is then pursued, and Bailey suggests that incoming students are now less diverse than previously. Austin then picks up this topic again: D5: The numbersAA the numbers have changed but I've still got some very weak people there in that course and I can only say that last semester the course I hadA I had a number of almost native speakers who had spent time in foreign countriesA and erA they were doing all kinds of things .... [Austin] My somewhat subjective assessment of the locations of the discussants along the vertical parameter represented in Diagram 1 (page 90), from more to less exclusively oriented to issues arising from practical experience is as follows: Austin, Davies, Bailey and Corder. 11 4.2 Individual versus Group Identity Because of its importance for the issue of role relationships, the horizontal scale in Diagram 1 (page 90) will be exemplified here by reference to issues of group iden- 11 Some simple statistics concerning the frequency and distribution of tums, derived from the transcription, are given in the Appendix (p. 103), together with frequencies of usage of modal expressions of obligation. At this point we can see that the rough order just suggested on the vertical sca1e in diagram 1 (p. 90) globally matches tum-length positively, while there is a tendency towards a negative correlation conceming tum-frequency, back-channel activity, and the expression of teacher-oriented obligation. lt so happens, too, that the remarks concerning learning mentioned above were produced by Bailey and Corder. These remarks are not put forward as evidence for the ordering suggested, but as correlational tendencies. FLuL 27 (] 998) Subjective Parameters describing Teaching Roles ... 93 tity. The relevance is clear when we consider the make-up of the discussion group. Given that the whole discussion was designed to .focus on the role relationship between student and teacher, and given that all participants are teachers, and not students, there is interactional pressure towards consensus or solidarity, I suggest. lt is further likely that the presence of the author might have affected the expression of group solidarity. How then do the speakers (from now on, the author' s participation will be largely ignored) perceive and present themselves? How far is there a collective identity inside this group, and how is it to be characterised? Amongst colleagues, differences of opinion and of practice may presumably be expected, and are, moreover, likely to be known to operate,. such that the issue of group identity may conflict with the nature of the individual speaker's identity. Role structuring parameters operate therefore as in Diagram 2 the dimension ME versus US is located inside the opposition between US and THEM. 12 + ME US THEM Diagram 2: Group relationship parameters The use of personal pronouns is a major indication of how group relations are perceived. A first and obvious point is that first person plural pronouns are seldom used to refer to teacher and taught. ME and THEM are rarely referred to as "us". Descriptions of classroom ·practice are more öften presented as what 1/ WE do, and/ or what THEY do; but not as what 'we' in the sense of ME/ US-and-THEM do. Such an inclusive use of the first person plural pronoun occurs on four occasions. For example, Corder describes a strategy for handling disputes with students, which occur in class time, as follows: D6: no / \ we go outside while the others are busy doing something else .. Two further instances simply refer to class activities, necessarily involving both US and THEM: D7: Last week in course X we were Jooking at the artistic use of language... [Corder] D8: I give them one every week and we discuss it the week after [Davies] 12 Conceptual categories ·will be given in CAPITALS, in order to distinguish them from linguistic tokens that may be used to express such categories, although there will often be a direct matching, of course. In other words, 'we' will be used to refer to US. FLuL 27 (1998) 94 Willis J. Edmondson In these three instances, I suggest, the avoidance of an inclusive "we" would in fact be marked, such that it is merely the scarcity of such usages that is of interest (note for example that in D8, 'our' discussion is premissed on 'my' giving 'them' something). In a fourth instance, however, Corder refers to "we" in a context where "they" could equally well have occurred: D9: we're not here to have a good time, / \ though it's nice if we can .. [Corder]. Inside the context of this discussion, then, the US/ fHEM teacher/ pupil relationship is, rather predictably, in fact encoded as 'us' as distinct from 'them'. lt is, however, not always clear who WE are. Group characterisations embracing the speaker include the four teachers of English present, the five persons present during the discussion, native-speakers of English, all English language teachers in the institution concemed, all foreign language teachers in this establishment, and lecturers/ Lehrkräfte für besondere Aufgaben (as opposed to other professional groups, specifically professors). Shifts occur between these first person plural identifications, which are not always distinguishable, as inclusive and subset relations clearly hold. lt is of importance however to recognise that speaker group-identity fluctuates, as it influences the nature of the role relationship with THEM, i.e. with learners. Against this background let us turn then to the ME versus US parameter, where, I have suggested, the context of communication would seem likely to exert some influence on what is said in group discussion. At issue is how far individual opinions and/ or practices are compatible with a group identity. Thus, although individual differences of opinion occur, they are seldom articulated directly ("I dunno / \ I don't agree exactly with you/ \ at least in my experience / \ it seems to me that.." [Bailey] is a typically-hedged disagreement), and are seen as totally compatible with a collective group identity. Thus, it commonly occurs that an individual both claims group membership, and topicalises personal preferences or practices, which in fact distinguish him or her from that group. In theory, an utterance of the form [WE DO X, BUT I DO Y], where X and Y are possibly incompatible, and certainly not easily reconciled, should be logically outrageous. lt is not, of course. This is amongst other things a reflex of the influence of the context of cominunication, as suggested in Diagram 1 (page 90). Individual speakers try, on the whole, to steer their presentations away from ME towards US. lt is noticeable, for example, that back-channel activity becomes very marked, when a generic claim regarding US is made, even though or indeed especially when the individual voicing the opinion may do so with reduced modal force. Consider D 10: FLuL 27 (1998) Subjective Parameters describing Teaching Roles ... 95 D10: Yeah I think that is a point too to bring out A in fact that whether it's fifteen pages or three five-page papers is [AGREEMENT NOISES] A doesn't matter and I have a feeling that we all do A er have certain a certain level of similar requirements and I think what is important is A it isn't a low level A er because basically these kids these young adults have had nine years of English, and they have to do something otherwise they're bored they're passive otherwise it's uninteresting for them [AGREE] so the requirement can 't be that low and it isn't that low among any of us I think [AGREEMENT NOISES] [Davies] · In this turn differences in terms of work-requirements for different courses are smoothed over, and receive emphatic back-channel support. Note too the hedged expressions: "I have a feeling", "to a certain extent", "any of us I think", which, together with the stress strategies employed, function as appeals for confirmation, and as such are successful. In D 11, Bailey brings up a point raised three times, but only by this participant. Again, the agreement is triggered by a rising intonation on 'abilities': D11: ... we could say something about their personalities and about A their academic abilitiesA [AGREES] that we're able to do that.[Bailey] A further example of a willingness to provide back-channel support for any claim referring to the group is D12: D12: and that is negotiable [Mmm from Corder] A that's where the variation comes in A for exarnple what kind of medium they choose [Hm] and also within my final project they can either work from a tape they've collected themselves A a tape that I've .collected or they can use a video-tape and the speakers can be a combination of non-native-speakers A or native-speakers with non-native-speakers or native speakers A so there's lotsa of variation I think within this frarnework that A we set up and I think that's true for all of us [BACK-CHANNEL AGREES] no matter what it is A the course. [Bailey] Such behaviours contrast markedly with occasional, noticeable silences, following the expression of individual opinions. In D13, Davies describes the group collectively (it is a claim about US), the characterisation is positive ("we're just not like that"), and the speaker moreover sets the group off from other groups (not German, not school teachers). In principle, then, we should expect enormous solidarity, which, however, is not clearly given. Instead, Corder's reaction flouts this expectation, and is accorded a quite different reception. D13: Davies: they're interested A we don't represent the German authoritarian teacher figure to them A because we're just not that at all. [Back channel confusion] Corder: I don't think the German teachers are authoritative any more. I think that is one problem for some of our students when they first come here A it can sometimes take two or three weeks A especially some of the better students, to get warmed up, before they get to the point where they realise they can learn something from you and they want to A in the Eingang. [Silence] Edmond: Hmm FLuL 27 (1998) 96 Willis J. Edmondson Corder: I think sometimes their exposure to us A erm some of the more.liberalteachers in the German school system A especially here in Hamburg, haven't been beneficial to some of the better students [Pause] Austin: I support that At times, too, voiced claims of common goals and strategies seem less than transparent, which may, of course, lead to them being emphatically reinforced. Consider D14. The context for this contribution is the remark "if they get the chance", which might be interpreted as hinting that sometimes students do not get such a chance. At this point, Davies assumes the speaking-role: D14: Corder: ... I think it is [pleasant? ] when you get input from the students they decide they want to do something and that isn't actually so rare. that they really are prepared to take an active part in shaping the content AAi\ if they get the chance Davies: Yeah I would agree with that erm approach basically A in my classes I begin the classes with a program and it's clear what my plan A you know what my input is going to be A you know in terms of providing material and it's clear er what my expectation is from them A what will they be doing but I agree with you it's negotiable and the more transparent it is in terms of what I want from them and what I expect from them A it's negotiable in finding tasks which maybe you know force them to do the same type of work but allow them different er topics to do it in [AGREE- MENT noises] A so our approaches seem to be similar in fact Here, back-channel agreement confirms the speaker's intention of expressing solidarity. On a content level, it is not at all clear how the approach described is compatible with the previous one, as requirements, contents and goals are apparently laid out clearly beforehand. This is, on my interpretation, precisely why agreement is voiced at the relevant point in time. The context of communication thus affects what is said, how it is said, and how it is responded to. This conclusion has been arrived at by looking at the ME versus US group identity parameter set up in Diagram 2. In this way, the two performance parameters proposed in Diagram 1 (page 90) have been examined, and the different effects of both on individual behaviours have been suggested. We may now turn to our central concern: teacher-student role relationships, the US versus THEM parameter inside Diagram 2. 5. Parameters of Perception We shall pröceed in three steps. Firstly, some descriptive metaphors will be derived from the data, descriptions which are presented as both appropriate and problematical, in that qualification is imrnediately offered. Such metaphors seem, then, to suggest certain tensions, or incompatible attitudes towards the teaching task. Secondly, some perceived sources of 'authority' which operate inside the working relationship between students and teachers will be listed. Finally, an attempt will be made to parameterize these differences along three basic dimensions. FLuL 27 (1998) Subjective Parameters describing Teaching Roles ... 97 5.1 Defining the Relationship In the following three selected data segments the relationship between US and THEM is viewed in three different ways, each of which is qualified: METAPHOR 1: Tue students are clients, but quality control is carried out by teachers D15: Corder: Weil I think actually at the moment our role is changing. It's certainly changing on the British scene / \ it's probably going to change in the near future here as weil because students are being seen much more as clients. You can see that in the way universities approach potential students / \ they're really selling themse! ves / \ and I think that's changing staff attitudes as weil Edmond: yeah .. has that already wom off on you, as it were .. Corder: Definitely, yes Edmond: How Corder: trying to be reader/ user-friendly / \ [Laughter] I don't think we have a very authoritarian attitude anyway / \ erm / \ very many of us are on first-name terms with students in our courses / \ but that can also lead to some problems because there are some points at which you have to assert authority in order to make sure that work gets done. METAPHOR 2: Teacher is an Organiser/ Manager but at times an Autocrat D16: Bailey: I think if you do group work and if you're sort of the / \ a Jot, which I do / \ erm that you're the organiser or the manager basically of setting up the groups / \ and I sometimes ask students erm .. [omission] / \ or I say find three other people to work with and they have to find their own group. They're. I find that most of my students are pretty erm what shall I say ( .. ) lazy isn't quite the right word .. Edmond: [interrupting] passive Bailey: passive er yeah, and if I don't assign them to a group then they don't get up and go and find their own group / \ so it just goes faster if I assign them groups .. METAPHOR 3: Teachers offer their wares, and students choose their purchases, but teachers are also taskmasters D17: .. most of those people know what they're getting into ... [OMISSION] In the Haupstudium, they've been here long enough / \ they've experienced some of us / \ they know who they can get along with / \ who they can't get along with / \ and they usually go to those people ... [OMISSION] .. we are basically also taskmasters / \ because we are exposing them to / \ causing them to do things that they might not otherwise have done... [Austin] D18: .. and erm I think that's important that it is part of our task to give them something / \ to set up a leaming .. procedure and to make sure it happens actually [Davies] 5.2 Aspects of the Working Relationship Tue positions sketched in 5.1 all implicitly raise the issue of 'authority', as caveats added to the three metaphors employed. In other words, something is needed if the teacher' s perceived or assumed responsibility for 'quality control', for autocracy in FLuL 27 (1998) 98 Willis J. Edmondson the interests of efficiency, or for ensuring that the work gets done, is to work. This something I term authority. I therefore use the term 'authority' to mean whatever teachers possess as individuals or as a category (as ME or as one of US), which enables them to be accepted by students in their professional role or roles. The definition is necessarily loose, as the nature of the professional role is intimately bound up with the concept of authority operating. The question can, however, now be paraphrased as follows: how do teachers require/ hope / wish that THEY view US? The term 'authority' is therefore being used in a broad sense, and may embrace everyday concepts such as 'respect', or 'liking' or indeed, in principle, notions such as 'obligation', or 'anxiety'. Where might it come from? 5.2.1 Language Expertise lt may be assumed or claimed that teachers' expertise in English language issues is the basis of their authority: Cl9: .. I demand respect as the erm the expert in EnglishA in the classroom... I don't like I don't like to see myself as an authoritarian teacherA and I don't think my courses are set up that way .. but I do find myself feeling a bit irritated or even more than just a bit irritated with students who challenge my methodology or my criteria and .. [OMISSION] I don't like being challenged by by students about topics of exercises .. [Bailey] 5.2.2 Native Speakerhood A further source of authority in the broad sense intended here is seen to derive from the fact that the teachers to some extent embody that which they teach: C20: Davies: I think we have a plus too because we're native speakers and they deal with us in a different way than they might erm with a German English teacher don't you think Bailey: [NOISE] authority A Davies: well I wouldn't even put it as authority. I think it somehow has A a plus in terms of aura or appeal. [silence] .... [OMISSION] Corder: you gain an aura Bailey: right Davies: well to a certain extent A er this er zu viel gesagt A aber we are what they are aspiring to become in a senseA in a sense... in terms of the language A it's too much I understand but erm I do think they bring something positive 5.2.3 Work Ethic C21: I wanted to come back to one thing you said A you said the authority the respect you have as a teacher. I don't ! hink a teacher has from the start respect A respect has to be earned A which means basically you set goals for them but they have to see that these goals are possible A and the same things that you demand of them A you demand of yourself A for example that you also get things back punctually A that you are on time yourself A more or less A you are some kind of a role model A and you can't start FLuL 27 (1998) Subjective Parameters describing Teaching Roles ... 99 expecting things from them if you can't do them yourself... and that's when the respect thing actually starts / \ .when they see that you actually / \ are doing the same thing yourself that you expect from them [BACK-CHANNEL NOISE, mostly positive] [Austin] 5.2.4 Personalised Contact Another issue, topicalised implicitly via the issue of appropriate forms of address when addressing students in German, is how far the teacher/ student relationship is a 'personal' one, even one of friendship. In this respect, language teaching is 'different', and this is a source of authority, too: C22: I find a difference to other subjects / \ this is what my students have said in many cases / \ is that we get to know them better as people because there is more give and take... it's less a lecturing situation and so we know them and many of them come to us for letters of reference for scholarships and when we say / \ oh it would be better if you had a professor do it for you and they say weil the professors don't know me... [Bailey] 5.2.5 Institutionalised Role Language teachers also have, of course, institutionalised roles, whereby they are assigned authority inside an educational institution. The most tangible evidence for this is that teachers distribute or withhold course credits: C23: .. and I sometimes get students coming up and .. erm and saying [rising inton] / \/ \ things like erm I don't really want to do this or I don't have time to do this / \ and I just look at them and I say 'weil that's your decision' and they look at me like they're expecting me to get mad with them / \ and I er/ \ say but I'm afraid that is one of the requirements for getting a Schein so if you don 't do it then I can't give you a Schein and you will have to repeat the course. [Bailey] C24: . .1 find though that now that we have registration / \ and students sign up for courses that the ones who do sign up / \ do er wanna get the Scheine and therefore do the work. I haven't really noticed that they don't do it [Davies] 5.3 Role Parameters Widdowson 1987 distinguishes between two roles bound up in the term 'teacher': a social role, and a professional role. Thus a person with the relevant qualification is a teacher (professeur), even if he or she has never carried out the profossional activity normally associated with this title, while other persons can carry out the professional activity (einseignant), without 'really' being teachers, i.e. without having been institutionally labelled. Widdowson goes on to point out that the meaning of 'leamer' is not so complex, possible matching social roles for this activity role being for example those of 'schoolgirl' or 'student'. We should, I suggest, accept this distinction, whereby the social role is to be equated with an institutional role, the professional role with a set of activities. FLuL 27 (1998) 100 Willis J. Edmondson Thus the role relationship we are investigating is a dual one: it is a teacher/ leamer relationship, and also a teacher/ student relationship (1 have in fact switched between these two formulations throughout this paper so far). Additionally, I propose, we may use the ME versus YOU parameter invoked in Diagram 2, and refer rather simplistically to a teaching individual, with his or her personalised agenda, as opposed to his or her categorial roles, be they institutional or professional. In setting up this additional parameter, I do not mean to suggest that behind a professional role such as 'teacher' a bland uniformity exists, but that individual persons can and do perceive their personalities as eo-existent with their professional and/ or institutional roles to noticeably different degrees. If this is acceptable, we have then a three'-way distinction between individual person, socio-institutional role, and professional domain role. The sources of 'authority' listed in 5.2 appear to be associated with these different roles to different degrees. The 'institutionalised' authority referred to in 5.2.5, for example, is clearly inherent in the institutional role of teacher. As nativespeakerhood is in some German universities a prerequisite for appointment as foreign language instructor, this feature can also be seen as institutional. One might, however, wish to argue that it is part of a professional role, not an institutional one. My response to this view would be that native-speakerhood of itself can scarcely count as a professional qualification, though possibly you might wish to say that it is a prerequisite for acquiring certain kinds of professional qualification. However, the dispute is of no matter for my purposes in this paper. We may simply accept, I suggest, that the five different ways of viewing the teaching role in 5.2 above relate to different role categories as I have just developed them. Thus the issue of language expertise is clearly professional (cf. 5.2.1), and the issue raised in 5.2.2 (setting up a role-model/ work ethos) can be seen as professional, too or conceivably-institutional. The notion of personalised contact may be perceived as inherent in the language teaching profession, but is more plausibly interpreted, it seems to me, as a function of personality. I want to suggest, then, that in establishing role relations in their classrooms, individual teachers rely to different extents on the three 'roles' that make up their teaching persona: the institutional, the professional, and the personal. I would argue that this is a fruitful way, in fact, of conceptualising different kinds of tensions established in the literature (cf. footnote 6 above ), and referred to in the metaphors explored in 5.1. A focus by the individual teacher on one of these three facets of the teaching role does not imply a total neglect of the others. lt is not the case that persons who largely adopt and assume an institutional role relationship with their students are thereby to be deemed 'non-professional', but that their concept of professionalism will in large part be determined by their perceived institutional role. Similarly, a person who seeks to teach 'professionally' may to a certain degree disregard institutional constraints, but will normally have to function inside the given syllabus and examination requirements (or else lose his or her position). Finally, the 'individualist' may well be aware of both professional and institutional FLuL 27 (1998) Subjective Parameters describing Teaching Roles ... 101 aspects of teaching as restrictions. One way of describing the challenge presented by the teaching task is then to point to the need to give priorities inside these eoexistent three roles, and the further need to reconcile all three in teaching practice. Additional to the combinatorial priorities given to these three roles by the individual teacher, there is, I suggest, a further variable, or parameter, to be invoked. I shall name the poles of this parameter 'presentation' versus 'accommodation' (cf. the distinctions of Legutke and Kokonen reported on in 3.2 above). By 'presentation' is meant presentation of self, of one's own perceived role or roles as teacher. Via the term 'accommodation' I wish to suggest the possibility of redefining one's own role (within perceived professional, institutional and personal constraints), in the light of the situation in which this role is intended tobe effective. Clearly, this parameter represents a cline. By definition, every teacher 'presents' in the intended sense: at issue is perhaps the possibility of negotiation, the possibility of change, and inside what frameworks either or both may operate. Tue result of this welding of variables is suggested by Diagram 3, which, however, clearly fails to present four dimensions of difference in a graphically unambiguous fashion: PRESENTATION + professional - -/ personal +/ -------,' . +_ i. 1 mst1tut1ona ACCOMMODATION Diagram 3: Three teaching roles along a presentation/ accommodation axis 6. Summary I would wish to claim that in principle all differences of attitude, perception, and reported behaviour in the data available can be accounted for in terms of the parameters delineated in Diagram 3 (see above), taking additionally the variables suggested in Diagrams 1 and 2 into account. One issue which requires further investigation and a different research methodology, involving possibly in-depth interviews, and reconstructive data from such discussions concerns the presentation/ accommodation parameter. lt is not clear to me how far the adopted identity FLuL 27 (1998) 102 Willis J. Edmondson configuration (the relative weightings of professional, personal and institutional persona) is in fact independent of a location on the presentation/ accommodation parameter. If the teacher role configuration in large part determines locations along this axis, then clearly the parameter can be removed from the model. lt would seem a priori logical to suggest for example that a professional weighting of one's perception of the teaching task would imply a weighting towards accommodation on the presentation/ accommodation parameter, while heavy institutional weighting might go together with a presentational stance. The logic behind such a view would simply be that institutions, and the edicts goveming them, tend not to be flexible, while some form of accommodation is probably contained in most if not all theories of language teaching (cf. section 2.3 above). Same evidence that is compatible with this stance is given in the appendix of this paper (see page 103). A large range of possible role conceptions can clearly arise inside the complex of parameters I have developed. This seems to require no apology. Indeed, the data has led the author to propose this complexity, even though the data consists of just ninety minutes' sampled talk from only four language teachers, all of them operating inside one and the same institutional setting, What, then, can be said in terms of language teaching policy, when one is faced by such a variety of role conceptions within such a theoretically homogeneous group of teachers? C25 suggests an answer: C25: Bailey: But that's different according to the class type and also the teachers/ \ [AGREE- MENT NOISES] ... No. I think they have stereotypes like we all have/ \ and whether they're confirmed or not depends on the person / \ and that's the process they go through here / \ and I know that they do to a certain extent self-select Edmond: yeah Bailey: In other words they choose certain teachers that they know and keep going back to them and there are others that they know that they don 't go back to and I think that we represent a wide enough spectrum of er [types Corder: options Edmond: choices] to satisfy all prejudices Davies: huhum Bailey: so we all have followings / \ I think. Davies: weil it has to do with their interests, too/ \ I mean we have the division between British and American / \ [AGREE noises] and amongst the Americans there are different subject / \ linguistic orientation [or the cultural Corder: same with the brits] Davies: so there is variety in the program and the students go to the interests that they have .. This answer is voiced in different guises by all four participants, and constitutes a rather neat argument to the effect that the variety of teacher concepts available, the variety of teacher roles on offer, .is itself accommodating in the sense of Diagram 3 (page ). At the same time, of course, consensus on this issue is consistent with FLuL 27 (1998) Subjective Parameters describing Teaching Roles ... 103 the horizontal parameter in Diagram 1 (page 90), and with the data discussed in 4.2 above, to which, finally, C26 offers a postscript: C26: Bailey: look how compatible we are [laughs] Corder: that's cos we're all lying [laughs].13 Appendix Austin A: Total turns 57 B: Back channel turns 10 C: Content turns 1 (= A-B) 47 D: Interspersals 14 6 E: Content turns 2 15 38 F: Words used in E 16 2432 G: Turn length (= F/ E) 64 Bailey Corder Davies 140 118 66 24 17 12 116 101 54 10 12 14 96 75 43 3912 2017 1805 41 28 42 Table 1: Some Distributive Data derived from a transcribed Group Discussion 13 I am extremely grateful to the colleagues who shared their views with me, and have Jet me use the resulting tape. Such permission does not, of course, imply •any agreement with my interpretation of the data. 14 The term is meant to cover turns which are interspersed between two segments of speech by some other speaker, dealing with roughly the same topic. Sometimes these 'interspersals' may be viewed as ignored interruptions, and also as floor-gaining attempts. Interspersals cover both supportive and repair work, whereby for example an utterance by the previous/ ongoing speaker is checked on, paraphrased, clarified, or rephrased. The whole point is that the notion of 'turn' is radically affected by such behaviours, as three interspersals may be viewed as comments on ONE turn, or as separators for FOUR turns. lt is conceded that the identification of interspersals is somewhat subjective, and that the distinction between back-channel behaviours and interspersals is not clear-cut. 15 Here the figures in row C are adjusted downwards, inthe light of how often interspersals occur in the individual speaker's turns. This is of course independent of how often individual speakers themselves intersperse. However, clearly we may roughly expect that across the board L D + IE = IC, i.e. the number of interspersals made should match the number of turnreductions caused thereby. 16 Including false starts, articulated hesitations, and word repetitions. FLuL 27 (1998) f04 Willis J. Edmondson WE THEY Corder 10 1 Bailey 6 2 Austin 2 3 Davies 2 3 Table 2: Distribution of Expressions of Obligation 17 referring to Teachers (WE) or Students (THEY) References ALLWRIGHT, Richard (1984): "Why don't learners learn what teachers teach? The Interaction Hypothesis". In: SINGLETON, D.M./ LITTLE, D.G. (eds.) Language Leaming in Formal and Informal Contexts. Dublin: IRAAL, 3-18. DINSMORE, David (1985): "Waiting for Godot in the EFL Classroom". In: English Language Teaching Journal 39, 225-234. DREHER, Michael/ DREHER, Eva (1982): "Gruppendiskussion". In: HUBER/ MANDEL (eds.) 1982, 141-165. EDMONDSON, Willis/ HOUSE, Juliane (1993): Einführung in die Sprachlehrforschung. Tübingen: Francke. EDMONDSON, Willis (1996): "Subjective Theories of Second Language Acquisition". In: KLEIN, Jürgen/ VANDERBEKE, Dirk (eds.): Anglistentag 1995 Greifswald. Proceedings. Tübingen: Niemeyer, 453-464. ELLIS, Rod (1994): The Study of Second Language Acquisition. Oxford: Oxford University Press. GROEBEN, Norbert/ WAHL, Diethelm/ SCHLEE, Jörg/ SCHEELE, Brigitte (1988): Forschungsprogramm Subjektive Theorien: Eine Einführung in die Psychologie des reflexiven Subjekts. Tübingen: Francke. GROTJAHN, Rüdiger (1991): "The Research Programme Subjective Theories. A New approach in Language Acquisition". In: Studies in Second Language Acquisition 13, 187-214. HUBER, Günter / MANDEL, Heinz (eds.) (1982): Verbale Daten. Eine Einführung in die Grundlagen der Methoden. Weinheim/ Basel: Beltz. KOHONEN, Viljo (1992): "Experiential language learning: second language learning as cooperative leamer education". In: NUNAN 1992, 15-39. LEGUTKE, Michael (1993): "Room to Talk: Experiential Learning in the Foreign Language Classroom". In: Die Neueren Sprachen 92.4, 306--311. NUNAN, David (ed.) (1992): Collaborative Language Learning and Teaching. Cambridge: Cambridge University Press. SAVIGNON, Sändra (1991): "Research on the role of communication in classroom-based foreign language acquisition: On the interpretation, expression and negotiation of meaning". In: 17 FREED, Barbara (ed.): Foreign Language Acquisition Research and the Classroom. Lexington, MA: Heath, 31-45. lt transpires that HAVE TO is the only modal expression used. FLuL 27 ( 1998) Subjective Parameters describing Teaching Roles ... 105 SCHL0ITHAUS, Werner/ N0ELLE, Karin ( 1984): "Kommunikationsprobleme mit der Kommunikation? " In: BAURMANN, Jürgen/ HOPPE, Ottfried (eds.) (1994): Handbuch für Deutschlehrer. Stuttgart: Kohlhammer, 19-40. SCHUMANN, Francine / SCHUMANN, John ( 1977): "Diary of a Language Leamer: an Introspective Study of Second Language Leaming". In: BR0WN, H.D./ YüRIO, C.A. / CRYMES, R.H. (eds.): On TESOL '77. Teaching and Leaming English as a Second Language: Trends in Research and Practice. Washington, DC: TESOL, 241-249 [personal agenda]. WAGNER, Angelika (1987): "Knots in Teachers' Thinking". In: CALDERHEAD, James (ed.) (1987): Exploring Teachers' Thinking. London: Cassell, 161-178. WIDD0WSON, Henry (1987): "The Roles of Teacher and Leamer". In: English Language Teach 0 ing Journal 14.2, 83-88. WINKLER, Karl-Heinz (1988): "Englischunterricht an der Realschule. Versuch einer Bestandsaufnahme". In: Neusprachliche Mitteilungen 41, 20-25. ZIMMERMANN, Günther (1984): Erkundungen zur Praxis des Grammatikunterrichts. Frankfurt: Diesterweg. ZIMMERMANN, Günther (1990): Grammatik im Fremdsprachenunterricht der Erwachsenenbildung. Ismanirtg: Hueber. FLuL 27 (1998)