THE EFFECT OF TASK VARIATION IN TEACHER-LED GROUPS ON REPAIR OF ENGLISH AS A FOREIGN LANGUAGE By RICHARD FRANKLIN BERWICK B . A . , The U n i v e r s i t y o f C a l i f o r n i a , B e r k e l e y , 1966 M . E d , The U n i v e r s i t y o f B r i t i s h C o l u m b i a , 1979 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF EDUCATION i n THE FACULTY OF GRADUATE STUDIES Department o f A d m i n i s t r a t i v e , A d u l t and H i g h e r E d u c a t i o n We a c c e p t t h i s t h e s i s as c o n f o r m i n g to the r e q u i r e d s t a n d a r d THE UNIVERSITY OF BRITISH COLUMBIA O c t o b e r , 1988 ® R i c h a r d F r a n k l i n B e r w i c k , 1988 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Administrative, Adult and Higher Education Department of The University of British Columbia Vancouver, Canada October 14, 1988 Date DE-6 (2/88) ABSTRACT An experiment was conducted to determine how learners and teachers of English as a foreign language i n Japan cooperatively attempt to improve the comprehensibility of t h e i r t a l k i n English during performance of various conversational tasks. The basic p r a c t i c a l issue under study was the p o s s i b i l i t y that c e r t a i n kinds of teacher-led groups and tasks would be more e f f e c t i v e i n generating r e p a i r and negotiation of the language by which tasks are accomplished than others, and that these group-task combinations might eventually be employed as alternatives to t r a d i t i o n a l teacher-fronted forms of foreign language i n s t r u c t i o n . The study was operationalized i n a 2 x 5 between-and-within subjects, repeated-measures analysis of variance design. Two, six-dyad, teacher-led groups—homogeneous (Japanese teacher/Japanese learner) and mixed (native English speaking teacher/Japanese learner)—were formed i n order to compare the frequency of 12 repair exponents generated during performance of f i v e tasks. Teaching goals were represented i n two tasks, i n s t r u c t i o n i n use of the string-searching function of a laptop computer 1) with and 2) without the computer p h y s i c a l l y present. Non-teaching (s o c i a l and cooperative problem-solving) goals were embodied i n three additional tasks, 3) free discussion, and construction of a Lego (snap-together) toy accomplished with p a r t i c i p a n t s facing 4) away from and 5) towards each other. Task categories were also divided into e x p e r i e n t i a l and i i expository a c t i v i t i e s (respectively, Tasks 2 and 5, and Tasks 1 and 4) following a model for use of reference i n English. E x p e r i e n t i a l dyadic a c t i v i t y was related to the occurrence of exophoric (pointing out) reference and expository dyadic a c t i v i t y to the incidence of anaphoric (pointing back) reference i n the task t r a n s c r i p t s . Results of the analysis of variance indicated that while tasks d i f f e r e d on the basis of repair and reference, the groups did not: Dyadic t a l k was more responsive to the nature of the task than to the language background of the teacher. Further analysis suggested more frequent and elaborate r e p a i r during tasks which combine non-teaching goals and ex p e r i e n t i a l processes as compared with tasks emphasizing teaching goals and expository processes. Q u a l i t a t i v e analysis of task t r a n s c r i p t s supported t h i s d i s t i n c t i o n and elaborated s p e c i f i c discourse functions for such r e p a i r exponents as r e f e r e n t i a l questions and confirmation checks which c h a r a c t e r i s t i c a l l y co-occur i n conversational discourse. Based on these findings, i t was concluded that Japanese teachers are capable of generating appropriate conversational repair i n dyadic i n t e r a c t i o n with learners l a r g e l y on a par with t h e i r native English-speaking counterparts. To t h i s extent, t h e i r p o t e n t i a l contribution to learners' a c q u i s i t i o n of a foreign language i s of an equivalent value. Furthermore, teacher-led small groups can i i i be e f f e c t i v e contexts for generating a r i c h supply of conversational repair and. thus should be considered as a l t e r n a t i v e s to t r a d i t i o n a l teacher-fronted foreign language classroom i n s t r u c t i o n . F i n a l l y , tasks which support achievement of s o c i a l and problem-solving ( i . e . , non-teaching) goals through expe r i e n t i a l a c t i v i t y are e f f e c t i v e contexts i n which normal forms of conversational repair can be generated. Since such tasks can be adapted e a s i l y to classroom settings, they merit consideration among the range of task options avai l a b l e to teachers and other i n s t r u c t i o n a l planners. i v TABLE OF CONTENTS Page ABSTRACT i i TABLE OF CONTENTS V LIST OF TABLES X LIST OF FIGURES x i i ACKNOWLEDGEMENTS x i i i CHAPTER 1: INTRODUCTION 1 CHAPTER 2: FOUNDATIONS OF THE STUDY 10 F o r e i g n e r T a l k (FT) 10 F o r e i g n e r T a l k and Second Language A c q u i s i t i o n (SLA) 14 F o r e i g n e r T a l k i n I n s t r u c t i o n a l S e t t i n g s 21 I n t e r l a n g u a g e T a l k (IT) .26 D imens ions o f Task and I n t e r l a n g u a g e T a l k 30 C o n c e p t u a l D imensions of the Study 39 R e p a i r 39 Task 42 R e f e r e n c e 50 Summary 57 CHAPTER 3: RESEARCH DESIGN AND METHODOLOGY 60 The R e s e a r c h D e s i g n 60 Assumpt ions and R a t i o n a l e 60 An Overv iew o f the D e s i g n 63 D e s c r i p t i v e Measures and Dependent V a r i a b l e s used i n the Study 67 R e s e a r c h Q u e s t i o n s and Hypotheses 76 v TABLE OF CONTENTS (cont'd.) Page General Strategy for Data Analysis 80 Methodology 83 Selection and Treatment of Subjects 83 C o l l e c t i o n and Coding of Data 86 Preliminary Treatment of the Data 98 Summary 103 CHAPTER 4: QUANTITATIVE ANALYSIS OF THE DATA 104 The Descriptive Features of Talk by Group and Task....104 The Analysis of Variance: Repair by Group and Individual Task (H1-H2) 115 C l a r i f i c a t i o n Request: CCLAR 116 Comprehension Check: CCOM 117 Confirmation Check: CCON 118 D e f i n i t i o n : DDEF 120 Display Question: DDQ 121 Echo: EECH 122 Le x i c a l Uncertainty: LLEX 123 Other-expansion: OOEXP 125 Other-repetition: OOREP 125 Referential Question: RRQ 126 Self-expansion: SSEXP 127 S e l f - r e p e t i t i o n : SSREP 128 Anaphora: AANA 132 Exophora: EEXO 133 v i TABLE OF CONTENTS (cont'd.) Page The Analysis of Variance: Repair and Reference by Group for Combined and Selected Tasks....136 Repair During Ex p e r i e n t i a l and Expository A c t i v i t y (H3) 137 Anaphoric Reference During E x p e r i e n t i a l and Expository A c t i v i t y (H4a) 141 Exophoric Reference During E x p e r i e n t i a l and Expository A c t i v i t y (H4b) 143 Summary 145 CHAPTER 5: DISCUSSION OF THE ANALYSIS OF VARIANCE 148 The Use of Repair by Group and Task 148 The General Lack of Group Differences 148 The Relationship of Repair and Task: A l l o c a t i o n of Repair to Teaching and Non-teaching Tasks 150 The Use of Reference by Group and Task 157 S p e c i f i c A l l o c a t i o n s of Reference to Task 159 Repair During Combined and Selected Tasks: Towards a Framework for Complementary Task Structures 161 Reference During Combined and Selected Tasks 166 Anaphora During E x p e r i e n t i a l and Expository Tasks 167 Exophora During Ex p e r i e n t i a l and Expository Tasks 168 Summary 169 CHAPTER 6: QUALITATIVE ANALYSIS OF THE TASK TRANSCRIPTS 173 Display and Referential Questions 174 COM2: The Instructional Demonstration Task 175 v i i TABLE OF CONTENTS (cont'd.) Page DIS and LEG2: Discussion and Cooperative Problem-solving 183 Display and Referential Questions: Summarizing the Contrasts Between the Teaching and Non-teaching Tasks 197 Repair i n Complementary Task Structures 201 Group 1: Defin i t i o n s and Expressions of Lexical Uncertainty 202 Group 2: Confirmation Checks and Referential Questions 209 Summary 215 CHAPTER 7: SUMMARY, LIMITATIONS, IMPLICATIONS AND CONCLUSION 220 Summary 220 Conclusions Based on the Analysis of Variance 223 Conclusions Based on the Analysis of Transcripts 22 6 Limitations of the Study 229 Implications for Educational Practice 232 Implications for Task-based Research 236 Conclusion 237 REFERENCES 239 APPENDICES (A-K) 247 A: I n v i t a t i o n for Teachers to P a r t i c i p a t e i n the Research 248 B: I n v i t a t i o n for Learners to P a r t i c i p a t e i n the Research 249 C: Statement of Informed Consent 251 v i i i TABLE OF CONTENTS ( c o n t ' d . ) Page D: E n g l i s h Language P r o f i c i e n c y T e s t s used t o S e l e c t S u b j e c t s 253 E : I n s t r u c t i o n s t o R a t e r s and Index o f Dependent V a r i a b l e s ' . 256 F: T r a n s c r i p t i o n C o n v e n t i o n s 265 G: ANOVA F r a t i o s f o r S e l e c t e d T rans formed and Unt rans formed V a r i a b l e s L i s t e d i n T a b l e 3 266 H: S i g n i f i c a n t R e p a i r C a t e g o r i e s and S o u r c e s o f V a r i a n c e f o r E x p e r i e n t i a l and E x p o s i t o r y T a s k s U s i n g LEG1 as the E x p o s i t o r y Stem 267 I: ANOVA T a b l e s Comparing E x p e r i e n t i a l w i t h E x p o s i t o r y Tasks (C0M1 v s . LEG1 + COM2) 268 J : ANOVA T a b l e s Comparing E x p e r i e n t i a l w i t h E x p o s i t o r y Tasks (C0M1 v s . LEG2) 271 K: Means and S tandard D e v i a t i o n s f o r M i x e d - and Homogeneous-group Tasks 274 i x LIST OF TABLES Table Page 1 Summary of Conditions for Testing Hypotheses Relating to Group, Task, Reference and Repair 79 2 Level of Homogeneity Within Groups by Dependent Variable 99 3 Comparison of Selected Transformed and Untransformed Variables by Significance of ANOVA E f f e c t s 101 4 Means and Standard Deviations fo r Word-based Measures of Conversational A c t i v i t y by Mixed-group Task 106 5 Means and Standard Deviations fo r Word-based Measures of Conversational A c t i v i t y by Homogeneous-group Task 108 6 Means and Standard Deviations fo r Utterance-based Measures of Conversational A c t i v i t y by Mixed-group Task 110 7 Means and Standard Deviations fo r Utterance-based Measures of Conversational A c t i v i t y by Homogeneous-group Task I l l 8 Means and Standard Deviations fo r Turn-based Measures of Conversational A c t i v i t y by Mixed-group Task 112 9 Means and Standard Deviations fo r Turn-based Measures of Conversational A c t i v i t y by Homogeneous-group Task 114 10 E f f e c t s of Group Membership and Task on C l a r i f i c a t i o n Requests 116 11 E f f e c t s of Group Membership and Task on Comprehension Checks. 117 12 E f f e c t s of Group Membership and Task on Confirmation Checks 119 13 E f f e c t s of Group Membership and Task on Definitions 120 14 E f f e c t s of Group Membership and Task on Display Questions 121 x 15 E f f e c t s o f Group Membership and Task on Echoes 122 16 E f f e c t s o f Group Membership and Task on L e x i c a l U n c e r t a i n t y 124 17 E f f e c t s o f Group Membership and T a s k on O t h e r - e x p a n s i o n 125 18 E f f e c t s o f Group Membership and Task on O t h e r - r e p e t i t i o n 125 19 E f f e c t s o f Group Membership and Task on R e f e r e n t i a l Q u e s t i o n s 126 20 E f f e c t s o f Group Membership and Task on S e l f - E x p a n s i o n 127 21 E f f e c t s o f Group Membership and Task on S e l f - r e p e t i t i o n 129 22 S i g n i f i c a n t R e p a i r Exponents and S o u r c e s o f V a r i a n c e f o r a l l Tasks 130 23 E f f e c t s o f Group Membership and Task on A n a p h o r i c R e f e r e n c e 133 24 E f f e c t s o f Group Membership and Task on E x o p h o r i c R e f e r e n c e 133 25 S i g n i f i c a n t R e f e r e n c e C a t e g o r i e s and S o u r c e s o f V a r i a n c e f o r A l l Tasks 135 2 6 S i g n i f i c a n t R e p a i r C a t e g o r i e s and S o u r c e s o f V a r i a n c e f o r E x p e r i e n t i a l and E x p o s i t o r y T a s k s U s i n g COMl as the E x p o s i t o r y Stem 139 27 E f f e c t s o f Group Membership and S e l e c t e d Tasks on A n a p h o r i c R e f e r e n c e : C o n t r a s t i n g LEG2 w i t h COMl (EXPER-EXP0S2) 142 28 E f f e c t s o f Group Membership and S e l e c t e d T a s k s on E x o p h o r i c R e f e r e n c e : C o n t r a s t i n g LEG2 and COM2 w i t h COMl (EXPER-EXPOSl) 144 29 E f f e c t s o f Group Membership and S e l e c t e d Tasks on E x o p h o r i c R e f e r e n c e : C o n t r a s t i n g LEG2 w i t h COMl (EXPER-EXP0S2) 144 x i LIST OF FIGURES Figure Page 1 Range of contextual support and degree of cognitive involvement i n communicative a c t i v i t i e s 45 2 Extending the Knowledge Framework to problems i n observational research 54 3 F a c t o r i a l design and major conceptual d i s t i n c t i o n s used i n the study 64 4 Plot of means by task for CCLAR 117 5 Plot of means by task for CCOM 118 6 Plot of means by task for CCON 119 7 Plot of means by task for DDEF 120 8 Plot of means by task for DDQ 122 9 Plot of means by task for EECH 123 10 Plot of means by task for LLEX 124 11 Plot of means by task for RRQ 127 12 Plot of means by task for SSEXP 128 13 Plot of means by task for SSREP 129 14 Plot of means by group and task for AANA and EEXO 134 15 Plot of means by group and experiential-expository tasks for SSREP 140 16 Plot of means by group and experiential-expository tasks f o r AANA 142 17 Plot of means by group and experiential-expository tasks for EEXO 145 18 A l l o c a t i o n of four repair exponents to complementary task structures 164 x i i ACKNOWLEDGEMENTS No project of t h i s sort can be completed without the help of others. I want to mention several people i n p a r t i c u l a r whose help was unstinting and timely and who made the completion of the study a goal of t h e i r own. Bernie Mohan, my thesis advisor, has been with me from the s t a r t ; he directed more patience and i n c i s i v e thinking my way than I probably deserved. Dorothy Pedtke typed most of the t r a n s c r i p t s and ch e e r f u l l y accomplished more than she bargained f o r . Steve Ross was a big help with s t a t i s t i c s and a constant source of c r i t i c a l advice. Tom Sork and Dan Pratt, two c o r d i a l and encouraging adult educators, put me back on track and helped keep me there. My wife Taeko restored balance and common sense to the enterprise whenever I thought I'd j u s t about had my f i l l of i t . F i n a l l y , I am e s p e c i a l l y g r a t e f u l to the twenty-four volunteers, Japanese, American and B r i t i s h , who made the study possible. x i i i CHAPTER 1: INTRODUCTION This study i s about the ways teachers and learners of English as a foreign language help each other to keep t h e i r t a l k i n the language comprehensible. I t i s also about the v a r i a b i l i t y of these e f f o r t s during performance of d i f f e r e n t conversational tasks i n small, teacher-learner groups. Three observations form the basis of the study. The f i r s t i s that English i s regul a r l y used around the world by non-native speakers (NNSs) to communicate with each other: by Japanese and Kuwaiti technical s p e c i a l i s t s to rep a i r a malfunctioning hydraulic motor, by medical doctors from third-world countries attending an in t e r n a t i o n a l conference to exchange information about t h e i r s p e c i a l i z a t i o n , by multi-national residents of expatriate communities to pass the time of day. This observation i s not the same as saying that English i s a popular language, or spoken by a large number of people or that many people have studied i t i n school. I t i s more of an assertion that NNSs of English f i n d i t a useful medium f o r conducting the various facets of s o c i a l l i f e — o f exchanging information, expressing feelings and opinions, solving problems—and that they are able to do so outside of any d i r e c t experience i n cultures i n which English i s nativ e l y spoken. The second observation r e l a t e s to the f i r s t , but extends i t into the dimension of use: NNSs frequently learn 1 English, or, for that matter, any language not acquired as a mother tongue, during attempts to use i t with other NNSs i n settings completely unrelated to second or foreign language i n s t r u c t i o n . This may seem an odd assertion to make u n t i l i t i s r e a l i z e d that language i s c h a r a c t e r i s t i c a l l y acquired during use, during verbal and physical i n t e r a c t i o n between speakers and hearers, and that, on a global scale, foreign language a c q u i s i t i o n which occurs under conditions of formal classroom i n s t r u c t i o n i s a r e l a t i v e l y rare event. The usual perception of classroom i n s t r u c t i o n i s that i n s t r u c t i o n precedes learning which i n turn precedes use for ordinary communicative purposes. The s o c i a l worlds outside the classroom se t t i n g , however, make i t possible to turn t h i s procedural l i n e v i r t u a l l y on i t s head, so that use, or at le a s t attempted use, becomes the veh i c l e f o r learning. In t h i s view of language learning, p a r t i c i p a n t s i n conversations may function as "teachers", as i n t e r l o c u t o r -informants, who negotiate and repa i r t h e i r t a l k as a matter of course during elaboration of i t s pragmatic structure. The f i n a l observation extends the second. When language i s used for normal communicative purposes, i t i s very u n l i k e l y that language per se becomes the object of discussion. With the exception of some obvious examples (an inte r n a t i o n a l convention of l i n g u i s t s , perhaps), extra-classroom use of a foreign language i s most often concerned with things other than language. By contrast, i t i s most unusual to f i n d foreign 2 language i n s t r u c t i o n which does not focus e x p l i c i t l y on the language to be taught and learned, most t y p i c a l l y on the syntax and l e x i s which form the content basis f o r i n s t r u c t i o n i n a foreign language. Beyond the central r o l e of t h i s emphasis on language rather than on other areas of content i n foreign language classrooms, however, i s the more fundamental concern i n educational systems generally with the structure and organization of i n s t r u c t i o n . Even when language and content are merged into a common syllabus, i t i s the t r a d i t i o n a l teacher-fronted lesson which contextualizes the use of language i n the classroom and has s p e c i f i c consequences for the ways i n which learners may employ i t . These observations are intended to suggest the outline of a study concerned with how teachers (both native speakers and NNSs of English) and learners (NNSs of English) employ the language i n various dyadic settings of p o t e n t i a l use to planners of foreign language i n s t r u c t i o n a l syllabuses. These settings range from those which most resemble teacher-led, content-based i n s t r u c t i o n to those which resemble the kind of information exchange and problem-solving which occur when s o c i a l , non-instructional goals, predominate. The focus of the study i s repair, defined broadly here as the ways i n t e r l o c u t o r s use language to help themselves and each other make sense of t h e i r discourse as i t unfolds. The ce n t r a l purpose of the study i s to understand how learners and teachers re p a i r each other's t a l k during performance of 3 tasks expressly organized to accomplish i n s t r u c t i o n a l and non-instructional goals. A secondary, related purpose i s to d i s t i n g u i s h among tasks on empirical grounds—to propose a framework fo r the choice among teacher-led tasks which may eventually be of use to educational planners i n designing a c q u i s i t i o n - r i c h environments i n foreign language classrooms. Why r e p a i r within NNS-NNS discourse i s worth studying i s the point of the discussion to follow i n Chapter 2, the foundation of the e n t i r e study. For the moment, however, i t may be useful to note that a small body of evidence and argumentation points to the p o t e n t i a l l y b e n e f i c i a l e f f e c t s of NNSs cooperatively attempting to r e p a i r t h e i r t a l k i n small groups on t h e i r a c q u i s i t i o n of a second language (Duff, 1986; Gass & Varonis, 1985b; Porter, 1983, 1986; Rulon & McReary, 1986, Varonis and Gass, 1985). According to these studies, i t i s unscripted i n t e r a c t i o n between NNSs, and to a l e s s e r extent between native speakers (NSs) and NNSs, which seems to produce the conditions f o r negotiation and r e p a i r of the discourse (much as i s the case with ordinary conversational a c t i v i t y ) through such t a c t i c s as requests f o r c l a r i f i c a t i o n and indications of l e x i c a l uncertainty. Behind t h i s l e v e l of discussion i n the l i t e r a t u r e , however, i s a widely held assumption (asserted i n deductive terms i n Long, 1981; see also Long, 1983a, 1983b, 1985) that as opportunities f o r negotiation and r e p a i r i n a second language increase, the comprehensibility 4 of the language to which a learner attends also increases. Given s u f f i c i e n t i n t e r a c t i o n of t h i s sort, learners eventually acquire various forms of grammatical, pragmatic and s t r a t e g i c competence which can be said to comprise second language competence. Taken together, these forms of competence comprise a l e v e l of s o c i a l i z a t i o n which i s u n l i k e l y to be achieved through focusing the attention of language learners on the formal properties of a target language. Along these l i n e s , a number of second language a c q u i s i t i o n studies have pointed out the r e l a t i v e i n e f f i c i e n c y of second language i n s t r u c t i o n conducted by teachers i n t r a d i t i o n a l roles as di r e c t o r s of verbal exchange i n the classroom (Long, Adams, & Castanos, 1976; Long & Porter, 1985; Long & Sato, 1983; Pica, 1987; Pica & Doughty, 1985). Unfortunately no studies have examined systematically an al t e r n a t i v e r o l e f o r the NNS teacher of English as a conversational p a r t i c i p a n t and problem-solver i n cooperation with NNS l e a r n e r s — t h e kind of ro l e which i s frequently performed by NNSs i n commercial, te c h n i c a l and s o c i a l exchanges around the world. Moreover, none has compared the differences i n repair behavior, i f any, which might be found between groups which contain native and non-native teachers of English, i n addition to non-native learners. Although i t may be i n t e r e s t i n g to speculate about NNSs as prime candidates f o r "teachers" of language through n o n - l i n g u i s t i c content, native speakers are 5 t y p i c a l l y viewed as having the advantage as teachers because of t h e i r native competence. Studies focused s p e c i f i c a l l y on t h i s point may help foreign language p o l i c y planners to formulate p o l i c y based on empirical research. In addition to approaching some of these unresolved issues i n second language a c q u i s i t i o n research and p o l i c y planning, there i s also the opportunity to apply recently developed models of the forms of discourse occurring i n f i r s t and second language i n s t r u c t i o n a l s i t u a t i o n s (Cummins, 1983; Mohan, 1986) to problems i n foreign language i n s t r u c t i o n a l planning. I t i s s t i l l very much an open question as to which communication contexts and a c t i v i t i e s i n classrooms best promote the learning of foreign languages. Although i t i s now f a i r l y unexceptional to assert, f o r example, that negotiated i n t e r a c t i o n i s useful fo r language a c q u i s i t i o n , questions remain over the p a r t i c u l a r forms of t a l k which various a c t i v i t i e s may engender. How might foreign language classrooms be organized, f o r example, to promote the kinds of verbal i n t e r a c t i o n believed to be at the basis of second language acquisition? Given t h i s sort of planning problem, Mohan's (1986) formulation of verbal a c t i v i t y i n educational settings, ranging from discourse which emphasizes "general, t h e o r e t i c a l knowledge" on the one hand to " s p e c i f i c , p r a c t i c a l knowledge" on the other (p. 40), becomes a useful point at which to i n i t i a t e the research. This study 6 examines the evidence of repair within a l a r g e l y t h e o r e t i c a l d i s c o u r s e / p r a c t i c a l discourse framework. I t thus attempts to apply a system f o r thinking about second language educational discourse to a p r a c t i c a l problem encountered when planning the foreign language i n s t r u c t i o n a l syllabus: e s t a b l i s h i n g c r i t e r i a f or organizing t a l k i n foreign language classrooms. Chapter 3 describes the study's methodology; i t s main function i s to d e t a i l the pattern within which the study's f i v e hypotheses are operationalized. Accordingly, a 2 x 5 repeated-measures f a c t o r i a l design i s outlined at the beginning of the chapter and related to the fundamental concepts and studies discussed i n Chapter 1. The twelve rep a i r exponents and two forms of conversational reference which serve as dependent variables i n the study are examined i n considerable d e t a i l . Because the study i s based on an experimental design, s p e c i a l emphasis i s placed on the procedures and strategies employed to carry i t out. Chapter 4 begins with a discursive summary of the means and standard deviations for nine, non-inferential d e s c r i p t i v e codings appended to the task t r a n s c r i p t s , including word-based, turn-based and utterance-based measures of t a l k during the tasks. The major function of the chapter, however, i s to report the r e s u l t s of the tests of the hypotheses outlined i n Chapter 2 and c a r r i e d out through analysis of variance, the quantitative methodology by which the dyads' production of repair and reference on 7 f i v e t a s k s i s compared. C h a p t e r 5 d i s c u s s e s and i n t e r p r e t s t h e r e s u l t s f o l l o w i n g the t o p i c a l f o c u s o f each r e s e a r c h q u e s t i o n . The major p o i n t s r a i s e d d u r i n g the d i s c u s s i o n i n c l u d e a r a t i o n a l e f o r a l l o c a t i o n o f r e p a i r and r e f e r e n c e t o the v a r i o u s t a s k s and a d i s t i l l a t i o n o f t h e r e p a i r exponents i n t o complementary t a s k s t r u c t u r e s . The n o t i o n o f complementary t a s k s t r u c t u r e s i s put i n t o t h e p e r s p e c t i v e o f a t e n t a t i v e , e x p l o r a t o r y framework f o r p r e d i c t i n g how t a s k s i n e d u c a t i o n a l s e t t i n g s i n f l u e n c e the p r o d u c t i o n o f r e p a i r . C h a p t e r 6 p r o v i d e s a q u a l i t a t i v e a n a l y s i s o f t h e t a s k t r a n s c r i p t s . T h i s d i s c u r s i v e t r e a t m e n t o f r e p a i r ex tends t h e r e s u l t s o f t h e q u a n t i t a t i v e a n a l y s i s by f o c u s i n g on the t h e f u n c t i o n a l q u a l i t i e s o f the most s a l i e n t r e p a i r exponents p r o d u c e d d u r i n g p a r t i c u l a r t a s k s . Formal and f u n c t i o n a l compar isons between exponents a r e made w i t h e x t e n s i v e r e f e r e n c e t o the t r a n s c r i p t s . Core groups o f r e p a i r exponents a l l o c a t e d t o the complementary t a s k s t r u c t u r e s a r e examined as c o - o c c u r r i n g u n i t s w i t h i n t h e t r a n s c r i p t s and r e l a t e d t o the g o a l s and p r o c e d u r e s o f the t a s k s i n wh ich t h e y a r e c h a r a c t e r i s t i c a l l y f o u n d . C h a p t e r 7 summarizes and c o n c l u d e s t h e s t u d y . Because the s t u d y i s o f p o t e n t i a l use t o a p p l i e d l i n g u i s t s and t o o t h e r s i n t e r e s t i n problems o f f o r e i g n language i n s t r u c t i o n a l d e s i g n , a t t e n t i o n i s d i r e c t e d t o i t s l i m i t a t i o n s and i m p l i c a t i o n s f o r e d u c a t i o n a l p r a c t i c e . F i n a l l y , s p e c i a l encouragement i s o f f e r e d t o t e a c h e r s who 8 wish to undertake t h e i r own task-based studies of problems i n foreign language education. 9 CHAPTER 2: FOUNDATIONS OF THE STUDY This chapter summarizes conceptual and empirical work accomplished i n the areas of foreigner t a l k , interlanguage t a l k (talk between learners of a second or foreign language), repair, task and reference. The discussion w i l l emphasize the si g n i f i c a n c e of the work to t h i s study and prepare the ground for a description of the research design and methods i n Chapter 3. The nature of task and i t s r e l a t i o n s h i p to conversational repair w i l l be c l o s e l y examined, as w i l l the bases f o r s e l e c t i n g task categories employed during implementation of the research design. Foreigner Talk (FT) Ferguson's seminal characterizations of FT (1971, 1975) are important contributions to current understanding of how native speakers (NSs) and NNSs communicate with each other. Ferguson (1971, p. 143) described FT as "a r e g i s t e r of s i m p l i f i e d speech . . . which i s used by speakers of a language to outsiders who are f e l t to have a very l i m i t e d command of the language or no knowledge of i t at a l l . " FT i s thus geared to an appraisal of the NNS i n t e r l o c u t o r which the NS makes during conversation, very probably during the f i r s t moments of contact but also following the NS's assessment of the NNS's comprehension of the ongoing discourse (see also Gass and Varonis, 1985b). 10 Ferguson (1975) has also described how NSs of English adapt t h e i r speech to NNSs, or at le a s t how NSs would adapt, given a set of constraints on the speech s i t u a t i o n . The study i s an i n d i r e c t approach to the use of FT i n the sense that the NSs i n the study, a l l members of a s o c i o l i n g u i s t i c s course, were asked to rewrite 10 sentences i n ordinary English as i f they were speaking them on behalf of t h e i r NS group to a group of uneducated, non-European foreigners. Ferguson also excerpted sixty-one sentences from C. S. Lewis' novel Out of the S i l e n t Planet. These sentences were selected because they exemplified speech of an Englishman speaking English foreigner t a l k to Martians, that i s , the nov e l i s t ' s concept of FT. The excerpted sentences were then compared with the 10 sentences modified for native-foreigner communication. The r e s u l t s are i n t e r e s t i n g i n that they presage some of the actual modifications described i n l a t e r empirical studies (e.g., Freed, 1978, below). Ferguson found several major categories of modification both i n the novel and on the re-writing tasks—omissions, expansions, and replacements or rearrangements. Examples of omission included dropping the d e f i n i t e a r t i c l e 'the', omitting various forms of the verb 'to be', avoiding i n f l e c t i o n a l s u f f i x e s s i g n a l l i n g case and number, and elimination of coordinating and subordinating conjunctions. Expansions included addition of 'you' to imperatives, p a r t i a l r e p e t i t i o n of i n i t i a l sentences or phrases, and addition of 11 tags (you come. OK?, f o r example). Replacements included s u b s t i t u t i n g 'no' f o r a l l negative constructions (I no understand, f o r i n s t a n c e ) — b u t 'not' for contracted negative forms, changing normal nominative pronouns to accusative forms (me Tarzan, you Jane, an u n l i k e l y utterance i n ordinary conversation!—see Hatch 1983, p. 175f), l e x i c a l s u b s t i t u t i o n and a n a l y t i c paraphrase (one day gone to replace yesterday, papa for father), unmarked or more frequently occurring synonyms (take for carry), and decomposition of words into phrases with s i m i l a r semantic content (which place f o r where, bier head f o r leader) . On the basis of the l i m i t e d body of evidence a v a i l a b l e to him, Ferguson speculated that NSs acquire the FT r e g i s t e r as c h i l d r e n and r e t a i n the r e g i s t e r f o r s p e c i a l s i t u a t i o n s of contact with native speakers of another language (that i s , "foreigners"), modifying t h e i r speech i n a systematic, rule-governed way (FT as a conventionalized use of language, (Ferguson, 1975, p. 11). He d i d not, however, examine the possible functions of FT as a v e h i c l e f o r language input which would encourage a c q u i s i t i o n of a second language by the NNS, nor did he explore the communicative value of s i m p l i f i c a t i o n by the NS. In an exhaustive study, Freed (1978; see also Freed, 1980) examined the s t r u c t u r a l and functional q u a l i t i e s of FT produced i n 11, two-member conversation groups. The study drew extensively from Newport's (1976) data on English baby t a l k or motherese, the s i m p l i f i e d r e g i s t e r adults employ 12 with infants acquiring English as t h e i r f i r s t language. Freed compared FT with native t a l k (NT), i . e . , NS's speech i n casual conversation with other NSs, NT with baby ta l k , and FT with baby t a l k . Using the utterance as the basic uni t of speech segmentation and the number of S-nodes per utterance and per sentence (measures of propositional complexity based on the number of main verbs i n the segment under a n a l y s i s ) , Freed concluded that English FT i s i n fact a r e g i s t e r d i f f e r i n g from normal English i n terms of the r e l a t i v e frequencies of various forms and functions (1978, p. 235). She found, for example, that FT contained s i g n i f i c a n t l y more sentence fragments and stock expressions, and fewer grammatically acceptable utterances, than NT. In terms of s y n t a c t i c complexity, FT had s i g n i f i c a n t l y fewer S-nodes and sentences per utterance, whereas the mean length of utterance was s i g n i f i c a n t l y longer i n NT. Similar r e s u l t s obtained at the sentence l e v e l : fewer S-nodes per sentence and shorter mean length of sentence. Americans t a l k i n g to foreigners used s i g n i f i c a n t l y more Wh- and Yes/No questions, but f a r fewer declarative sentences than Americans t a l k i n g to other Americans. Further analysis revealed that the Yes/No questions employed during FT were s i g n i f i c a n t l y more l i k e l y to have no subject-auxiliary inversion, and to contain deletions of do and/or you, than NT. Freed's functional analysis highlighted information exchange as the main purpose of both FT and NT, although 13 behavior i n d i c a t i n g a need to keep the conversation going was e s p e c i a l l y evident i n FT. For example, NSs speaking to NNSs used more conversation continuers (e.g., mmm, r e a l l y ) to show i n t e r e s t i n or attention to t h e i r partner's utterances undergoing a sometimes tortuous construction. FT, moreover, was much more l i k e l y to be c l a r i f i e d than NT. NSs' attempts at c l a r i f i c a t i o n included repeating previous utterances i n whole or i n part, and paraphrasing previously used words and phrases. NSs engaged i n FT also emphasized such conversational support as supplying a word or phrase to the NNS when needed. Freed's discussion of natu r a l l y occurring FT stressed the underlying functional s i m i l a r i t y of FT and NT, i n addition to the s p e c i f i c d i s s i m i l a r i t i e s , and thus supported Ferguson's e a r l i e r characterization of FT as a r e g i s t e r (rather than a dia l e c t ) which, she noted, i s intended by i t s users to improve the qu a l i t y of information and s o c i a l exchange during a p a r t i c u l a r conversation. Unlike the control and d i r e c t i v e functions of baby t a l k used i n mother-infant exchanges, FT i s "motivated by the need to i n i t i a t e and maintain conversation appropriate to the s o c i a l and cognitive presence of . . . foreign partners. That i s , the Americans saw t h e i r foreign partners as conversational peers . . . ." (p. 2 3 6). Foreigner Talk and Second Language A c q u i s i t i o n (SLA) This notion of in t e r a c t i o n between in d i v i d u a l s with developed cognitive a b i l i t i e s and the competence to explore 14 t o p i c s f r e e l y through verbal exchange i s also central to the FT research which focuses on SLA. Krashen (1980, 1982), for example, argued that NS input directed to learners i s made more comprehensible through conversational negotiation and eventually thus leads to SLA. Long (1980, 1981), among others, even more emphatically emphasized the r o l e of NS-NNS in t e r a c t i o n which occurs during two-way conversational exchange i n a c q u i s i t i o n of a second language. Hatch (1983) outlined a middle p o s i t i o n which puts conversational and classroom i n t e r a c t i o n at the source of input modifications leading to SLA, although the extent of negotiation and modification seems c l e a r l y r elated to such variables i n the communicative environment as task (Crookes, 1986; Duff, 1986; Long, 1980; Long, 1985a; Pica, 1987) and pr o f i c i e n c y or apparent comprehension of the NNS as assessed by the NS (Long, 1983a; Long & Pica, 1986; Long & Porter 1985; Varonis & Gass, 1982). This emphasis on i n t e r a c t i o n — o n i t s sources and e f f e c t s — i s l a r g e l y missing from the early FT l i t e r a t u r e . More recent discussions, however, place NS-NNS in t e r a c t i o n at the center of the SLA process ( a l b e i t by deduction more than by evidence) and stress the importance of NS responsiveness to the perceived conversational needs of the NNS partner (Long, 1983a). A related body of l i t e r a t u r e i n the area of communication strategies has also found an i n t e r a c t i o n perspective useful i n developing t h e o r e t i c a l accounts of how learners use t h e i r interlanguage with i n t e r l o c u t o r s i n communicative, problem-posing 15 s i t u a t i o n s to negotiate common understandings (Tarone, 1983; also Faerch & Kasper, 1983; Haastrup & P h i l l i p s o n , 1983; Wagner, 1983). The d i v e r s i t y of NS responsiveness i s well documented by Long (1980, 1981; see also S c a r c e l l a & Higa, 1981). Although early description of FT tended to stress s i m p l i f i c a t i o n of input to NNSs as the predominant means of conveying meaning during conversation, Long (1980), argued that s i m p l i f i c a t i o n of input was only one type of modification NSs are l i k e l y to make when speaking to NNSs. The more important l e v e l of conversational a c t i v i t y — important from the perspectives of both SLA and i n s t r u c t i o n a l methodology—was shown to be i n t e r a c t i o n a l modifications constructed cooperatively by conversational partners. Long found that questions occurred s i g n i f i c a n t l y more frequently i n NS-NNS dyads than i n NS-NS dyads during conversational tasks requiring exchange of information (two-way tasks). Questions t y p i c a l l y took the form of confirmation checks, comprehension checks, c l a r i f i c a t i o n requests and other-repetition, and served to sustain the conversation by increasing the NNS's p a r t i c i p a t i o n . Long also found s i g n i f i c a n t differences between NS-NNS and NS-NS dyads i n the frequency of s e l f - r e p e t i t i o n s and expansions of previously occurring utterances. Long explained that NS-NNS i n t e r a c t i o n i s characterized, among other things, by communication breakdowns. Confirmation checks, c l a r i f i c a t i o n requests, s e l f - and other-16 r e p e t i t i o n s are a l l i n t e r a c t i o n a l resources a v a i l a b l e to the NS (and to NNSs) to repair the discourse when breakdowns occur. Comprehension checks, s e l f - and other - r e p e t i t i o n are among the devices NSs can use to avoid breakdowns, and so may be expected to be more frequent where communicative trouble i s anticipated, as i s the case with much NS-NNS in t e r a c t i o n . (p. 152) F i n a l l y , Long compared groups of tasks which required information exchange with those which did not. Again, he found a s i g n i f i c a n t l y greater frequency of the "repair and trouble-avoidance devices" (p. 152) i n the information exchange group of tasks. These r e s u l t s point to the range of modification speakers have at t h e i r disposal and a c t u a l l y do invoke to continue the exchange of information. C l e a r l y NS input to NNSs i n the form of s i m p l i f i e d speech i s not the only, nor even apparently most important, means of maintaining the NNS's a c c e s s i b i l i t y to the target language. Long (1981, p. 275) makes the p o s i t i o n i n favor of modified i n t e r a c t i o n e x p l i c i t : " p a r t i c i p a t i o n i n conversation with NS, made possible through the modification of int e r a c t i o n , i s the necessary and s u f f i c i e n t condition f o r SLA." Triggers to Modification of FT Discussion of NS responsiveness has also turned to the question of what tr i g g e r s or otherwise influences FT both p r i o r to and during conversation. This s p e c i f i c kind of NS adjustment to NNSs was examined i n Gass and Varonis (1984), 17 Gass and Varonis (1985b), Long (1983a), Long (1985a), Pica and Long (1986) and Varonis and Gass (1982) . Gass and Varonis (1984) found that NNS speech i s more l i k e l y to be understood, and thus l e s s l i k e l y to be negotiated, by NSs who were f a m i l i a r with NNS speech. This f i n d i n g suggests that NSs who are teachers may tre a t the language of t h e i r NNS students d i f f e r e n t l y — a n d thus modify t h e i r input d i f f e r e n t l y — t h a n NSs who have no sp e c i a l conversational experience with NNSs. Pica and Long (1986), however, found i n general no re l a t i o n s h i p between years of teaching experience and such input features as the d i s t r i b u t i o n of questions, statements and imperatives, and the length or synta c t i c complexity of teachers 1 utterances which are direct e d towards t h e i r NNS students. On the other hand, Pica and Long d i d f i n d that experienced teachers use various other FT features more frequently than inexperienced teachers, including more yes/no questions and fewer Wh-questions, although experienced teachers employ one device of conversational adjustment, other-repetition, s i g n i f i c a n t l y more frequently than inexperienced teachers. The evidence i s thus mixed on the question of whether the teacher's experience influences the kind and frequency of FT directed towards the NNS. On the issue of how NS perceptions of NNS p r o f i c i e n c y influence NS conversational adjustment, Varonis and Gass (1982) found that whereas such factors as NNS pronunciation and grammar seem to t r i g g e r modifications i n the language of 18 NSs, the physical appearance of the NNSs does not. They concluded that NNS pronunciation and grammar were the major contributors to the comprehensibility of NNS speech to NS conversation partners. In other words, NNS's pr o f i c i e n c y i n the second or foreign language, as r e a l i z e d i n pronunciation and grammar, seems to be one basis f o r NS's use of FT. Comprehensibility of NNS speech as established through NS sampling of the NNS's pronunciation and grammar at the outset of a conversation, however, would seem to be only one among a number of sources of adjustment. Given the in t e r a c t i v e nature of discourse i n NS-NNS conversation, the l e v e l of comprehension which NSs a t t r i b u t e to NNSs during a conversation could also be a major factor leading to q u a l i t a t i v e and quantitative adjustments i n NS speech. Along t h i s l i n e of thinking, Long (1983a) has also examined the causes of NSs l i n g u i s t i c and conversational adjustments to NNSs, including the perceived foreignness of the NNS, features of the NNS's interlanguage, the NS's perception of the NNS's comprehension of the NS's speech, i n addition to the NS's perception of the comprehensibility of the NNS's speech. Based on a summary of l i t e r a t u r e devoted to NS conversational adjustments to NNSs, Long argued that a combination of factors lead to adjustment—comprehensibility of the NNS's interlanguage, the l i n g u i s t i c q u a l i t i e s of the interlanguage, and, s i g n i f i c a n t l y , NS comprehension of the NNS. Comprehension, or the lack of i t , occurs throughout a conversation and i s at the basis of conversational 19 adjustment. I t i s the qu a l i t y of a conversation as discourse, not as a c o l l e c t i o n of i s o l a t e d utterances, which permits re p a i r or avoidance of misunderstanding. Thus the feedback which NNSs provide during conversation i s , as Long noted, an important source f o r NS evaluation of NNS comprehension. What a c t u a l l y t r i g g e r s NS adjustment when NNSs sign a l trouble during a conversation? In a study which con t r o l l e d f o r the e f f e c t s of grammar, vocabulary and physical appearance, and which responded to the issue of ongoing adjustment of FT discourse, Gass and Varonis (1985b) focused on the e f f e c t of NNS comprehensibility and pr o f i c i e n c y as factors i n NS speech modification. Their study used data from Abunahleh et a l . (1982) i n which eight NNSs at pr o f i c i e n c y l e v e l s ranging from beginning to intermediate each made random telephone c a l l s to NSs. The NNS c a l l e r s followed a s c r i p t of eight questions on food preparation and consumption, with the t h i r d and seventh questions requiring the c a l l e r to say Pardon me? to whatever the NS responded. This technique was designed to e l i c i t a c l a r i f i c a t i o n from the NS and thus constituted a d i r e c t means of determining the qu a l i t y and quantity of NS modification r e s u l t i n g from NS perception of NNS pr o f i c i e n c y over the course of the conversation. The study showed that the frequency of negotiation exchanges—"exchanges i n which there i s some overt i n d i c a t i o n that understanding between p a r t i c i p a n t s has not 20 been complete and . . . a resultant attempt to c l a r i f y the nonunderstanding" (Gass & Varonis, 1985b, p. 3 9 ) — i s related to p r o f i c i e n c y ; NSs i n i t i a t e d negotiation routines about three times more frequently with low-level NNSs than with high-level NNSs. The quantity of speech, moreover, seems at le a s t p a r t i a l l y r elated to profic i e n c y : Following the f i r s t request f o r c l a r i f i c a t i o n , s i g n i f i c a n t l y more speech was direc t e d to high-level subjects than to low-level ones. Gass and Varonis suggested that t h i s a dditional amount of speech resulted from a more severe reassessment of the high-l e v e l speakers' p r o f i c i e n c y than of the low l e v e l speakers' p r o f i c i e n c y . Thus the authors concluded that "perceived" comprehensibility t r i g g e r s NS speech modification (p. 55), although i t should also be noted that the comprehensibility NSs a t t r i b u t e to NNS speech during a conversation seems re l a t e d to ongoing revisions of t h e i r i n i t i a l perceptions about a speaker's p r o f i c i e n c y — t h a t i s , to t h e i r perception of the NNS's comprehension. Foreigner Talk i n Instructional Settings One f i n a l area of research into FT and what has come to be c a l l e d foreigner t a l k discourse (Long, 1980, 1981, 1983a), i s the study of t a l k i n classrooms between NS teachers and NNS students. In general t h i s l i n e of research compares the discourse i n teacher-fronted lessons, which o r d i n a r i l y stress language i n s t r u c t i o n , with the non-i n s t r u c t i o n a l discourse occurring i n small groups of NNS. For the present, however, t h i s discussion w i l l look only at 21 those c h a r a c t e r i s t i c s of FT which other studies have associated with second and foreign language i n s t r u c t i o n a l s ettings. Chaudron's (1983) study of FT i n high school and un i v e r s i t y subject-matter classes f o r English as a second language (ESL) students examined how teachers s i m p l i f i e d classroom language l i n g u i s t i c a l l y and c o g n i t i v e l y (cf. Ferguson, 1971). Chaudron selected vocabulary, anaphoric ("pointing back") reference, topic development, explanations and questions f o r q u a l i t a t i v e analysis. He found that teachers attempt to simplify vocabulary by elaborating on i t and making i t much more redundant than i n non-ESL classrooms. He also noted, however, that elaboration can create new meanings for learners to deal with and thus lead to ambiguity. S i m p l i f i c a t i o n through anaphoric reference was p a r t i c u l a r l y problematic. Although teachers apparently did not hesitate to use anaphoric pronouns, they tended to complicate the comprehensibility of t h e i r explanations by overuse of such pronouns and by assuming that students had learned the appropriate rules for r e l a t i n g referent to pronoun when, i n fact, they had not. Similar problems were noted with regard to marking and changing topics, and sim p l i f y i n g i n s t r u c t i o n s . Chaudron found, for example, that teachers' elaborations sometimes entailed excessive and confusing rephrasing or excessive redundancy. Chaudron (1983) also found that teachers' " s p e c i f i c procedural" or "obliquely l o g i c a l " questions (p. 135) 22 directed towards ESL students—presumably intended to sim p l i f y the structure of knowledge that a teacher wants to convey—often exceeded the a b i l i t y of the ESL students to process the language directed towards them. Others have noted the higher frequency of questions i n FT generally (Freed, 1978, for example) and of questions intended by teachers i n second and foreign language classrooms to t e s t learner' knowledge (see Long & Sato, 1983, summarized below) as compared with ordinary conversational and i n s t r u c t i o n a l settings. Chaudron, however, has a d d i t i o n a l l y r aised the issue of how accurately teachers are able to p i t c h t h e i r FT discourse to NNSs i n i n s t r u c t i o n a l settings. The frequency and functions of teachers 1 questions i n second language classrooms was examined by Long and Sato (1983). The authors pointed out the importance of questions i n FT discourse to sustaining NNS p a r t i c i p a t i o n i n conversation by s i g n a l l i n g an open turn, making conversational topics c l e a r e r by "compelling" responses and, generally, i n opening opportunities to modify the i n t e r a c t i o n a l structure of conversation. Classroom data comprising t r a n s c r i p t s of s i x elementary l e v e l ESL lessons were compared with t r a n s c r i p t s of conversations between NSs and NNSs i n 36 dyads, or two-member groups. Among the t o t a l of 938 questions i n the classroom corpus, Long and Sato found a s i g n i f i c a n t l y larger number of display questions than r e f e r e n t i a l questions, i . e . , questions f o r which the teacher already knows the answer over questions designed to 23 e l i c i t unknown information or to check or otherwise c l a r i f y conversational material. Moreover, the frequency of display questions was s i g n i f i c a n t l y higher i n the classroom speech than i n the conversational dyads. Other comparisons between the i n s t r u c t i o n a l and non-i n s t r u c t i o n a l settings showed further s i g n i f i c a n t differences. R e f e r e n t i a l questions, for example, constituted more than three-quarters of the t o t a l number of questions i n the NS-NNS dyads, but only 14 percent of the t o t a l number of questions i n the ESL classes. At the same time, the teachers asked fewer questions o v e r a l l than the NSs i n conversational settings. S i g n i f i c a n t differences were also obtained f o r verbal marking of present and non-present temporal reference: The ESL teachers preferred verbs marked for present tense by a s i g n i f i c a n t margin over NSs i n the NS-NNS dyads. Long and Sato concluded that i n s t r u c t i o n a l t a l k i n second language classrooms i s a greatly d i s t o r t e d version of i t s NS-NNS counterpart i n conversational settings and suggested that i f the difference i s important i n terms of SLA, as they c l e a r l y think i t i s , further research be conducted to determine "how the i n t e r a c t i o n a l structure of classroom NS-NNS conversation can be changed" (p. 284). Additional evidence of the r e l a t i v e l y d i s t o r t e d nature of FT i n i n s t r u c t i o n a l s e t t i n g s — r e l a t i v e to treatment of NS students i n content area classrooms—was reported by Shinke-Llano (1983). Shinke-Llano examined teacher t a l k directed to both NSs and NNSs i n f i f t h and s i x t h grade classes. The FT 24 directed to the students of li m i t e d English p r o f i c i e n c y (LEP) provided s i g n i f i c a n t l y less i n t e r a c t i o n than the "normal" i n s t r u c t i o n a l r e g i s t e r used for non-LEP students. The attention that LEP students did receive tended to be managerial rather than i n s t r u c t i o n a l , and, i n general, much b r i e f e r than that received by t h e i r t h e i r non-LEP counterparts. This l e v e l and qual i t y of FT suggests a conclusion s i m i l a r to the ones reached by Chaudron (1983) and Long and Sato (1983) for a d u l t - l e v e l i n s t r u c t i o n , namely that the i n s t r u c t i o n a l r e g i s t e r which teachers t y p i c a l l y employ f o r NNSs i s q u a l i t a t i v e l y and qua n t i t a t i v e l y d i f f e r e n t from the la r g e l y well-modified v a r i e t y of FT which occurs i n non-i n s t r u c t i o n a l settings. Although findings have reported appropriate adjustment of the teacher's classroom speech to NNSs l i s t e n i n g to s t o r i e s (Henzl, 1974, 1979) and lectures (Wesche & Ready, 1985)—i.e., learners mainly attending to the teacher's expository behavior—the weight of evidence suggests, to the contrary, that FT i n classrooms i s a r e l a t i v e l y i n e f f i c i e n t medium by which to a s s i s t construction of discourse which i s useful to language learners (see also Long, Adams, & Castanos, 1976; Long & Porter, 1985). FT occurring i n natural or non-instructional settings seems better adjusted to ongoing discourse and to the NS's perception of the NNS's l e v e l of understanding, generalizations which suggest a possible r o l e f o r non-i n s t r u c t i o n a l conversational tasks between NSs and NNSs i n 25 second and foreign language classrooms. The discussion so fa r has examined the nature and functions of FT i n a v a r i e t y of contexts and noted the uses of n a t u r a l l y modified NS speech i n helping to sustain NNS p a r t i c i p a t i o n i n conversation. The NS's repa i r or avoidance of troublesome conversational material i s part of t h i s process (Long, 1980). However, the ro l e of other NNSs as conversational partners and sources of input and i n t e r a c t i o n i n a second or foreign language has not yet been considered. The p o s s i b i l i t y that NNSs could function i n much the same way as NSs for other NNSs, and that they could contribute to a learner's a c q u i s i t i o n of a second language, has received some attention i n the l i t e r a t u r e . I t i s to t h i s small but important body of l i t e r a t u r e that the discussion turns next. Interlanguage Talk (IT) Conversation between NNSs i n a non-native language has been va r i o u s l y described as Interlanguage Talk (IT) (Krashen, 1980, 1981, 1982; Long & Porter, 1985), Interlanguage Communication (the term can also r e f e r , non-s p e c i f i c a l l y , to the developmental character of learners' t a l k to ei t h e r NNSs or NSs—see Faerch & Kasper, 1983) and Learner Language (Porter, 1983). IT has received attention recently i n the SLA l i t e r a t u r e because, l i k e FT discourse, i t apparently increases opportunities to negotiate meaning during conversational exchange, thus leading i n p r i n c i p l e to SLA. (No unambiguous evidence yet exists for t h i s claim, 26 although most SLA researchers argue a strong case f o r the causative p o s i t i o n of negotiated discourse i n SLA. (See Long, 1981, 1985a). IT has also been examined f o r i t s p o t e n t i a l as a pedagogical t o o l i n second language classrooms, p a r t i c u l a r l y as an al t e r n a t i v e to teacher-fronted forms of d e l i v e r i n g i n s t r u c t i o n a l material. Studies of FT discourse, i t w i l l be r e c a l l e d , have noted the r e l a t i v e i n e f f i c i e n c y of FT used by teachers i n second language classrooms i n comparison with FT directed to learners i n non-instructional settings. Can IT provide the same opportunities for int e r a c t i o n , negotiation of meaning and r e p a i r as non-instructional FT? What are some of the l i m i t a t i o n s of IT as a medium for possible second language a c q u i s i t i o n and what are i t s l i m i t a t i o n s as a method of organizing instruction? Porter (1983; also 1986) compared t a l k generated by dyads during problem-solving tasks at three l e v e l s of pr o f i c i e n c y i n English f o r s p e c i f i c c h a r a c t e r i s t i c s of input, i n t e r a c t i o n and appropriateness. Six intermediate learners, s i x advanced learners and s i x native speakers of English were paired so that in d i v i d u a l s spoke with others at t h e i r own l e v e l and at the other two l e v e l s . A l l learners were native speakers of Spanish. The 27 dyadic conversations centered on a frequently used i n s t r u c t i o n a l technique requiring i n d i v i d u a l s to rank order a l i s t of solutions to a problem or items which could be used i n the the s o l u t i o n of a problem ("You have j u s t crash landed i n 27 the Sonora desert . . . ." Porter, 1983, p. 217), and then to discuss t h e i r ranking with a conversational partner. Each p a r t i c i p a n t negotiated a preferred s o l u t i o n f o r each of three d i f f e r e n t tasks with a d i f f e r e n t conversation partner. The tape recorded discussions were transcribed and rated by teams of judges for such q u a l i t i e s as comprehensibility by rater, fluency, pronunciation, grammaticality, and l e x i c a l p r e c i s i o n and breadth. In addition, t r a n s c r i p t i o n s of the discussions were coded and analyzed for t o t a l words, the percentage of words contributed by each p a r t i c i p a n t , and the number of f a l s e s t a r t s (a greater frequency f o r t h i s measure of fluency, Porter hypothesized, would r e s t r i c t the l i s t e n e r ' s comprehension). The t r a n s c r i p t i o n s were also coded f o r monitor—the speakers' attention to the q u a l i t y of t h e i r own and others' speech (as measured by the frequency of s e l f - and other-corrections of grammatical and l e x i c a l e r r o r s ) , o t her-repetition rate (a measure of comprehension), rep a i r rate (a measure of negotiation i n the discussion including c l a r i f i c a t i o n requests, confirmation checks, v e r i f i c a t i o n s of meaning, d e f i n i t i o n requests, indications of l e x i c a l uncertainty, and comprehension checks), and for the prompt rate (a measure of conversational cooperativeness and willingness to keep the conversation going). F i n a l l y , Porter examined the appropriateness of learner t a l k i n comparison with native-native t a l k as a baseline: To what extent had the learners acquired s o c i o l i n g u i s t i c rules as indicated by the occurrence of inappropriate t a l k 28 i n t h e i r discussions? Porter's findings are of p a r t i c u l a r i n t e r e s t to i n s t r u c t i o n a l planners who have generally assumed (following Krashen, 1978, 1982) that NS speech constitutes the only source of high-quality language input a v a i l a b l e to learners i n second and foreign language classrooms. Perhaps the most important f i n d i n g of Porter's study was that learner-learner conversation, e s p e c i a l l y between learners at advanced and intermediate l e v e l s of pro f i c i e n c y , i s at le a s t as e f f e c t i v e as NS-learner t a l k i n terms of providing opportunities to repair or avoid misunderstanding, and to a s s i s t i n t e r l o c u t o r s i n continuing s u c c e s s f u l l y through a topic of mutual i n t e r e s t . Only a t i n y f r a c t i o n of the errors occurring during learner t a l k was repeated by a non-native conversation partner; many errors were successfully monitored and corrected by the partner. Several of Porter's s p e c i f i c findings are worth reporting here. Regarding the i n t e r a c t i o n a l q u a l i t y of IT, Porter found the rate of monitoring and such repairs as c l a r i f i c a t i o n requests, confirmation checks and comprehension checks to be e s s e n t i a l l y i d e n t i c a l i n both learner-learner and NS-learner conversation, and the rate at which learners prompted each other to be much higher than the rate at which they prompted NSs. On these points Porter (1986, p. 214) concluded: "both types of in t e r l o c u t o r s [learners and NSs] are equally e f f e c t i v e conversation partners. The fin d i n g f o r . . . prompts [however] suggests 29 that learners make better partners . . . . 1 1 Comprehensibility during learner t a l k was s i g n i f i c a n t l y better than during NS-learner t a l k . Comprehensible input thus would seem to be assisted by IT when p a r t i c i p a n t s share the same interlanguage phonology. Moreover, input provided by the advanced learners speaking to other learners was s i g n i f i c a n t l y better i n qu a l i t y than that provided by NSs as measured by the judges' ratings. Learners at a l l l e v e l s , but p a r t i c u l a r l y at the advanced l e v e l , produced s i g n i f i c a n t l y more language f o r other learners than f o r NSs—a f i n d i n g which c l e a r l y suggests the p o t e n t i a l benefit of IT as input i n i n s t r u c t i o n a l settings. One addit i o n a l finding, however, showed IT to be a r e l a t i v e l y i n e f f i c i e n t means f o r language learners to acquire rules of s o c i o l i n g u i s t i c competence. Porter found that IT d i d not provide learners " s o c i o c u l t u r a l l y appropriate models" (p. 194) for the three language functions examined i n the q u a l i t a t i v e analysis: expressing opinions, agreements and disagreements. This suggests a spe c i a l r o l e f o r teachers i n classrooms or for NSs outside the classroom, namely providing adequate s o c i o c u l t u r a l input for language learners who apparently are unable to provide i t to each other. Dimensions of Task and Interlancruacre Talk One-way and Two-way Tasks. The e f f e c t of task on the q u a l i t y of IT ( p a r t i c u l a r l y on the frequency of repairs undertaken during conversation 30 on task) has also been examined by Gass and Varonis (1985a) and Duff (1986), among others. (An extensive review of the l i t e r a t u r e on task, Crookes, 1986, w i l l be examined i n Conceptual Dimensions of the Study, below.) Following Varonis and Gass' (1985) model describing the form and process of negotiation of meaning i n nonnative discourse, Gass and Varonis (1985a) observed how one-way and two-way tasks made d i f f e r e n t communicative demands on intermediate-level NNSs i n conversational dyads and thus influenced the q u a l i t y of negotiated i n t e r a c t i o n . They defined a one-way task as "an i n t e r a c t i o n which involves the g i v i n g of information from only one p a r t i c i p a n t to the other" (p. 149) and a two-way task as "an i n t e r a c t i o n which involves exchanges of information . . . exchanges i n which both p a r t i c i p a n t s have information which must be shared i n order to complete a given task" (p. 149). The dependent va r i a b l e used i n the study was the number of pushdowns, or indica t i o n s of d i f f i c u l t y i n understanding, i n i t i a t e d by a l i s t e n e r . Pushdowns were the basis of nonunderstandinq routines—"exchanges i n which there i s some overt i n d i c a t i o n that understanding between pa r t i c i p a n t s has not been complete" (p. 151)—and were expected to vary with the task employed. The one-way task entailed one member of the dyad describing a picture while the other member attempted to reconstruct i t without d i r e c t reference to the o r i g i n a l p i c t u r e (but with the feedback of the person describing i t ) . The two-way task required the dyad members to piece 31 information together which they possessed i n d i v i d u a l l y towards solu t i o n of a crime; the members had to exchange information cooperatively i f the crime were to be solved. The authors reported no s i g n i f i c a n t differences between t h e i r one-way and two-way tasks. This r e s u l t seems to contradict Long's (1980) findings i n which two-way tasks resulted i n s i g n i f i c a n t l y greater negotiated i n t e r a c t i o n (via comprehension checks, requests f o r c l a r i f i c a t i o n , etc.) than one-way tasks. The authors pointed out, however, that the l e v e l of shared assumptions distinguished the two sorts of tasks, with two-way tasks req u i r i n g l e s s negotiation than one-way tasks (although, i n t h e i r study, apparently not s i g n i f i c a n t l y less) because of the greater amount of information shared by the p a r t i c i p a n t s : As the amount of information independently possessed by p a r t i c i p a n t s increasingly overlaps, they have less need to share i t during performance of the task. (Gaies, 1982 also makes t h i s point, explaining that p a r t i c i p a n t s ' shared knowledge of each other reduces the chance of conversational breakdown and dropping of topics.) The explanation that p a r t i c i p a n t s ' shared assumptions or knowledge reduces the need to negotiate over conversational "trouble" i s a t t r a c t i v e i f not persuasive. However, i t does not deal d i r e c t l y with the problem of what i s being negotiated (information per se versus comprehensibility of the p a r t i c i p a n t ' s t a l k — comprehensibility or lack of i t being a f o c a l issue i n Long, 32 1980 and 1983, for example) or with the problem of d i r e c t i o n a l i t y (one-way tasks can, arguably, require p a r t i c i p a n t s to negotiate s i g n i f i c a n t l y more than two-way tasks due to the r e l a t i v e l y smaller number of assumptions which p a r t i c i p a n t s share). Thus, i t i s not cl e a r whether negotiation over nonunderstanding i n IT i s a function of the degree of shared assumption permitted by a task or the need to share information i n order to complete a task s u c c e s s f u l l y (by d e f i n i t i o n , a c h a r a c t e r i s t i c of two-way tasks) or, perhaps, the need to make the language by which information i s to be shared more comprehensible. Convergent and Divergent Tasks. Duff (1986) provides an additional view of the task-i n t e r a c t i o n r e l a t i o n s h i p i n IT, examining the degree to which dyadic tasks support shared-goal (convergent) or independent-goal (divergent) i n t e r a c t i o n . Like Porter (1983, 1986) and Gass and Varonis (1985a), Duff employed teacherless tasks i n the form of cooperative problem-solving and debate. Unlike other researchers, however, Duff used two-way tasks exclusively i n the study and thus d i d not attempt to r e p l i c a t e research which examined the e f f e c t of one-way and two-way tasks on repair behavior. Subjects i n the study included four native speakers of Mandarin Chinese and four of Japanese. Quantity of input was measured by the number of words and c-units ("a word, phrase, or sentence that i n some way contributed pragmatic or semantic meaning to a conversation", p. 153). Quality was measured by the 33 frequency of turns, questions and S-nodes (a measure of syn t a c t i c complexity; see Freed, 1978). S p e c i f i c measures of i n t e r a c t i o n included comprehension checks, c l a r i f i c a t i o n requests, confirmation checks, c o l l a b o r a t i v e checks ( " e x p l i c i t feedback or agreement or disagreement i s sought", p. 152), i n addition to several question forms. The i n t e r a c t i o n features were s i m i l a r to those reported i n Long (1980, 1981) Pica (1987), Pica and Doughty (1985), Pica and Long (1986) and Porter (1983) except f o r the elaboration of question types and the addition of c o l l a b o r a t i v e checks. Duff found that the debates (the divergent tasks) produced s i g n i f i c a n t l y more words per turn, fewer c-units, more words per c-unit and more S-nodes per c-unit than the problem-solving (convergent) tasks. Debate, i n general, was thus found to produce longer and more complicated discourse than problem-solving. As f o r the i n t e r a c t i v e q u a l i t y of the tasks, Duff found s i g n i f i c a n t l y more subject questions i n the form of confirmation checks and r e f e r e n t i a l questions i n problem-solving than i n debate, although task differences did not reach s i g n i f i c a n c e for comprehension checks and c l a r i f i c a t i o n requests. Ethnic differences were also noted, with the Chinese subjects taking more frequent turns and asking questions more frequently than t h e i r Japanese counterparts. This finding suggests that r e l a t i v e l y voluble i n d i v i d u a l s may create the opportunities f o r t h e i r l e s s voluble partners to p a r t i c i p a t e i n the negotiation of conversational material. F i n a l l y , with shorter turns and 34 more frequent and immediate feedback, Duff concluded that problem-solving was more conducive to SLA than debate, p a r t i c u l a r l y with respect to the greater amount of questioning (and thus c l a r i f i c a t i o n of meaning) engendered by t h i s task type. The major implication of Duff's study i s that tasks can be distinguished by the degree to which they stimulate cooperative, interlanguage exchange on the t o p i c . Divergent tasks would seem to encourage a more expository, abstract and i n s t r u c t i v e approach to t a l k with conversation partners, whereas convergent tasks seem conducive to the frequent, cooperative exchange of conversational material which i s made concrete and personally relevant from moment-to-moment. This d i s t i n c t i o n has also been discussed by Kramsch (1985) who noted the "dual nature of the language learning task" (p. 170) and the v a r i a t i o n of tasks along the i n t e r a c t i o n continuum between "position-centered teaching and learning, . . . i n which information i s delivered and received . . . [and] person-centered communication, i n which information i s exchanged and meanings are negotiated" (p. 171) . Required versus Optional Information Exchange. Kramsch's d i s t i n c t i o n i s echoed i n research conducted by Doughty and Pica (1986) and Pica (1987). Doughty and Pica compared tasks which required the exchange of information and tasks which l e f t information exchange optional, i n teacher-led, small (four-member) group and 35 dyadic settings. They found that the requirement of information exchange was the key va r i a b l e i n producing s i g n i f i c a n t l y more modified i n t e r a c t i o n i n English (as measured by the frequency of c l a r i f i c a t i o n requests, confirmation checks and comprehension checks) i n a l l settings, but that the NNS-NNS p a r t i c i p a t i o n patterns i n small groups and dyads produced much more modified i n t e r a c t i o n than those led by the NS teacher. Noting the large number of ungrammatical utterances (p. 322) i n the various intermediate p r o f i c i e n c y l e v e l student groups, however, Pica and Doughty cautioned that the teacher remains the only source of grammatical input o r d i n a r i l y a v a i l a b l e to the learners. Beyond t h i s caveat, which finds only l i m i t e d support i n the l i t e r a t u r e and must contend with contradictory evidence (Long, 1980, f o r example), the study's general conclusion was that i t i s the task-obligatory exchange of information, e s p e c i a l l y i n but not l i m i t e d to NNS groups, which seems to create the conditions fo r negotiated exchange of information and meaning. Information Exchange versus Decision-Making Tasks. These findings are consistent with those reported i n Pica (1987) who contrasted the number of c l a r i f i c a t i o n requests, confirmation checks and comprehension checks ( i n d i c a t i v e of the degree of modified interaction) i n teacher- and learner-directed groups for both decision-making and information exchange tasks. Pica found that teacher-directed p a r t i c i p a t i o n was generally the l e a s t 36 productive of modified i n t e r a c t i o n . When contrasting the two sorts of task, however, Pica found much larger differences i n both teacher-led and learner-led groups when information had to be exchanged i n order to complete a task su c c e s s f u l l y than when members of a group simply discussed a problem. When par t i c i p a n t s were obliged to share information about the loc a t i o n of flowers on a board, modified i n t e r a c t i o n became the key to successful completion of the task. In addition, Pica noted the apparent influence of r o l e r e l a t i o n s h i p s on the task-based t a l k when the task did not require an equal exchange of information among pa r t i c i p a n t s . The teacher's normally dominant status i n the classroom or a given learner's a b i l i t y to dominate a group's conversational time, for example, were less l i k e l y to r e s t r i c t the occurrence of modified i n t e r a c t i o n when the task encouraged p a r t i c i p a n t s to exchange information on a more-or-less equal basis. In summary, then, d i f f e r e n t tasks have been found to influence the qu a l i t y and quantity of IT. Although Long (1980) found two-way tasks more e f f e c t i v e than one-way tasks i n generating negotiated i n t e r a c t i o n between members of NNS dyads, Gass and Varonis (1985a) found no s i g n i f i c a n t differences between the two forms of task, noting that the degree of shared background and experience which learners bring to a task seems to control the c h a r a c t e r i s t i c s of in t e r a c t i o n . Recasting the d i s t i n c t i o n between task types, Duff (1986) found s i g n i f i c a n t differences between tasks 37 which require learners to reach a common sol u t i o n (convergent tasks) and those which encourage independent goals f o r each member (divergent tasks). Others, including Doughty and Pica (1986) and Pica (1987), have concluded that tasks can be distinguished on the basis of whether or not they require, and not merely i n v i t e , an exchange of information i n order to be completed succes s f u l l y . Thus, there i s l i t t l e consensus on how tasks can be c l e a r l y distinguished to serve p r e d i c t i v e functions i n second and foreign language research, although i t i s c l e a r that the qua l i t y and quantity of IT i s influenced by the nature of the task i n which learners are asked to engage. More generally, the organizing of NNS-NNS i n t e r a c t i o n has been found a serious a l t e r n a t i v e to NS-NNS exchanges i n classrooms (Porter, 1983, 1986) and a source of input and negotiated i n t e r a c t i o n at l e a s t as e f f e c t i v e as that which occurs i n NS-NNS dyads. Instruction which i s teacher-fronted, and maintains the t r a d i t i o n a l teacher-pupil status differences, has been found i n some studies to produce an " i n f e r i o r " and les s focused version of FT, s p e c i f i c a l l y i n the dysfunctional use of display questions (see Chaudron, 1983; Long & Sato, 1983; also Doughty & Pica, 1986; Pica, 1987 f o r comparisons of teacher-fronted and small, NNS-NNS group conversational a c t i v i t y ) . The discussion next moves to several key conceptual underpinnings for the study, examining i n greater d e t a i l the nature and uses of repair, task and reference. 38 Conceptual Dimensions of the Study Repair The f i r s t conceptual f i e l d to be examined here i s repair, a s u r p r i s i n g l y mercurial term given i t s frequency of occurrence i n the l i t e r a t u r e and uses as a measure of i n t e r a c t i o n a l adjustment. In general, r e p a i r has been viewed e i t h e r as a process for negotiating conversational "trouble" (Gass and Varonis, 1985a, 1985b) or a r e l a t e d group of i n t e r a c t i o n strategies which p a r t i c i p a n t s use to improve the comprehensibility of t h e i r t a l k (Long, 1980; Porter, 1983, 1986). Other studies have distinguished between forms of repair which are e s s e n t i a l l y l e x i c o -grammatical i n nature and those which modify the propositional content of the discourse (Kasper, 1985; Porter, 1986; Schachter, 1985; Schwartz, 1980). Schegloff, Jefferson, and Sacks (1977) characterized r e p a i r as "the s e l f - r i g h t i n g mechanism fo r the organization of language use i n s o c i a l i n t e r a c t i o n " (p. 381) which occurs when conversational p a r t i c i p a n t s perceive a source of trouble i n e i t h e r t h e i r own or t h e i r partner's t a l k . Repair of anticipated or a c t u a l l y occurring trouble was distinguished from simple correction or replacement of error and found to be overwhelmingly the r e s p o n s i b i l i t y of the "trouble source". That i s , s e l f - r e p a i r i s o r d i n a r i l y preferred to other-repair, although Schegloff et a l . speculated that other-correction for the not-as-yet competent 39 . . . appears to be one ve h i c l e for s o c i a l i z a t i o n . I f that i s so, then i t appears that other-correction i s not so much an a l t e r n a t i v e to s e l f - c o r r e c t i o n i n conversation i n general, but rather a device f o r dealing with those who are s t i l l learning or being taught to operate with a system which requires, f o r i t s routine operation, that they be adequate self-monitors and s e l f - c o r r e c t o r s as a condition of competence, (p. 381) Indeed, Kasper (1985) not only found a preference f o r both o t h e r - i n i t i a t e d and other-completed repair i n language-centered i n s t r u c t i o n ( i . e . , i n s t r u c t i o n dominated by the teacher), but also found the more conventional pattern of s e l f - i n i t i a t e d and self-completed repair by NNSs during non-i n s t r u c t i o n a l discourse. The repair preference thus seems to depend on the context i n which t a l k occurs. Although not s p e c i f i c a l l y concerned with the context i n which re p a i r occurs, Porter (1983, 1986) reported higher rates of both s e l f - and other-correction (monitoring) by NSs i n conversation with NNSs, a finding i n support of Schegloff et a l . These s p e c i f i c findings are l a r g e l y r e l a t e d to the lexico-grammatical character of conversational r e p a i r (but see Schwartz, 1980 for a treatment of s e l f - and other-repair i n negotiated and i n s t r u c t i o n a l discourse). Of even more in t e r e s t to the present study are the various exponents of repair which have been reported as contributing to i n t e r a c t i o n a l 40 modification. Here there i s a well-studied, frequently r e p l i c a t e d group of exponents, although the term repair i s not yet rou t i n e l y or consistently used to describe them. Long (1980), f o r example, found that NS-NNS dyads r e l i e d on such r e p a i r devices as confirmation checks, c l a r i f i c a t i o n requests, s e l f - and other-repetition to repair breakdowns which had already occurred, whereas comprehension checks and r e p e t i t i o n functioned to avoid breakdown. Porter (1983) grouped c l a r i f i c a t i o n requests, comprehension checks, v e r i f i c a t i o n s of meaning, d e f i n i t i o n requests and l e x i c a l u ncertainties under repair. (Tarone (1983) has also described the l a t t e r three devices as exponents of the communication strategy appeals for assistance). Porter found r e p a i r frequencies for NNS groups s i m i l a r to those i n NS groups. About two-thirds of a l l repairs were found to be concentrated among confirmation checks, c l a r i f i c a t i o n requests and comprehension checks. Further consideration of repair frequency as an ind i c a t o r of i n t e r a c t i o n a l modification i s found i n Gass and Varonis (1985b). Repair occurs within "negotiation exchanges . . . i n which there i s some overt i n d i c a t i o n that understanding between parti c i p a n t s has not been complete and there i s a resultant attempt to c l a r i f y the nonunderstanding" (p. 39). Gass and Varonis (1985a) defined nonunderstanding routines i n a s i m i l a r fashion and, i n discussion of t h e i r model of unaccepted input, noted that negotiation i n nonnative-nonnative discourse i s 41 triggered and then indicated by a hearer's incomplete understanding, followed by the o r i g i n a l speaker's response and the hearer's optional reaction to the response (p. 151f). Schwartz (1980) made the i d e n t i f i c a t i o n of repa i r and negotiation e x p l i c i t on the basis of her q u a l i t a t i v e study, de f i n i n g repair as "a process of negotiation, invo l v i n g speakers conferring with each other to achieve understanding" (p. 151). Thus i t seems that repair i s viewed both as the p a r t i c u l a r utterances speakers make when dealing with p o t e n t i a l or actual trouble spots i n conversations and the process by which pa r t i c i p a n t s attempt to reach a common understanding of problematic conversational material. The working d e f i n i t i o n of repair used i n t h i s study combines elements of both views: Conversational re p a i r i s a group of in t e r a c t i o n strategies speakers employ to make t h e i r own and other's t a l k more comprehensible i n a n t i c i p a t i o n of or response to communication d i f f i c u l t i e s . Although i t i s not always possible to know i n advance of observation what pa r t i c i p a n t s consider to be "trouble", i t i s possible to examine the evidence of breakdown under various conversational circumstances by examining—as others have done—the record of repair behavior during performance of d i f f e r e n t tasks. Task The concept of communication task and i t s use as a unit of analysis i n SLA research and teaching has been examined 42 comprehensively by Crookes (1986). Crookes defined communication task as "a piece of work or an a c t i v i t y , u sually with a s p e c i f i e d objective, undertaken as part of an educational course, at work, or used to e l i c i t data f o r research" (p. 1), and noted that a number of other terms, including a c t i v i t i e s , jobs, procedures, processes, have also been used to denote organizational formats of use to researchers i n operationalizing t h e i r research designs. From the researcher's point of view, then, tasks which appear to have unique c h a r a c t e r i s t i c s can be employed to e l i c i t language data for l a t e r analysis. Tasks can be varied to produce systematic v a r i a t i o n i n the language used to navigate through the task. From the i n s t r u c t i o n a l planner's point of view, however, tasks represent s p e c i a l environments i n which to operationalize i n s t r u c t i o n a l objectives. Thus, implementation of various tasks leads eventually to various, foreseeable changes i n knowledge or attitudes among learners. Unfortunately, as Crookes noted, there i s i n f a c t very l i t t l e understanding of the task-behavior r e l a t i o n s h i p i n second language research and p r a c t i c e . Applied l i n g u i s t s have applied the equivalent of c r i t e r i a for face v a l i d i t y to tasks borrowed from i n s t r u c t i o n a l p r a c t i c e (but see Shortreed, 1986 for an attempt to d i s t i n g u i s h among tasks on conceptual grounds established i n the l i t e r a t u r e of group and s o c i a l psychology). Thus, i t i s very d i f f i c u l t to know, except perhaps with hindsight, which task categories are 43 worth pursuing for data c o l l e c t i o n , or to ensure that a given task used i n one piece of research i s the same as that used i n another (a problem noted by Varonis and Gass, 1985). Without reference to a defensible t h e o r e t i c a l point of view, i t i s also d i f f i c u l t to know whether tasks are best distinguished by how obligatory information sharing-is or by the degree of shared assumptions learners bring to the conversational s e t t i n g . There are c e r t a i n l y other ways to di s t i n g u i s h among tasks used i n classrooms or planned f o r research purposes, including, f o r example, the r e l a t i v e degree of cognitive complexity one task has over another. Even i f complexity can be operationalized (see Shortreed, 1986; Crookes, 1986), researchers are s t i l l faced with j u s t i f y i n g i t s importance i n t h e o r e t i c a l and p r a c t i c a l terms. One way of approaching the problem of s e l e c t i n g tasks fo r research or i n s t r u c t i o n a l purposes i s to begin with a t h e o r e t i c a l framework f o r se l e c t i n g and then characterizing tasks. (Apart from preliminary c l a s s i f i c a t i o n within such a framework, such tasks may well have been i n classroom use for a considerable period). This rather deductive strategy i s not the only approach, of course, although i n contrast to the inductive strategies of much recent research i n the f i e l d , i t can help to va l i d a t e the s e l e c t i o n of task factors within which data w i l l be examined. This way of thinking i s exemplified i n two related views of communication task which have been proposed by Cummins (1983) and Mohan (1986). 44 Cummins was interested i n the ways communication i s affected by changes i n "contextual support f o r a given . . . exchange or b i t of discourse, and . . . the degree of cognitive e f f o r t required f o r comprehension and expression" (Cummins, 1983, p. 108). According to Cummins, language p r o f i c i e n c y can be expected to vary along two orthogonal continuua: Range of Contextual Support and Degree of Cognitive Involvement i n Communicative A c t i v i t i e s . His framework i s reproduced below. COGNITIVELY UNDEMANDING CONTEXT EMBEDDED CONTEXT REDUCED B COGNITIVELY DEMANDING Figure 1. Range of contextual support and degree of cognitive involvement i n communicative a c t i v i t i e s , (p. 120) Cummins noted that context-embedded communication i s t y p i c a l of everyday (non-instructional) t a l k outside of classrooms i n that par t i c i p a n t s negotiate meaning by 45 o f f e r i n g feedback about the i n t e l l i g i b i l i t y of the t a l k as i t unfolds. Context-reduced si t u a t i o n s are more c h a r a c t e r i s t i c of academic or school settings i n which a premium i s placed on abstract reasoning, p r e c i s e l y elaborated messages and c a r e f u l control of learners' verbal p a r t i c i p a t i o n i n order to avoid misinterpretation. Cognitive Involvement "can be conceptualized i n terms of the amount of information that must be processed simultaneously or i n close succession by the i n d i v i d u a l i n order to carry out the a c t i v i t y " (p. 121). At the c o g n i t i v e l v undemanding end of the continuum are mainly automatized communicative tasks which require r e l a t i v e l y l i t t l e active involvement or creative use of language. Cognitivelv demanding tasks, by contrast, require more active communication and negotiation of the discourse. The discourse becomes open to manipulation by the p a r t i c i p a n t s , thus allowing them to achieve such l o c a l conversational purposes as c l a r i f y i n g what a co-conversationalist says or checking to see i f the l i s t e n e r has comprehended an utterance. With NNS-NNS communication i n mind, Cummins' framework suggests conditions under which negotiation and conversational repair are l i k e l y to be e s s e n t i a l conditions of the discourse (quadrant B) and conditions under which they are l e a s t l i k e l y to occur (quadrant C). Quadrant B a c t i v i t y could reasonably occur, for example, i n a company se t t i n g i n which technical s k i l l s are being transferred through the medium of English as a second language: A 46 t r a i n e r and trainee are standing i n front of a piece of chemical analysis equipment; the t r a i n e r i s a r e l a t i v e l y good speaker of English and thus has not memorized (although i s f a m i l i a r with) the relevant section of the i n s t r u c t i o n manual on chemical analysis of non-organic p r e c i p i t a t e s ; the trainee, highly motivated to complete the a c t i v i t y successfully, i s not sure he has understood what the t r a i n e r s a i d about f i l l i n g a graduated cylinder to a c e r t a i n l e v e l , so he nominates a ce r t a i n figure f o r the t r a i n e r to confirm. Verbal i n t e r a c t i o n of a si m i l a r , although simulated, sort could be examined under controlled circumstances, with tasks varied according to the requirements of a t h e o r e t i c a l framework such as the one Cummins has proposed. A terminology for task analysis i n SLA research has not generally been developed on the basis of frameworks or models of the sort discussed here (but see Duff, 1986; Faerch & Kasper, 1983; Wagner, 1983 for conceptual thinking of use to design of empirical study). Mohan (1986), however, o f f e r s a broad t h e o r e t i c a l perspective f o r describing a c t i v i t i e s and, s p e c i f i c a l l y , uses of language i n educational settings which can be applied to s e l e c t i o n of tasks f o r research purposes. I t should be emphasized that Mohan was interested i n educational processes and d i d not attempt to explain processes i n SLA. The typology of language and content learning i s based on a knowledge framework (p. 35f.) which i s divided into general t h e o r e t i c a l knowledge and s p e c i f i c p r a c t i c a l knowledge. 47 Knowledge i s communicated through an a c t i v i t y which, Mohan noted, "combines theory (background knowledge) and pra c t i c e (action situations) . . . . Verbal, expository learning i s es s e n t i a l f o r understanding theory and symbolic knowledge, but i t needs to be associated with l i f e experience and p r a c t i c a l knowledge" (p. 45). Thus the d i s t i n c t i o n between expository and exp e r i e n t i a l approaches to teaching and learning i s , at i t s broadest, the difference between content expressed through t h e o r e t i c a l discourse over knowledge which ex i s t s independently of the s i t u a t i o n i n which i t i s discussed (as i n lectures, textbooks, classroom discussions, fo r example) and content expressed through p r a c t i c a l discourse over objects which can be referred to i n the communicative s i t u a t i o n (laboratory work, demonstrations, cooperative games, for example). The expository-experiential d i s t i n c t i o n i s c e r t a i n l y more complex than suggested by the broad outline presented so f a r . For the purposes of t h i s discussion, however, a simple, l i m i t e d and incomplete characterization of task i n r e l a t i o n to the d i s t i n c t i o n w i l l be adopted. A communication task i s expository when pa r t i c i p a n t s can communicate about the topic of conversation by means of discourse only. A telephone conversation would be an obvious example. (This i s c l e a r l y not expository i n the sense of categories of rh e t o r i c sometimes employed to describe prose, e.g., "expository" versus "narrative" prose.) A communication task i s ex p e r i e n t i a l when 48 p a r t i c i p a n t s can communicate about the t o p i c of conversation through various media (visual presentation, gesture and action, as well as verbalization) and when they can d i r e c t l y experience f o r themselves what i s talked about. An example would be t a l k while j o i n t l y constructing a Lego toy. Although i t i s convenient to speak i n terms of a d i s t i n c t i o n between two task types, i t i s probably more r e a l i s t i c to view tasks along a dimension permitting p a r t i c i p a n t s various degrees of d i r e c t , shared experience and shared perception i n the task s i t u a t i o n . A l l things being equal, e x p e r i e n t i a l tasks are l i k e l y to lead to more repa i r than expository tasks on the grounds that p a r t i c i p a n t s have more sources of information which may indicate conversational trouble. I t should be also pointed out, however, that shared experience and perception of material resources may, at the extreme, begin to obviate p a r t i c i p a n t s ' need to negotiate trouble, a problem posed by Gass and Varonis (1985a) and raised again below during discussion of how the Knowledge Framework may be applied to problems of observational research. Proposing new task terminology and r e l a t i n g i t to conceptual discussion i n the l i t e r a t u r e only p a r t i a l l y j u s t i f i e s i t s i n c l u s i o n i n the research. I t does not follow, however, that an experiential-expository characterization of tasks i s a v a l i d one, or even that the framework which supports i t i s a useful means of studying the a l l o c a t i o n of IT during tasks. How can these additional 49 c r i t e r i a f o r d i s t i n g u i s h i n g between ex p e r i e n t i a l and expository tasks be invested i n the research? What additi o n a l body of research can be employed to t e s t the r e a l i t y of e x p e r i e n t i a l and expository behavior during performance of given tasks? Reference One way of approaching these questions i s to examine how elements of spoken texts gain cohesion during discourse and force what Brown and Yule (1983a, p. 190) r e f e r to as co-interpretation. Halliday and Hasan (1976) have discussed cohesion i n texts as "a semantic r e l a t i o n between an element i n the text and some other element that i s c r u c i a l to the in t e r p r e t a t i o n of i t " (p. 8). They note, "Where the in t e r p r e t a t i o n of any item i n the discourse requires making reference to some other item i n the discourse, there i s cohesion" (p. 11). Reference i s a form of cohesion which l i n k s the i d e n t i t y of a thing ( i t s r e f e r e n t i a l meaning) with other elements i n a text variously crafted to represent i t . Halliday and Hasan's taxonomy distinguishes between exophoric (situational) and endophoric (textual) forms of reference. Exophoric reference i s an e s p e c i a l l y i n t e r e s t i n g part of the taxonomy because i t s use i s e n t i r e l y r e s t r i c t e d to the s i t u a t i o n i n which i t occurs; i t s use i s thus e n t i r e l y external to e f f o r t speakers expend on creating a cohesive text. During exophoric reference speakers t y p i c a l l y r e f e r to objects which can be viewed or otherwise located through the use of language. In the absence of a v i s u a l 50 record or a supplemental text, therefore, non-participants are forced to imagine what the i n i t i a l referent might have been. The authors noted that "language-in-action" s i t u a t i o n s e n t a i l a high proportion of exophoric reference, s i t u a t i o n s i n which at l e a s t one of the p a r t i c i p a n t s makes reference to things i n the immediate environment and assumes that the co-participant i s able to follow the verbal (and often physical) "pointing out". When an addressee i s unable to do so, as may be the case when adults are dealing with very young childre n who assume that everyone wo whom they speak shares t h e i r own focus of attention, r e f e r e n t i a l presuppositions must be resolved, negotiated i n e f f e c t , before the adult w i l l allow the conversation to move on. The following exchange (excerpted from Halliday & Hasan, p. 34) i l l u s t r a t e s t h i s point: C h i l d : Why does THAT one come out? Parent: That what? Ch i l d : THAT one. Parent: That what? Chil d : That ONE! Parent: That one what? Chil d : That l e v e l there that you push to l e t the water out. This rather narrow focus of negotiation would not t y p i c a l l y happen i n adult conversation, e s p e c i a l l y i n cases of peer group members who share considerable knowledge and maintain c e r t a i n expectations about the things l i k e l y to be 51 pointed out during conversation. In fac t i t i s p r e c i s e l y the "reservoir of shared experience" (p. 36) which makes exophoric reference a natural, expected feature of the discourse but an enigma to those who do not share the same l e v e l of experience and the same material context. One function of endophoric reference i n discourse, then, i s to supply cohesion to the spoken text which exophoric reference i s unable to do. Anaphora, the "pointing back" form of endophoric reference, for example, lends cohesion to texts by r e f e r r i n g to things (objects, ideas, states) which are removed i n space (in the case of written texts) and time from the i n i t i a l presupposition. This c h a r a c t e r i s t i c of anaphora makes t a l k more portable, i n a sense, allowing conversational p a r t i c i p a n t s to share meaning (assuming they share a s i m i l a r r e f e r e n t i a l competence) without dependence on the s i t u a t i o n . Halliday and Hasan point out that speakers-to-be (next speakers i n a conversation) o r d i n a r i l y have the competence to judge whether reference i s exophoric or anaphoric, i . e . , whether or not i t serves a cohesive function i n a text, and to i d e n t i f y which part of the text i s the referent. Although t h i s sort of competence may be generally a v a i l a b l e to speakers-hearers i n any language group, i t i s demonstrably a learned competence which permits people i n p a r t i c u l a r language groups to recognize what kind of reference i s at work i n a conversation—whether reference functions to point out or to point back—and to respond to i t appropriately. 52 What happens t o t h i s l a n g u a g e - s p e c i f i c s t r u c t u r e o f r e f e r e n c e when NNSs engage i n c o n v e r s a t i o n and how does i t r e l a t e t o t h e t a s k s t h e y a r e asked t o p e r f o r m i n i n s t r u c t i o n a l s e t t i n g s ? G i v e n Mohan's d i s t i n c t i o n between p r a c t i c a l and t h e o r e t i c a l knowledge and H a l l i d a y and H a s a n ' s d e s c r i p t i o n o f how s i t u a t i o n a l and t e x t u a l r e f e r e n c e f u n c t i o n , i t becomes p o s s i b l e t o p r o p o s e a s e t o f terms wh ich move from c o n c e p t u a l framework t o b e h a v i o r a l exponents wh ich a r e o f use i n t h e r e s e a r c h d e s i g n ( F i g u r e 2 ) . The l i s t o f terms i s n e i t h e r e x h a u s t i v e nor i n d i c a t i v e o f s u b t l e d i f f e r e n c e s among s p e c i f i c t y p e s o f t a s k s which c o u l d be s e l e c t e d f o r r e s e a r c h ( o r , f o r t h a t m a t t e r , i n s t r u c t i o n a l ) p u r p o s e s . I t d o e s , however, pu t r e c e n t t h i n k i n g on t a s k i n t o p e r s p e c t i v e . I t a l s o p r o v i d e s a f o u n d a t i o n f o r making s u g g e s t i o n s about t h e r e l a t i o n s h i p o f such v e r b a l b e h a v i o r as r e f e r e n c e and r e p a i r i n d i s c o u r s e c o n t e x t s t o approaches t o t e a c h i n g and l e a r n i n g . F i g u r e 2 i s o f f e r e d s i m p l y as a g u i d e f o r e x p l o r i n g p o s s i b l e c o n n e c t i o n s among i t s p a r t s . I t a ims t o s u g g e s t t e n d e n c i e s bu t i t i s not i n t e n d e d t o c l a i m a b s o l u t e , c l e a r c u t r e l a t i o n s h i p s . The f i g u r e thus s u g g e s t s t h a t t h e r e i s a tendency f o r e x p e r i e n t i a l approaches t o be a s s o c i a t e d w i t h e x o p h o r i c r e f e r e n c e more than w i t h a n a p h o r i c r e f e r e n c e . There i s no s u g g e s t i o n , however , t h a t e x p e r i e n t i a l approaches a r e l i m i t e d o r r e s t r i c t e d t o e x o p h o r i c r e f e r e n c e , o r t h a t a n a p h o r i c r e f e r e n c e cannot appear i n e x p e r i e n t i a l t e x t s . The same may be s a i d f o r the o t h e r l i n k s i n F i g u r e 53 2; they are a l l tendencies, some of which may be treated empirically. Many of the terms i n Figure 2 w i l l be re-introduced i n Chapter 3 with the research design. For the present, i t should be noted that the figure distinguishes h o r i z o n t a l l y between t h e o r e t i c a l and p r a c t i c a l knowledge, and v e r t i c a l l y between concept and s i t u a t i o n . Task i s thus roughly at the in t e r s e c t i o n between what the researcher (or teacher) intends and what a c t u a l l y occurs i n the discourse s e t t i n g . Movement down the figure brings increasing s p e c i f i c i t y , so that at the point where the reference types and repa i r FRAMEWORK COMPONENTS (Mohan, 1986) Theoretical Expository • ~ I Background Knowledge P r a c t i c a l I I E x p e r i e n t i a l <- KNOWLEDGE BASE <- APPROACH TO TEACHING/ LEARNING INTERFACE: FRAMEWORK/ BEHAVIOR Action S i t u a t i o n I <- ACTIVITY/ TASK PREDICTED VERBAL BEHAVIOR I Anaphoric I I Display Question Exophoric I I Referential Question <- DISCOURSE REFERENCE <- REPAIR EXPONENT Figure 2. Extending the Knowledge Framework to problems i n observational research. exponents are l i s t e d , i t i s possible to think i n terms of how parts of the framework might be extended into o p e r a t i o n a l i z i n g task-based research. I t i s possible to propose, f o r instance, that exophoric reference would be 54 more frequent than anaphoric reference when an ex p e r i e n t i a l approach to organizing a task i s employed, or that more display questions w i l l be produced under expository (rather than experiential) conditions. At the same time, Figure 2 does not attempt to r e l a t e s p e c i f i c task a t t r i b u t e s discussed i n the l i t e r a t u r e to learners' verbal behavior. Thus, although two-way tasks, as Long (1980) and others have found, are f o c a l points f o r negotiation of language, Figure 2 proposes, instead, that e x p e r i e n t i a l and expository a c t i v i t y be viewed as more fundamental bases f o r learners' verbal behavior. The figure i s not e s p e c i a l l y s e n s i t i v e , moreover, to the p o s s i b i l i t y that some tasks w i l l be mixed a f f a i r s and that e x p e r i e n t i a l and expository a t t r i b u t e s may be blended i n the same task. This p o s s i b i l i t y suggests the dimensionality of approaches to teaching and learning, and the importance of eventually r e f l e c t i n g dimensionality i n a research design which claims some reasonable l i n k to the world of educational p r a c t i c e . F i n a l l y , the figure does not indicate that some task a t t r i b u t e s may have a complex, c u r v i l i n e a r r e l a t i o n s h i p with points along the experiential-expository dimension. A very high l e v e l of shared s i t u a t i o n a l knowledge, f o r instance, would reduce the negotiation over meaning p a r t i c i p a n t s would otherwise have to accomplish during use of an ex p e r i e n t i a l approach. Largely shared perception of the s i t u a t i o n would occur when par t i c i p a n t s have a common physical, v i s u a l access to the objects they are t a l k i n g about. Highly 55 e x p e r i e n t i a l ( l i t e r a l l y "hands-on") a c t i v i t y , then, would predictably e n t a i l shared v i s u a l perception and permit expository reference to be the norm (Halliday & Hasan, 1976 made p r e c i s e l y t h i s point; see also Gaies, 1982). A high l e v e l of negotiation would thus occur somewhere between completely shared and completely atomized knowledge of the s i t u a t i o n : When the i d e n t i f i c a t i o n of topics becomes a problem f o r pa r t i c i p a n t s to work out, when gaps i n s i t u a t i o n a l or background knowledge must be compensated for, t a l k w i l l very l i k e l y have to be repaired. The general argument developed to t h i s point, then, i s that c e r t a i n kinds of knowledge (theoretical and p r a c t i c a l ) are l i k e l y to be communicated by c e r t a i n approaches to teaching and learning (experiential and expository) which are given form i n p a r t i c u l a r tasks. Tasks are the settings i n which behavior i s enacted and i n which various forms of reference (anaphora and exophora, for example) and rep a i r occur. This formulation of the argument, or at l e a s t parts of the argument, can be tested by empirical means. Thus, f o r example, an important focus of the foregoing discussion has been on reference, on the des c r i p t i v e system of reference which has served as a basis for nearly a l l l a t e r consideration of the subject (Brown & Yule, 1983a; Martin, 1983) and on the possible r e l a t i o n s h i p among conversational reference, repair and task. In the case of reference, i t i s the more established framework (reference, i n p a r t i c u l a r 56 e x o p h o r i c and a n a p h o r i c forms o f r e f e r e n c e ) which can be u s e d t o v a l i d a t e the more n o v e l way o f t h i n k i n g about t a s k s — t h a t t a s k s r e f l e c t the v a r y i n g degrees t o which t e a c h e r s a p p l y e x p e r i e n t i a l and e x p o s i t o r y approaches t o t h e i r communicat ion w i t h l e a r n e r s . E l a b o r a t i o n o f how t h i s was done and t h e r e s u l t s o b t a i n e d from t h e v a l i d a t i o n p r o c e d u r e w i l l be p r e s e n t e d i n the f o l l o w i n g two c h a p t e r s . Summary A r e v i e w o f l i t e r a t u r e i n s e v e r a l r e l a t e d f i e l d s has h i g h l i g h t e d t h e impor tance o f i n t e r a c t i o n a l m o d i f i c a t i o n f o r second and f o r e i g n language l e a r n i n g . E a r l y FT l i t e r a t u r e s u g g e s t e d a c e n t r a l r o l e f o r s i m p l i f i c a t i o n by NSs i n NS-NNS communicat ion i n o r d e r t o a c h i e v e a b a s i c l e v e l o f c o m m u n i c a t i o n . FT which o c c u r s i n c o n v e r s a t i o n a l s e t t i n g s was a l s o shown t o s e r v e key s o c i a l f u n c t i o n s , i n p a r t i c u l a r t h e exchange o f i n f o r m a t i o n between c o n v e r s a t i o n a l p e e r s . FT i n i n s t r u c t i o n a l s e t t i n g s , o t h e r w i s e known as t e a c h e r t a l k , has been c h a r a c t e r i z e d as a d i s t o r t e d and r e l a t i v e l y i n e f f i c i e n t medium f o r a s s i s t i n g l e a r n e r s t o a c q u i r e a second l a n g u a g e , a l t h o u g h s e v e r a l s t u d i e s have s u g g e s t e d a p p r o p r i a t e ad jus tment o f t e a c h e r t a l k t o f o r e i g n l e a r n e r s d u r i n g l e c t u r e and n a r r a t i o n o f i n s t r u c t i o n a l m a t e r i a l . FT p r o d u c e d i n n o n - i n s t r u c t i o n a l s e t t i n g s , however , seems more s e n s i t i v e t o t h e s i t u a t i o n and the l e a r n e r , w i t h NS c o n t r i b u t i o n s ( input ) a d j u s t e d a c c o r d i n g t o t h e l e a r n e r ' s p r o f i c i e n c y , p r o n u n c i a t i o n and demonst ra ted c o m p r e h e n s i o n . Much r e c e n t work compar ing FT and IT has f o c u s e d on the 57 i n t e r a c t i o n a l q u a l i t i e s of the language produced i n various kinds of discourse settings and the p o t e n t i a l contribution such i n t e r a c t i o n makes to SLA. Several studies have shown IT to be at le a s t as useful as FT i n generating negotiation over troublesome or misunderstood conversational material. When compared with t r a d i t i o n a l , teacher-fronted classroom i n s t r u c t i o n , information exchange i n small-groups of NNSs has proved to be a superior means of developing negotiated exchanges which require the pa r t i c i p a n t s , i n general, to c l a r i f y information and check comprehension. Among the key variables i n studies of NS-NNS and NNS-NNS i n t e r a c t i o n are repair, task and reference. The nature of conversational repair was examined and found to be a frequent focus of attention i n studies examining how members of NS-NS, NS-NNS and NNS-NNS groups r e f i n e and c l a r i f y conversational trouble. Repair i s c l e a r l y an e s s e n t i a l feature of small group communication i n that i t allows members to widen the scope of material which can be discussed. Although the repair process requires sometimes complex negotiation over incomplete understanding, r e p a i r i s also s i g n a l l e d by a l i m i t e d number of recurring and commonly used exponents which have been reported i n the l i t e r a t u r e . Task has been viewed as both an i n s t r u c t i o n a l resource and as a means of studying the c h a r a c t e r i s t i c s of learner language (interlanguage). Tasks have been varied to study e f f e c t s on learner language and on the language used to negotiate t a l k between learners. Theoretical viewpoints on 58 the nature of educational discourse have generally not informed the s e l e c t i o n of tasks for research purposes, nor has task-based language been the subject of comparative, small-group study which distinguishes between native and non-native teacher-led groups on performance of tasks. Based on conceptual reasoning, i t has been argued here that tasks seem most fundamentally to vary on a dimension of e x p e r i e n t i a l and expository a c t i v i t y and that the d i s t i n c t i o n can be tested empirically through a p p l i c a t i o n of a now widely held understanding of reference within the discourse s i t u a t i o n . The issue of how task influences various forms of repair and reference i n small, teacher-led groups i s the subject of the e n t i r e discussion which follows. 59 CHAPTER 3: RESEARCH DESIGN AND METHODOLOGY This chapter contains a description of the research design and the methodology used to implement i t . Major topics taken up i n the section on research design include a b r i e f r a t i o n a l e , a tabular summary and rela t e d discussion of the design, a description of the major variables used during operation of the design, a l i s t of hypotheses and a discussion of the general strategy used to t e s t the hypotheses. The section on methodology focuses on se l e c t i o n and treatment of subjects, c o l l e c t i o n and coding of data and various approaches to analysis of the data. The Research Design Assumptions and Rationale Two assumptions regarding the nature of rep a i r between NNS conversation partners have guided construction of the ov e r a l l design. F i r s t , a d i s t i n c t i o n i s drawn between repair undertaken by teacher-learner groups (which may include e i t h e r a NS or a NNS teacher) and by learner-learner groups. Several studies have made i t c l e a r that the proper baseline for making sense of interlanguage t a l k i s t a l k between native and non-native speakers. One of the strengths of Porter's (1983) study, f o r example, was that each NNS-NNS dyad had NS-NNS and NS-NS counterparts to allow for multiple l e v e l s 60 of comparison on the dependent va r i a b l e s . Long (1981) e x p l i c i t l y noted the importance of comparing mixed (NS-NNS) dyads with NS-NS dyads i n order to make useful comparisons between a r e l a t i v e l y unstudied phenomenon (NS-NNS interaction) and a r e l a t i v e l y well-studied phenomenon (NS-NS i n t e r a c t i o n ) . A question arises, however, when the object of research i n t e r e s t i s the language of repair i n teacher-led groups: What sort of comparison ought to be represented i n the research design? Because there i s no "well-studied" group to serve as a natural baseline i n t h i s study, c r i t e r i a f o r s e l e c t i o n of l e v e l s i n a group factor should, as an al t e r n a t i v e , r e f l e c t something of the natural world i n which English-as-a-foreign language (EFL) teachers operate. Given the t y p i c a l overseas EFL s i t u a t i o n , t h i s would mean that research groups would, at a minimum, include learners of English who speak a common l o c a l language, and a combination of native and non-native ( i . e . , local) teachers of spoken English who are p r o f i c i e n t i n the language they teach. An important implication of comparing two kinds of teacher-led groups i n an EFL sett i n g , then, i s that baseline comparative data for repair of NNS-NNS (teacher-learner) t a l k i s rep a i r of NS-NNS (teacher-learner) t a l k and that comparisons between NS-NNS and NNS-NNS teacher-led groups must be set up at the stage of research design. The second assumption i s that the tasks selected for the research design are q u a l i t a t i v e l y d i f f e r e n t from each other and can be expected to produce q u a l i t a t i v e l y d i f f e r e n t 61 r e p a i r p r o f i l e s . This assumption i s based i n a very general sense on the well-argued s o c i o l i n g u i s t i c perception that the forms and organization of conversation are dependent on c h a r a c t e r i s t i c s of the speech s i t u a t i o n (Cazden, 1972; Hymes, 1972; Turner, 1976, for example). But the more s p e c i f i c point made here i s that the frequency with which p a r t i c i p a n t s r e p a i r t h e i r own and others' t a l k i s se n s i t i v e to the kind of conversational environment they are operating i n . Although tasks can be categorized i n many ways, one system developed for categorizing tasks i n educational settings, Mohan's (1986) Knowledge Framework was found to be of pot e n t i a l value i n di s t i n g u i s h i n g the kinds of language generated by exp e r i e n t i a l and expository approaches to teaching and learning. E x p e r i e n t i a l and expository approaches to tasks i n the research s i t u a t i o n , namely tasks intended to e l i c i t research data, w i l l extend uses of the framework beyond i n s t r u c t i o n a l planning and into the area of research design. Based on the framework, i t i s assumed that e x p e r i e n t i a l approaches to tasks i n the research s i t u a t i o n w i l l o r d i n a r i l y require more repair than expository ones, although c e r t a i n kinds of ex p e r i e n t i a l approaches, those i n which s i t u a t i o n a l information must be negotiated, w i l l e n t a i l more repair than others. The implication of t h i s p o s i t i o n for the research design i s that a conceptual basis e x i s t s f o r d i s t i n g u i s h i n g between tasks and can a s s i s t predictions about the r e l a t i v e frequency of repair. Furthermore, the v a l i d i t y of 62 d i s t i n g u i s h i n g between tasks on the basis of t h e i r e x p e r i e n t i a l and expository c h a r a c t e r i s t i c s can be tested by examining the frequency of reference within the various kinds of tasks selected for the study. In general, exophora should be found more frequently i n tasks emphasizing experience; anaphora should be found more frequently i n tasks emphasizing exposition. An Overview of the Design Group and Task Categories Figure 3 outlines the combined between-and-within subjects, repeated-measures f a c t o r i a l design used i n the study. Although the design i s pri m a r i l y intended to support a serie s of 2 x 5 repeated-measures analyses of variance (ANOVA), i t i s also the basis of a q u a l i t a t i v e analysis of t r a n s c r i p t s coded for repair and reference. Group. the between-subjects factor, has two l e v e l s , Mixed (NS-NNS) and Homogeneous (NNS-NNS), with s i x equivalent values within each l e v e l ( i . e . , s i x mixed and s i x homogeneous dyads). The dyads (N=12) are the basic between-subjects source of comparison; each contains one teacher and one learner. Communication Task, the within-subjects factor and repeated measure, has f i v e basic l e v e l s (computer i n s t r u c t i o n , C0M1; computer demonstration, COM2; t o p i c a l discussion, PIS; Lego constructed back-to-back, LEG1; Lego conducted face-to-face, LEG2). 63 Factor 2: Communication Task Teaching tasks (educational Non-teaching tasks goals) ( s o c i a l goals) Subject S o c i a l Problem-i n s t r u c t i o n exchange solving Factor 1: Group COMl COM2 DIS LEG1 LEG2 Mixed 1 NS-NNS 2 NS-NNS 3 NS-NNS 4 NS-NNS 5 NS-NNS 6 NS-NNS Homogeneous 1 NNS-NNS 2 NNS-NNS 3 NNS-NNS 4 NNS-NNS 5 NNS-NNS 6 NNS-NNS Figure 3. F a c t o r i a l design with major conceptual d i s t i n c t i o n s used i n the study. The task factor contains conceptual d i s t i n c t i o n s of use during the ANOVA and the q u a l i t a t i v e analysis which follows the ANOVA. The f i r s t d i s t i n c t i o n i s between teaching and non-teaching tasks. Teaching tasks emphasize achievement of objectives intended to increase the learner's knowledge or competence through e x p l i c i t i n s t r u c t i o n of subject matter 64 which an educational authority considers worth learning ( i . e . , broadly, educational goals). Both of the teaching tasks selected for the study, C0M1 and COM2 are oriented around subject-matter rather than target language i n s t r u c t i o n . The Non-teaching tasks employed here include free discussion, DIS, which emphasizes s o c i a l exchange and two, LEG1 and LEG2, which center on problem-solving. Although both sets of non-teaching tasks depend on p a r t i c i p a n t s ' cooperative, consensual behavior to achieve t h e i r goals (they are intended to achieve interpersonal or s o c i a l goals), DIS emphasizes expressive discussion allowing free development of propositional content and presumes that p a r t i c i p a n t s have more-or-less equal r i g h t s to volunteer contributions ("autonomous contributions", see E l l i s , 1984, p. 13 0); the point of the discussion i s t y p i c a l l y development or exploration of s o c i a l r e l a t i o n s h i p s . The Lego tasks, on the other hand, center on exchange of information which i s normally intended to a s s i s t i n the s o l u t i o n of a problem; i t i s the problem which motivates cooperative use of the target language. These d i s t i n c t i o n s among tasks r e f l e c t the categories E l l i s (1984, 1985) has developed to describe the prospective goals of i n t e r a c t i o n i n second language classrooms, s p e c i f i c a l l y message-oriented, s o c i a l , and a c t i v i t y - o r i e n t e d goals (cf. tasks based, respectively, on subject-matter i n s t r u c t i o n , s o c i a l expression, and problem-solving). Malamah-Thomas (1987) draws a related, although b r i e f e r , 65 d i s t i n c t i o n between c l a s s r o o m language used t o a c h i e v e e i t h e r e d u c a t i o n a l o r s o c i a l p u r p o s e s . F o u r o f t h e t a s k s a r e a l s o c l a s s i f i e d a c c o r d i n g t o t h e e x t e n t t o which they emphasize the r o l e o f e x p e r i e n c e o v e r e x p o s i t i o n d u r i n g per formance o f the t a s k . " E x p e r i e n c e " has been s i m p l i f i e d i n p r a c t i c e t o r e f e r t o whether o r no t dyad members c a n p o i n t out o r m a n i p u l a t e and see t h i n g s i n t h e t a s k env i ronment ( i . e . , + o r - " d o i n g " and + o r -" s e e i n g " ) — a l e v e l o f s i m p l i f i c a t i o n a t e a c h e r might employ, f o r example , when p l a n n i n g c l a s s r o o m a c t i v i t i e s . T h u s , t a s k s range from the most i n t e n s e l e v e l o f e x p e r i e n t i a l a c t i v i t y t o the most i n t e n s e l e v e l o f e x p o s i t o r y a c t i v i t y (LEG2 and COM2 [+ d o i n g , + s e e i n g ] -> LEG1 [+ d o i n g , -s e e i n g ] -> COMl [- d o i n g , - s e e i n g ] ) , a s e r i e s o f d i s t i n c t i o n s which were a p p l i e d t o p a r t o f t h e h y p o t h e s i s -t e s t i n g p r o c e d u r e (Research Q u e s t i o n s and H y p o t h e s e s , b e l o w ) . PIS l i e s o u t s i d e t h i s c l a s s i f i c a t i o n , s i n c e i t has t h e p o t e n t i a l t o t a k e on o r drop e x p e r i e n t i a l c h a r a c t e r i s t i c s depend ing on the i n t e n t i o n s o f the p a r t i c i p a n t s and the development o f the d i s c u s s i o n . Because PIS i s , a t t h e o u t s e t , an u n d i r e c t e d , n o n - t e a c h i n g t a s k wh ich c o n t a i n s t h e p o t e n t i a l t o become whatever i t s p a r t i c i p a n t s want t o make o f i t , i t can s e r v e as a u s e f u l r e f e r e n c e a g a i n s t which the o t h e r t a s k s may be compared. G r o u p i n g o f t a s k s i n t o t h e s e c a t e g o r i e s a l l o w s f o r p u r s u i n g d i r e c t i o n a l hypotheses from a c o n c e p t u a l b a s e . The c l a s s i f i c a t i o n s o u t l i n e d here (and r e f l e c t e d i n F i g u r e 66 3) overlap at a number of points, most notably i n the combination of teaching tasks with expository approaches to i n s t r u c t i o n and of non-teaching (cooperative problem-solving) tasks with expe r i e n t i a l approaches toward solution of the problems. Although these do not exhaust the p o s s i b i l i t i e s , they have been mentioned to suggest l i n k s between the research design and the kinds of discourse contexts which might be encountered i n both classroom and non-classroom settings. A more det a i l e d d e s c r i p t i o n of the tasks i s found i n Methodology, below. Treatment Each dyad experienced a sequence of a l l tasks at one s i t t i n g , hence the repeated-measures designation of the task fa c t o r (see Ferguson, 1981: "repeated measurement of the same subjects under a number of d i f f e r e n t conditions or treatments", p. 317). The sequence of tasks was randomized, however, following a standard L a t i n Square assignment of tasks to the dyad. Descriptive Measures and Dependent Variables Used i n the Study Descriptive Measures In order to achieve a general d e s c r i p t i v e picture of the data, nine non-inferential measures of conversational a c t i v i t y grouped into three categories were applied to the t r a n s c r i p t data p r i o r to the analysis of variance: 1) word-based measures ( t o t a l words, the number of words uttered per minute, the number of unique words uttered, and type-token 67 r a t i o — u n i q u e words:total words, i . e . , an i n d i c a t i o n of increasing l e x i c a l s o p h i s t i c a t i o n as the c o e f f i c i e n t derived from the r a t i o increases), 2) utterance-based measures ( t o t a l utterances, and words per utterance), and 3) turn-based measures ( t o t a l turns, words per turn and utterances per turn). Although hypothesis t e s t i n g could be based on these d e s c r i p t i v e features of the discourse, very l i t t l e previous work has found them s i g n i f i c a n t predictors of repa i r behavior. On the other hand, they comprise a useful group of terms f o r characterizing the qu a l i t y of t a l k i n NS-NNS and NNS-NNS conversations (see, f o r example, Porter, 1983; Long, 1985 a for i l l u s t r a t i o n s of word-based measures, Arthur et a l . , 1980; Porter, 1983; Long, 1980 f o r app l i c a t i o n of turn-based measures, including type-token r a t i o ) . In the present study, the de s c r i p t i v e categories make e x p l i c i t reference to the f a c t o r i a l structure of the research design but, at the same time, do not require raters to i n f e r the occurrence of a p a r t i c u l a r kind of repa i r behavior ( i . e . , they are "low-inference" measures). Dependent Variables In addition to the desc r i p t i v e measures l i s t e d above, the design used two categories of dependent var i a b l e s , which are b r i e f l y defined and described here (see Appendix E, Instructions to Raters and Index of Dependent Variables, for addi t i o n a l d e s c r i p t i o n and exemplification). The f i r s t category of dependent variable i s r e a l l y a group of 12 68 r e l a t e d discourse strategies p a r t i c i p a n t s employ to maintain the comprehensibility of the ongoing t a l k . These Repair Exponents (REs) were selected l a r g e l y on the basis of t h e i r appearance i n previously reported research and t h e i r u t i l i t y i n focusing on the q u a l i t i e s of teacher-learner i n t e r a c t i o n . The l i s t of REs i s neither exhaustive nor i s i t intended to break new ground i n the description of r e p a i r behaviors. On the contrary, the l i s t i s intended to apply categories which have been already i d e n t i f i e d i n the l i t e r a t u r e , or which are conservative extensions of e x i s t i n g categories, to examination of novel research questions. From an i n t e r a c t i o n a l perspective, moreover, the REs, considered i n d i v i d u a l l y , are p a r t i c u l a r points i n the discourse and thus cannot suggest the complex, negotiated character of the t a l k . In order to do t h i s , a q u a l i t a t i v e examination of re p a i r i n context w i l l be presented i n Chapter 6. For the present, however, emphasis i s on the comparative frequency of r e p a i r and reference within the c e l l s of the design. The following REs served as dependent variables i n the study; the associated description also served as working guidelines for coding of t r a n s c r i p t s . 1. C l a r i f i c a t i o n Request (CCLAR). (See Brulhart, 1985; Doughty & Pica, 1986; Duff, 1986; Long, 1980, 1981; Long and Sato, 1983; Pica, 1987; Pica & Doughty, 1985; Pica et a l . , 1987; Porter 1983, 1986.) A request f o r c l a r i f i c a t i o n i s focused on the preceding speaker's utterance and requests new or reformulated information from the previous speaker. 69 Although a question o r d i n a r i l y conveys the c l a r i f i c a t i o n request (nominally, Would you say that i n other words?) i t i s possible for the request to come i n the form of an interpretable statement such as I don't quite understand. 2. Comprehension Check (CCOM). (See Brulhart, 1985; Doughty & Pica, 1986; Duff, 1986; Long, 1980, 1981; Long & Sato, 1983; Pica, 1983, 1986, 1987; Pica & Doughty, 1985; Pica et a l . , 1987.) Speakers are normally interested i n knowing i f l i s t e n e r s have understood them. A comprehension check s a t i s f i e s t h i s i n t e r e s t by allowing a speaker to query the l i s t e n e r ' s understanding of a current utterance. The nominal form of a confirmation check i s Have you understood . . . ?. although, i n pr a c t i c e , such i n d i r e c t forms of confirmation check as OK? may serve j u s t as well . R i s i n g intonation t y p i c a l l y signals a comprehension check and thus makes i t possible to d i s t i n g u i s h such t o p i c a l boundary markers as r i g h t or OK (used with f a l l i n g intonation) from an attempt to check comprehension. 3. Confirmation Check (CCON). (See Brulhart, 1985; Doughty & Pica, 1986; Duff, 1986; Long, 1980, 1981; Long & Sato, 1983; Pica, 1983, 1986; Pica, 1987; Pica & Doughty, 1985; Pica et a l . , 1987.) A confirmation check i s made by a l i s t e n e r to check understanding or hearing of the speaker and can be reduced to the nominal form Have I understood? As i n the case of a comprehension check, a confirmation check i s made with r i s i n g intonation, but also e n t a i l s p a r t i a l or complete r e p e t i t i o n of a preceding utterance 70 ( s p e c i f i c a l l y , the immediately preceding utterance) as i n the case of other-repetition. 4. D e f i n i t i o n CDDEF). (See Gaies, 1983; Schwartz, 1980; Snow, 1987; also Porter, 1983, 1986 re: requests for d e f i n i t i o n . ) Like prompts, d e f i n i t i o n s t y p i c a l l y serve to f i l l a gap l e f t by one of the p a r t i c i p a n t s or may be produced even before any s p e c i f i c request has been received from an i n t e r l o c u t o r , depending on the speaker's perception of the l i s t e n e r ' s l e v e l of comprehension. A d e f i n i t i o n i s o r d i n a r i l y accomplished by a speaker producing a statement on the meaning of an i d e n t i f i e d object which i s unfamiliar to the l i s t e n e r but included within a class of objects which i s presumably f a m i l i a r to the l i s t e n e r (e.g., A wrench i s a kind of t o o l ) . In conversational s i t u a t i o n s where pr o f i c i e n c y l e v e l s d i f f e r between pa r t i c i p a n t s , a d e f i n i t i o n would generally be produced by a more p r o f i c i e n t speaker i n a i d of a le s s p r o f i c i e n t speaker's understanding. D e f i n i t i o n thus serves as a marker of how the more p r o f i c i e n t speaker perceives the l e x i c a l competence of the l e s s p r o f i c i e n t speaker. 5. Display Question (DDO). (See Brulhart, 1985; Duff, 1986; Long & Sato, 1983.) Sometimes known as r h e t o r i c a l , t e s t , evaluative or known-information questions, display questions request demonstration of knowledge or information already possessed by the speaker—and known by the l i s t e n e r to be possessed by the speaker. In teaching s i t u a t i o n s , display questions are frequently intended to serve an 71 i n s t r u c t i o n a l purpose, and thus the p a r t i c u l a r content covered by the question would form part of the i n s t r u c t i o n a l syllabus. Display questions may occur i n settings other than those which are e x p l i c i t l y i n s t r u c t i o n a l . One i m p l i c i t assumption behind a display question, regardless of the se t t i n g i n which i t i s asked, i s that a conversational partner probably does not know, but ought to know, the s p e c i f i c content on which the question i s based. A l t e r n a t i v e l y , an additional assumption behind a display question i s that even though a l i s t e n e r may know the content focus of a display question, i t i s necessary to t e s t the knowledge. A question of the sort What do I have i n mv hand? ( l i s t e n e r s are able to see what i s i n the hand) i s an i l l u s t r a t i o n of the point. 6. Echo (EECH). (See Gass & Varonis, 1986.) One other form of other-repetition, echo, i s s i g n a l l e d by f l a t or f a l l i n g intonation and thus does not seem to serve as an in d i c a t i o n of incomplete understanding, but rather functions to pick out or reinforce the introduction of new information by one of the pa r t i c i p a n t s . I t thus has the po t e n t i a l to encourage addi t i o n a l t a l k . An echo i s , by d e f i n i t i o n , an exact ( p a r t i a l or complete) r e p e t i t i o n , of an immediately preceding utterance. 7. Lex i c a l Uncertainty (LLEX). (See Porter, 1983, 1986; Schwartz, 1980; Tarone, 1983.) Indications of l e x i c a l uncertainty represent possible t r i g g e r s f o r such conversational behavior as d e f i n i t i o n s , comprehension checks 72 or prompting and may take such forms as a search f o r a s p e c i f i c word or pausing to indicate to an i n t e r l o c u t o r that l e x i c a l material i s not immediately at hand. Indications of l e x i c a l uncertainty open up opportunities for other-repair which may or may not be taken up by a partner i n a given context. 8. R e f e r e n t i a l Question (RRQ). (See Brulhart, 1985; Duff, 1986; Long & Sato, 1983.) A r e f e r e n t i a l question i s designed to e l i c i t information which i s unknown to the speaker but which may be possessed by the hearer. An i n t e r l o c u t o r ' s p o t e n t i a l response to the r e f e r e n t i a l question, moreover, must be of i n t e r e s t to the source of the question. Referential content i s t y p i c a l l y generated by the t o p i c being considered; that i s , i t i s not part of the p a r t i c i p a n t ' s negotiation of meaning (which takes place outside of and temporarily removed from the t o p i c a l content). Given t h i s formulation, a request f o r c l a r i f i c a t i o n would be external to the t o p i c and thus distinguishable from a r e f e r e n t i a l question. 9. Self-expansion (SSEXP). (See E l l i s , 1984, 1985.) A self-expansion i s a p a r t i a l or complete rephrasing of one's own utterance and i s thus distinguished from elaboration of another speaker's utterance (see other- expansion, below), a form of other-repair. I t can be viewed as a form of s e l f - r e p a i r which t y p i c a l l y occurs within the current speaker's turn but may occur within the speaker's next a v a i l a b l e turn (see, also, s e l f - r e p e t i t i o n and other-73 e p e t i t i o n , below). Self-expansion extends the i n t e r p r e t a b i l i t y and refines the meaning of the speaker's i n i t i a l utterance. 10. S e l f - r e p e t i t i o n (SSREP). (See Brulhart, 1985; Long, 1980; Long, 1983b; Pica & Doughty, 1985; also Doughty & Pica, 1986 re: the r o l e of r e p e t i t i o n i n NS-NNS and NNS-NNS interaction.) Exact, p a r t i a l or semantic (equivalent) s e l f - r e p e t i t i o n (not including a grammatical functor) within f i v e turns of an i n i t i a l utterance indicate that the speaker wishes to emphasize or recycle conversational material of use i n a current context. This form of r e p e t i t i o n i s distinguished from a f a l s e s t a r t or s t u t t e r within an utterance i n order to emphasize i t s p o t e n t i a l function i n maintaining conversational cohesiveness at a l e v e l a speaker perceives necessary for a l i s t e n e r . 11. Other-expansion (OOEXP). (See Campbell et a l . , 1977 (cited i n Freed, 1978; Long, 1980), Ferguson, 1975; Long, 1980). The term expansion normally r e f e r s to expansion of an i n t e r l o c u t o r ' s utterances and has also been applied s p e c i f i c a l l y to r e p e t i t i o n and/or rephrasing of part or a l l or part of a previous utterance i n order to supply obligatory grammatical functors (Long, 1980, p. 84). Use of the term here i s applied to rephrasing and/or extension, but not exact r e p e t i t i o n alone, of eit h e r grammatical or propositional content i n the previous speaker's utterance. 12. Other-repetition (OOREP) (See Brulhart, 1985; Doughty & Pica, 1986; Long, 1980, 1981, 1983b; Pica & 74 Doughty, 1985; Pica & Long, 1986; Porter, 1986.) Exact, p a r t i a l or semantic r e p e t i t i o n of another p a r t i c i p a n t ' s utterance within f i v e turns nominally indicates incomplete understanding and a desire to begin r e c y c l i n g the problematic conversational material. This form of other-r e p e t i t i o n i s o r d i n a r i l y accompanied by r i s i n g intonation. The second category of dependent v a r i a b l e has been discussed under the notion of conversational reference (Halliday & Hasan, 1976; Brown & Yule, 1983a) and includes the following two exponents. 1. Exophoric Reference (EEXO). Exophoric ("pointing out" or "si t u a t i o n a l " ) reference, takes a number of forms during conversation depending on the background and s i t u a t i o n a l perception which pa r t i c i p a n t s share. Among the most common exponents of t h i s form of reference, however, are context-bound, demonstrative pronouns which point to p a r t i c u l a r objects i n the perceptual range of the speaker and hearer: i t , t h i s , that, these, those (push t h i s [e.g., function key]), i n addition to the d e f i n i t e a r t i c l e used to r e f e r to a " p a r t i c u l a r i n d i v i d u a l or subclass . . . i d e n t i f i a b l e i n the s p e c i f i c s i t u a t i o n . . . . pass me the towel; . . . the snow's too deep" (Halliday & Hasan, 1976, p. 71). As Halliday and Hasan have pointed out, however, i t i s not necessary that the thing being referred to be "ph y s i c a l l y present i n the interactant's f i e l d of perception" (p. 49). The only fundamental requirement i s that p a r t i c i p a n t s are able to share i d e n t i f i c a t i o n of the 75 t h i n g b e i n g r e f e r r e d t o . Numerous o t h e r c l a s s e s o f exponents c o u l d a l s o be i n c l u d e d (such p o s s e s s i v e d e t e r m i n e r s as mine , y o u r s , o u r s , h i s , h e r s , t h e i r s , and i t s , i n a d d i t i o n t o such p o s s e s s i v e m o d i f i e r s as my, y o u r , o u r , h i s , h e r s , e t c . ) , a l t h o u g h f o r p r e s e n t p u r p o s e s i t i s s u f f i c i e n t t o emphasize t h a t s i n c e e x o p h o r i c r e f e r e n c e l e n d s no c o h e s i o n t o spoken d i s c o u r s e i t i s marked by i t s r e l a t i v e i s o l a t i o n i n t h e here -and-now o f t h e s i t u a t i o n . 2 . A n a p h o r i c R e f e r e n c e (AANA). A n a p h o r i c ( " p o i n t i n g back" o r " t e x t u a l " ) r e f e r e n c e , i s e n t i r e l y c o h e s i v e i n f u n c t i o n ; o r d i n a r i l y i t cannot be i n t e r p r e t e d w i t h o u t r e f e r e n c e t o something a t a p r e v i o u s p o i n t i n t h e spoken o r w r i t t e n t e x t . F o r example, t h e t h e y i n Computers can be u s e d f o r wordprocess incr . a l t h o u g h t h e y a r e b e t t e r s u i t e d t o numbercrunch ing r e f e r s , o f c o u r s e , t o "computers" i n t h e i m m e d i a t e l y p r e c e d i n g independent c l a u s e and would be c o u n t e d as an i n s t a n c e o f a n a p h o r i c r e f e r e n c e . The t h e y i n They go on the r i g h t s i d e , by c o n t r a s t , would c o n s t i t u t e a c a s e o f e x o p h o r i c r e f e r e n c e i f the group o f o b j e c t s t o which t h e p a r t i c i p a n t s r e f e r can o n l y be i d e n t i f i e d by v i s u a l i n s p e c t i o n o r some o t h e r form o f s h a r e d p e r c e p t i o n , and i f t h e r e i s no p r i o r r e f e r e n c e i n the spoken t e x t . R e s e a r c h Q u e s t i o n s and Hypotheses T h i s s e c t i o n i n t r o d u c e s a s e r i e s o f r e s e a r c h q u e s t i o n s and h y p o t h e s e s which r e f l e c t the f o r e g o i n g d i s c u s s i o n and which a r e l i n k e d t o a s t r a t e g y f o r a n a l y s i s o f t h e d a t a (see T a b l e 1 ) . Each r e s e a r c h q u e s t i o n (Qn) i s i n t e n d e d t o f o c u s 76 a t t e n t i o n on an i s s u e o f p r a c t i c a l o r t h e o r e t i c a l i n t e r e s t and t o s u p p o r t e x a m i n a t i o n o f e n s u i n g q u e s t i o n s and h y p o t h e s e s . Each h y p o t h e s i s (Hn) i s s t a t e d i n d i r e c t i o n a l o r n o n - d i r e c t i o n a l form depending on the r e s e a r c h purpose and numbered f o l l o w i n g the s t e p s o f the a n a l y s i s s t r a t e g y . A b r i e f d e s c r i p t i o n o f t h e s t a t i s t i c s employed t o t e s t t h e h y p o t h e s e s and the forms o f t a b u l a r a n a l y s i s f o l l o w s t h e l i s t i n g o f h y p o t h e s e s . The f i r s t two q u e s t i o n s a r e p r o c e d u r a l (PQn) and w i l l no t be c o n s i d e r e d p a r t o f the h y p o t h e s i s t e s t i n g r e l a t e d t o r e p a i r and r e f e r e n c e . They w i l l , however, form t h e i n i t i a l p a r t o f t h e a n a l y s i s s t r a t e g y , s i n c e t h e answers t h e y g e n e r a t e w i l l h e l p t o de te rmine how dependent v a r i a b l e s i n t h e r e s e a r c h d e s i g n w i l l be t r e a t e d (whether o r no t t h e y w i l l have t o be t r a n s f o r m e d , f o r example) p r i o r t o h y p o t h e s e s t e s t i n g . PQ1: How homogeneous a r e the t e a c h e r - l e d g roups? PQ2: What i s the d i s t r i b u t i o n o f f r e q u e n c i e s f o r r e p a i r and r e f e r e n c e between and w i t h i n g roups? Q l : Do group t y p e and t a s k i n f l u e n c e the use o f r e p a i r ? HI : The f r e q u e n c y o f r e p a i r i n dyads does no t v a r y s i g n i f i c a n t l y by group membership o r t y p e o f t a s k p e r f o r m e d . Q2: Do group t y p e and t a s k i n f l u e n c e the use o f r e f e r e n c e ? H2: The f r e q u e n c y o f r e f e r e n c e i n dyads does no t v a r y s i g n i f i c a n t l y by e i t h e r group membership o r t a s k p e r f o r m e d . 77 Q3: Is e x p e r i e n t i a l a c t i v i t y a b e t t e r s o u r c e o f r e p a i r b e h a v i o r t h a n e x p o s i t o r y a c t i v i t y ? H3: R e p a i r o c c u r s s i g n i f i c a n t l y more f r e q u e n t l y d u r i n g t a s k s wh ich emphasize e x p e r i e n t i a l a c t i v i t y t h a n d u r i n g t a s k s which emphasize e x p o s i t o r y a c t i v i t y . Q4: How a r e a n a p h o r i c and e x o p h o r i c r e f e r e n c e r e l a t e d t o e x p o s i t o r y and e x p e r i e n t i a l t a s k a c t i v i t y ? H4a: A n a p h o r i c r e f e r e n c e o c c u r s s i g n i f i c a n t l y more f r e q u e n t l y d u r i n g t a s k s which emphasize e x p o s i t o r y a c t i v i t y t h a n d u r i n g t a s k s wh ich emphasize e x p e r i e n t i a l a c t i v i t y . H4b: E x o p h o r i c r e f e r e n c e o c c u r s s i g n i f i c a n t l y more f r e q u e n t l y d u r i n g t a s k s wh ich emphasize e x p e r i e n t i a l a c t i v i t y than d u r i n g t a s k s which emphasize e x p o s i t o r y a c t i v i t y Q5: What a r e t h e t e x t u a l p r o f i l e s o f the most f r e q u e n t l y o c c u r r i n g forms o f r e p a i r , p a r t i c u l a r l y t h o s e which c h a r a c t e r i s t i c a l l y appear i n c o m b i n a t i o n w i t h each o t h e r ? No h y p o t h e s e s were t e s t e d f o r t h i s q u e s t i o n , a l t h o u g h g e n e r a l i z a t i o n s about t h e q u a l i t i e s o f r e p a i r i n c o n t e x t w i l l be d e v e l o p e d f o l l o w i n g r e p o r t o f the r e s u l t s o b t a i n e d t h r o u g h the a n a l y s i s o f v a r i a n c e . A summary o f c o n d i t i o n s under which t h e s e h y p o t h e s e s were t e s t e d , i n c l u d i n g t e s t s t a t i s t i c s , d i r e c t i o n a l i t y , and t h e number and l o c a t i o n o f t a b l e s o r t a b u l a r summaries, i s 78 l i s t e d i n Table 1. Table 1 Summary of Conditions for Testing Hypotheses Relating to Group, Task, Reference and Repair C r i t i c a l values Question/ Number hypothesis S t a t i s t i c a l of number te s t s d i r e c t i o n p_ tables* PQ1 Friedman 2-way ANOVA/ Chi-square 2 - t a i l e d < .025 1 Sum. PQ2 Sq. root/log. trans./Comp. alpha l e v e l s 1 Sum. (App. G) HI ANOVA/F 2-t a i l e d < .025 12, 2 x 5 1 Sum. H2 II II II 2, 2 X 5 1 Sum. H3 II 1 - t a i l e d < .05 1 (Sum.) 6, 2 X 2 (App. I) 6, 2 X 2 (App. J) H4a II II • i 1, 2 X 2 H4b II II II 1, 2 X 2 * The number of tables indicated does not include post-hoc analyses. Since re p a i r has 12 exponents and reference has 2 exponents i n t h i s study, the number of analyses i s considerably larger than the number of hypotheses. In a l l , 33 ANOVA tables were produced (including the main and summary tables found i n the text or appendices), i n addition to a number of post-hoc comparisons made whenever major sources of variance needed to be located. In general, the research design was conceived to move s t r a t e g i c a l l y through the process of analysis, requiring that c e r t a i n hypothesis-testing or v a l i d a t i o n procedures be completed before beginning others. This issue i s more f u l l y developed i n the following section. General Strategy for Data Analysis The f i r s t phase of the strategy (PQ1-PQ2) was directed towards the i n t e r n a l v a l i d i t y of the design and a decision over whether or not to transform dependent variables which showed a skewed d i s t r i b u t i o n . Two preliminary treatments of the data were performed. The f i r s t treatment concerned the degree of homogeneity found within each of the groups (n = 6) formed fo r the study. B a s i c a l l y , the problem was to determine the sources of any differences within groups with respect to use of repair and reference during the f i v e tasks. Although the composition of dyads within the groups was c o n t r o l l e d f o r p r o f i c i e n c y i n English and status (teachers and students i n each dyad) i t was assumed that i n d i v i d u a l differences would probably emerge on some of the variables during performance of some of the tasks. Given t h i s point of view, i t became necessary to determine the sources and account for any pattern of differences within the groups p r i o r to conducting and attempting to i n t e r p r e t analyses of variance. In order to do t h i s , a Friedman Two-way Analysis of Variance by Ranks (Siegel, 1956) was 80 conducted f o r each group (see Table 2 for a summary of the s t a t i s t i c a l a n a l y s i s ) . The second preliminary treatment of the data determined the need f o r transformation of the frequencies recorded f o r each dependent variable used i n the study. Tabachnick and F i d e l l (1983) note that the "F t e s t i s robust to v i o l a t i o n s of normality and homogeneity of variance, as long as sample sizes are r e l a t i v e l y equal, but not to skewness" (p. 77). Although excessively skewed d i s t r i b u t i o n s are candidates for transformation, the authors also note that i n practice the advantages of attempting to normalize d i s t r i b u t i o n s may be small, p a r t i c u l a r l y when the transformed variables are rendered more d i f f i c u l t to in t e r p r e t . The approach used here was to se l e c t randomly seven of the 14 dependent variables, apply both square root and logarithmic transformations to each of these, and then compare the e f f e c t s for group and task i n repeated-measures ANOVAs performed on the variables i n both transformed and untransformed states. The s e l e c t i o n included variables with severe p o s i t i v e skewness, moderate skewness and near-normal skewness. As i n the case of the discussion of within-group homogeneity, a b r i e f summary of the r e s u l t s of these t e s t s (Table 3) i s presented below i n Preliminary Treatment of the Data. The second l e v e l of strategy (H1-H2) entai l e d a conservative approach to t e s t i n g group and task differences i n the use of repair and reference. The two n u l l hypotheses 81 at t h i s l e v e l propose no differences between group and tasks and require a l e v e l of s i g n i f i c a n c e which i s twice as stringent as that normally required for d i r e c t i o n a l hypotheses (p_ < .025). This i s due to the use of a r e l a t i v e l y novel conceptual approach to tasks and groups i n the design, and thus to the exploratory nature of the research questions and associated hypotheses. At the same time i t should be stressed that t h i s l e v e l of analysis i s the key to further treatment and i n t e r p r e t a t i o n of the data. Results obtained from t h i s phase of the study would be used to construct the s p e c i f i c components of an approach to hypothesis t e s t i n g i n the following phase. Accordingly, once these i n i t i a l hypotheses were tested, i t was then possible to pursue the r e s u l t s more a g g r e s s i v e l y — t o argue, i n e f f e c t , that the add i t i o n a l hypotheses were founded i n the previously tested ones—and t e s t (with d i r e c t i o n a l hypotheses and l e s s stringent p r o b a b i l i t y levels) additional hypotheses about the r e l a t i o n s h i p of group and task to repair and reference. This t h i r d phase of the strategy (H3-H4b) i s based on combining and s e l e c t i n g tasks on both conceptual and empirical ( i . e . , p r i o r hypothesis-testing) grounds. In p a r t i c u l a r , i t was designed to d i r e c t analysis of repair and reference to tasks which appear as concentrated sources of e x p e r i e n t i a l or expository behavior. The fourth phase of the study extended the r e s u l t s of the previous phase into a q u a l i t a t i v e examination of 82 patterns of r e p a i r i n t r a n s c r i p t i o n s excerpted from two overlapping areas of the research design: teaching tasks which employ extensive expository behavior and non-teaching tasks characterized by p a r t i c i p a n t s ' e x p e r i e n t i a l behavior. The procedures involved at t h i s l e v e l of q u a l i t a t i v e analysis included describing and contrasting formal and .functional q u a l i t i e s of the selected sets of v a r i a b l e s . The e s s e n t i a l point of t h i s phase of the study was to capture patterns and r e g u l a r i t i e s within the data which were not pursued or adequately described through analysis of variance. Methodology Selection and Treatment of Subjects Subjects f o r the study were selected from the membership of two public u n i v e r s i t y English Speaking Society (ESS) clubs ( t o t a l membership = 45) located i n the Osaka-Kobe area of western Japan, and from a l i s t of 14 u n i v e r s i t y English teachers (seven Japanese and seven native speakers of English) known to the researcher. The object of s e l e c t i n g from among t h i s group of 59 prospective subjects was to form an equal number of teacher-led NS-NNS and NNS-NNS dyads. A l l prospective subjects received a general explanation of the research and i n v i t a t i o n to p a r t i c i p a t e under s p e c i f i c conditions (Appendix B). The conditions d i f f e r e d somewhat depending on whether the prospective subjects were ESS members or teachers. ESS members were asked to take a standardized English language 83 p r o f i c i e n c y t e s t , the CELT (Comprehensive English Language Test) - Structure (see Appendix D). Members scoring i n an intermediate range (65 to 80 percent, see norms published i n Harris & Palmer, 1986a) would be asked to take a standardized, o r a l p r o f i c i e n c y examination i n English, the LPI—Language Proficiency Interview (see Appendix D; Educational Testing Service, 1982), to confirm the i n i t i a l f i n d i n g of intermediate p r o f i c i e n c y based on the CELT and to e s t a b l i s h a l e v e l for conversational a b i l i t y — t h a t i s , f o r a l e v e l of competence which would be exercised during performance of the communication tasks. Members scoring between 1+ and 2 on the LPI following a 15-minute telephone conversation (roughly an intermediate range on the scale between 0, no a b i l i t y to communicate i n the language and 5, equivalent to an educated native speaker) would be i n v i t e d to p a r t i c i p a t e i n a series of dyadic conversations with e i t h e r a Japanese or a native speaker of English. The Japanese teachers were also asked to take the LPI and i n v i t e d to p a r t i c i p a t e only i f t h e i r score was 3 or greater (professional competence i n the language). No such t e s t s were administered to the NSs of English. The main purpose of s e l e c t i n g subjects by English p r o f i c i e n c y was to ensure that a l l dyads would consist of p a r t i c i p a n t s at comparable l e v e l s , that i s , a learner at an intermediate l e v e l of p r o f i c i e n c y and a teacher of native or near-native p r o f i c i e n c y . 84 Assignment to Dyads Once a pool of prospective p a r t i c i p a n t s had been obtained, the process of assignment to eit h e r the Japanese-led or native-led dyads was i n i t i a t e d . The names of a l l ESS members whose pro f i c i e n c y was tested at an intermediate l e v e l were shuffled and randomly assigned to eit h e r a NS or a NNS teacher. This process continued u n t i l each of the 14 teachers was matched with an ESS member. Next, i n d i v i d u a l s i n each matched group were contacted i n order to arrange for a recording date. Whenever ESS members indicated that t h e i r schedule would not i n f a c t permit matching and recording with a teacher at any of the dates, times and places suggested by the researcher, the member was dropped from further p a r t i c i p a t i o n and the next member on the l i s t was contacted and asked to p a r t i c i p a t e . Because of scheduling d i f f i c u l t i e s , two matched and scheduled groups could not be accommodated and were dropped from the study. Eventually 12 dyads were scheduled and completed p a r t i c i p a t i o n i n the balance of the study. Although no attempt was made to a l l o c a t e s p e c i f i c proportions of male and female learners to the mixed and homogeneous groups, one male and f i v e females were allocated to the native-led group and two males and f i v e females to the Japanese-led group. This representation of learners within the research groups approximates the r a t i o of males to females i n the two ESS clubs, about 1:4, although opportunistic s e l e c t i o n of the teachers resulted i n a r a t i o 85 of 11:1 across both research groups. A l l subjects completed a form i n d i c a t i n g informed consent to p a r t i c i p a t i o n (Appendix C). Data C o l l e c t i o n Sites Although the 12 teachers indicated a willingness to p a r t i c i p a t e at any convenient l o c a l s i t e , i t was f e l t that the ESS members would o f f e r more relaxed assistance i f they could be recorded on t h e i r own campus. Accordingly, most ESS members were recorded i n conversation with a teacher (to whom they had not been introduced previously) on t h e i r own campuses; two were recorded on the a l t e r n a t i v e campus because of scheduling d i f f i c u l t i e s . Eight recording sessions were conducted i n a p a r t i t i o n e d area of the researcher's o f f i c e . Four sessions were conducted on the a l t e r n a t i v e campus i n the area of a language laboratory reserved for small group conversations. In a l l cases, the researcher was present i n the same room as the subjects, although the technical nature of the recording (see Data C o l l e c t i o n Procedures, below) permitted the researcher to "ignore" the dyad—to s i t apart from the dyad and engage i n a c t i v i t y unrelated to the dyadic conversations. C o l l e c t i o n and Coding of Data Task Order A l l dyads experienced the same f i v e communication tasks i n an order dictated by a standard L a t i n Square assignment of task order to each of the dyads i n eit h e r the mixed or homogeneous group types (see Ferguson, 1981; Eames et a l . , 86 1985). In pra c t i c e , t h i s meant that the order of the f i r s t dyad's tasks would be rotated by one task f o r the second dyad's scheduled combination, and so on u n t i l the l a s t dyad f o r the group (NS-NNS or NNS-NNS) had worked through i t s scheduled tasks. The rati o n a l e f o r doing t h i s was to reduce the carry-over e f f e c t s which may be produced when a l l subjects undergo the same treatment order (Eames et a l . , 1985). Task Description The f i v e communication tasks used i n the study included 1) COMl, a lecture on how to f i n d character s t r i n g s i n a text through use of the word processing program of a small, "laptop" computer (the NEC 8201A) without the computer p h y s i c a l l y present, 2) COM2, a demonstration of how to f i n d character strings on the NEC 8201A with the computer p h y s i c a l l y present, 3) DIS, informal discussion of any topic of mutual i n t e r e s t to the pa r t i c i p a n t s (tr a v e l , holiday plans, computer, and so on), 4) LEG1, reconstruction of a small Lego (snap-together) toy with the pa r t i c i p a n t s s i t t i n g back-to-back and using a set of sequenced, graphic i n s t r u c t i o n s supplied with the toy, and 5) LEG2, face-to-face reconstruction of a Lego toy of s i m i l a r d i f f i c u l t y (re: LEG1) with one pa r t i c i p a n t giving the inst r u c t i o n s as the other assembled the pieces. During the computer-based tasks, the teacher supplied information i n an attempt to i n s t r u c t the learner i n use of the s t r i n g search function, although the teacher had been 87 b r i e f e d to request periodic summaries from the learner. In addition, the hands-on demonstration involving the search function encouraged rather free exchange, questioning and gesturing by both p a r t i c i p a n t s . Both computer tasks focused on the teaching and learning of a p a r t i c u l a r computer function. The Lego reconstruction tasks, i n contrast, were not so much instruction-oriented as problem-oriented. Although the teacher was asked to convey i n s t r u c t i o n s without allowing the learner to see them, the Lego task required verbal cooperation from both p a r t i c i p a n t s i n order to work towards reconstruction of the object. The e s s e n t i a l difference between the two Lego tasks was that one (LEG1) required p a r t i c i p a n t s to communicate without v i s u a l feedback whereas the other (LEG2) made v i s u a l feedback the center of the a c t i v i t y . (See Wagner, 1983 f o r further discussion of Lego i n conversation strategy research; Littlewood, 1981 for a d e s c r i p t i o n of Lego used i n communicative language teaching.) The f i n a l task, open discussion of any i n t e r e s t i n g topic, resembled Long's (1980) unguided dyadic conversations: Cooperation i s not directed towards the s o l u t i o n of a problem, nothing need be taught, no physical materials are hidden from view and yet none are t y p i c a l l y a v a i l a b l e as conversational resources. This task was selected to allow p a r t i c i p a n t s a chance to structure t h e i r t a l k as background and knowledge dictated. 88 Data C o l l e c t i o n Procedures P r i o r to recording the dyadic conversations, a l l pa r t i c i p a n t s were br i e f e d as to the nature of the data c o l l e c t i o n . The teachers were sent a packet containing a de s c r i p t i o n of each task and additional materials which i l l u s t r a t e d use of the s t r i n g search function on the computer. The teachers and the researcher met a few days before a scheduled recording session and further b r i e f e d on the operation of the tasks with the learners. In p a r t i c u l a r , teachers had an opportunity to p r a c t i c e use of the computer and assembly of at l e a s t one of the Lego sets. A few minutes p r i o r to the scheduled recording, the learners also received a general b r i e f i n g on the nature of the tasks and t h e i r p a r t i c i p a t i o n i n them. Although they were not b r i e f e d i n d e t a i l on what to expect, they were encouraged to p a r t i c i p a t e a c t i v e l y i n the conversations and to ask for information of use i n completing a task whenever they needed i t . Just before recording commenced the teacher and learner were introduced and, depending on the scheduled f i r s t task, sat facing away from each other, towards each other, or side-by-side. The seating arrangements f o r a l l tasks are diagrammed as follows: 89 C0M1: [][] (side-by-side) A A T L COM2: T>[][][] LEG2: T>[]" and " A " symbols. Depending on the order of tasks established f o r a given dyad, the researcher arranged desks and b r i e f e d p a r t i c i p a n t s for t h e i r f i r s t task. During breaks between the remaining tasks, desks were arranged i n the appropriate pattern and p a r t i c i p a n t s b r i e f e d as to t h e i r a c t i v i t y on the ensuing task. A l l tasks were conducted with the p a r t i c i p a n t s s i t t i n g and facing i n the appropriate d i r e c t i o n . Once the p a r t i c i p a n t s had taken t h e i r i n i t i a l p o s i t i ons, they were n o t i f i e d that they would continue each of the tasks uninterrupted f o r seven minutes and that they would have a two-minute break between each task. Task-specific instructions given to the p a r t i c i p a n t s j u s t before beginning the tasks included the following: 1) C0M1: The teacher was asked to teach the learner how to operate the computer's s t r i n g search without d i r e c t manipulation of the computer or recourse to the text f i l e s i t contained. (The teacher had previously studied a three-90 page de s c r i p t i o n of the function and rehearsed i t s operation.) The learner was n o t i f i e d that the teacher would occasionally request a summary of the in s t r u c t i o n s . 2) COM2: The teacher was asked to teach use of the search function through manipulation of the appropriate keys and use of any text f i l e i n the computer's memory. Partic i p a n t s were n o t i f i e d that they could point to or touch anything of use i n the task s i t u a t i o n , including the keys and screen display. 3) DIS: Participants were asked to agree on a topic of mutual i n t e r e s t shortly before the task began and to discuss the selected t o p i c " f r e e l y " , that i s , without any attempt to teach or learn anything i n p a r t i c u l a r . Both p a r t i c i p a n t s were s p e c i f i c a l l y encouraged to contribute to the discussion whenever i t seemed appropriate to do so. The researcher supplied a topi c whenever the pa r t i c i p a n t s were unable to make t h e i r choice during the break. 4) LEG1: The teacher was handed a set of v i s u a l ( i . e . , non-text) in s t r u c t i o n s for the Lego toy and asked to r e l a t e information on the correct assembly of the toy to the learner. The teacher was also asked to help the learner to assemble the toy without looking at the learner's work. The disassembled toy was scattered on the learner's desk and the learner asked to assemble the toy i n response to information supplied by the teacher. The learner was also instructed to keep the teacher informed as to whether an i n d i v i d u a l step i n the procedure had been completed. 91 5) LEG2: The teacher received the v i s u a l i n s t r u c t i o n s but was asked not to show them to the learner. The disassembled toy was scattered on the learner's desk and the learner asked to assemble the toy i n response to the teacher's i n s t r u c t i o n s . However, both p a r t i c i p a n t s were informed that they were free to point to objects, but that the teacher could not p h y s i c a l l y pick up and assemble pieces on the learner's behalf. Each task was started by a verbal i n s t r u c t i o n to the p a r t i c i p a n t s to begin. As the signal to commence work was given, the researcher moved to the opposite end of the recording room, sat down at a desk and started a stopwatch and the recording equipment. No further communication between the researcher and the p a r t i c i p a n t s occurred during performance of the task, except for the verbal s i g n a l to the p a r t i c i p a n t s to stop t h e i r work a few seconds beyond the seven-minute mark. Video and audio cassette recordings were made of a l l tasks. Videotaping was intended to provide a p a r a l l e l record of the tasks which could be used to i n t e r p r e t problematic points i n the audio record. Although videotaping can be a more obtrusive method of data c o l l e c t i o n than audiotaping, care was taken to make the videotaping as unobtrusive as possible. The video camera was placed approximately 2.5 meters from the subjects, focused j u s t before conversation began and then l e f t untouched f o r the duration of the f i v e tasks. Also, 92 videotaping operations were started and stopped remotely from the researcher's p o s i t i o n . Audio taping employed two, c l i p - t y p e microphones (one fo r each dyad member) with 3-meter cords feeding independently into a junction and then plugged into the microphone jack of a cassette recorder. Recorder operation was c o n t r o l l e d from the researcher's p o s i t i o n by use of a remote control switch and cable. The two-minute breaks between tasks were used by the p a r t i c i p a n t s to relax, or by the researcher to accomplish such housekeeping functions as checking the equipment and giving i n s t r u c t i o n s f o r the next task. A complete session thus required l e s s than one hour. Total recording time f o r both of the groups (mixed and homogeneous) was about 420 minutes or seven hours. Coding and Treatment of the Data The seven hours of recorded t a l k on the tasks were transcribed ( t r a n s c r i p t i o n conventions are l i s t e d i n Appendix F; t r a n s c r i p t i o n samples are i n Appendix E) and recorded on a floppy disk f o r l a t e r manipulation with a personal computer. Each task was transcribed as a separate f i l e and required an average of six-and-a-half typed pages of text. About 32.5 pages were required f o r each dyad and approximately 390 pages for the en t i r e corpus. Text corresponding to the f i r s t minute of t r a n s c r i p t i o n was l e f t uncoded; the following s i x minutes were coded and served as the basis f o r determining frequencies for r e p a i r and reference, the dependent varia b l e s . Coded t r a n s c r i p t s for 93 each dyad's s i x minutes of t a l k averaged about 5.5 pages i n length or about 330 pages fo r the coded corpus. The following coding categories (in parentheses) were then added to the typed t r a n s c r i p t s : 1) repair: c l a r i f i c a t i o n request (CCLAR), comprehension check (CCOM), confirmation check (CCON), d e f i n i t i o n (DDEF), display question (DDQ), echo (EECH), l e x i c a l uncertainty (LLEX), other-expansion (OOEXP), other-repetition (OOREP), r e f e r e n t i a l question (RRQ), self-expansion (SSEXP), s e l f -r e p e t i t i o n (SSREP); 2) reference: anaphora (AANA), exophora (EEXO). Repairs and forms of reference o r i g i n a t i n g with the learner were a d d i t i o n a l l y coded with an S following the main code (thus, f o r example, RRQS). Coding R e l i a b i l i t y . Nine NSs of English were trained by the researcher to recognize seven of the 14 repair and reference categories i n context: comprehension checks, confirmation checks, display questions, indications of l e x i c a l uncertainty, r e f e r e n t i a l questions, exophoric reference and anaphoric reference. Although the REs were selected randomly, both categories of reference were included d e l i b e r a t e l y because of t h e i r key conceptual p o s i t i o n i n the study. The nine coders were given 21 short excerpts (one to three turns long) of transcribed t a l k selected from the dyadic conversations of both groups and asked to a l l o c a t e each excerpt to one category among the seven ava i l a b l e . S p e c i f i c a l l y , the coders were asked to decide on a category f o r a word or 94 phrase underlined i n the excerpt. Three examples of each category appeared on the raters ' forms; these were randomly selected by the researcher for order of i n c l u s i o n on the forms. Using index numbers for categories i n the nine sets of coded excerpts i n addition to the researcher's o r i g i n a l coding of the same texts (k = 10), Kendall's C o e f f i c i e n t of Concordance W was calculated at .908 (Chi-square = 181.68, df = 20, p_ < .0001 ), a l e v e l of i n t e r - r a t e r r e l i a b i l i t y considered adequate for the study. Analysis of Transcripts. The frequency of each code i n the discourse of each transcribed task was counted by running the word frequency program of a s p e l l i n g checker widely used with a v a r i e t y of word processors (The WORD Plus, see Holder, 1982). Ranked frequencies by task for a l l dependent variables were then compared with a Friedman Two-way Analysis of Variance by Ranks within each group type (mixed and homogeneous) following the plan outlined above i n General Strategies for Data Analysis. Analysis of variance within the scope of the general strategy was based on comparison of means for the various exponents of re p a i r and reference, i n d i v i d u a l l y , by task and group: One exponent was the basis of each ANOVA table (excluding summary ta b l e s ) . Repair frequencies were compiled f o r each speaker i n each dyad—making teacher-learner comparisons f e a s i b l e f o r future use of the data; beyond t h i s and the counting of d e s c r i p t i v e data by dyad 95 p a r t i c i p a n t , the hypothesis t e s t i n g proceeded on the basis of comparing group by task without d i s t i n g u i s h i n g between teachers and learners. The basic unit of analysis f o r the quantitative phases of the study thus remained the dyad. Whenever an e f f e c t reached s i g n i f i c a n c e according to the pre-set c r i t e r i a (see Table 1, above), Tukey's HSD multiple-comparison procedure was run on a l l combinations of group means within the ANOVA to locate the main sources of variance. Sources of variance (at both p_ < .05 and < .01) were ranked from highest to lowest and appear on the summary ANOVA tables i n Chapter 4. Tukey's HSD t e s t was selected because i t represents a balance between power and conservative approaches to multiple comparison (Huck, et a l . , 1974; Nie et a l . , 1975), and i s a widely known approach to post hoc analysis for groups with equal n's (Ferguson, 1981). A l l analyses were conducted on a Macintosh Plus personal computer using StatView 512+ (Feldman & Gagnon, 1986) and CLR ANOVA (Clearlake Research, 1985). The sampling procedure and focus of the q u a l i t a t i v e analysis (Chapter 6) were based on the r e s u l t s of the analysis of variance. The f i r s t c r i t e r i o n f o r s e l e c t i o n of t r a n s c r i p t s was the s i g n i f i c a n t l y frequent use of a repair exponent i n a p a r t i c u l a r task. Given post-hoc analysis of s i g n i f i c a n t e f f e c t s , i t was possible to locate the main source(s) of variance by task. Thus, REs showing the most s i g n i f i c a n t e f f e c t s served as pointers to a r e l a t i v e l y small number of t r a n s c r i p t s , so that, for example, examination 96 of the form, function and context of display questions would focus on the 12 t r a n s c r i p t s of the task i n which display questions occurred most frequently. An a d d i t i o n a l c r i t e r i o n for s e l e c t i o n required a preliminary sampling of a group's t r a n s c r i p t s to see whether REs appeared to co-occur, that i s whether re p a i r was accomplished i n some patterned way so as to suggest a closer look at how such co-occurring REs were involved i n negotiated exchanges. One way of looking at t h i s basis for s e l e c t i o n i s that i t leads to examination of n a t u r a l l y occurring p a i r s which would not otherwise be studied i n a research design emphasizing treatment of dependent vari a b l e s , one at a time. Prospective candidates f o r study of how d i f f e r e n t REs function together, for example, might include expression of l e x i c a l uncertainty co-occurring with d e f i n i t i o n . One extension of t h i s way of organizing examination of the t r a n s c r i p t s i s the notion of overlapping task categories (see the discussion of complementary task structures i n Chapters 5 and 6). This notion was mentioned b r i e f l y i n connection with the categorical structure of the research design. I t re-emerges i n connection with the q u a l i t a t i v e strategy because i t o f f e r s a method fo r s e l e c t i n g a very l i m i t e d number of REs which occur frequently together i n t r a n s c r i p t s which f i t the overlapping descriptions. An example of t h i s would be the set of 12 t r a n s c r i p t s which f a l l within the category of expository approaches to teaching tasks, that i s , s e l e c t i o n of the 12 97 t r a n s c r i p t s for COMl. The choice of which parts of these t r a n s c r i p t s to excerpt and compare was handled o p p o r t u n i s t i c a l l y ; that i s , i t was based on the researcher's best judgement following a process of es t a b l i s h i n g and r e v i s i n g categories i n which to elaborate the various forms of the RE (or set of REs) under consideration. This sort of pragmatic (as opposed to p r o b a b i l i s t i c ) sampling i s further discussed i n Goetz and Le Compte (1984) and Merriam and Simpson (1984). Preliminary Treatment of the Data Assessing Homogeneity within Groups Table 2 summarizes the l e v e l of homogeneity within each group (that i s , the l e v e l of i n d i v i d u a l dyadic differences within each group) by a l l dependent variables as tested by Friedman's Two-way Analysis of Variance by Ranks. A l l Chi-square values have been corrected f o r t i e d ranks (thus increasing somewhat the chances of obtaining a s i g n i f i c a n t value for Chi-square). Three of the REs and one form of reference l i s t e d i n Table 2—confirmation checks, display questions, indications of l e x i c a l uncertainty and exophora—were employed more frequently by one or more of the dyads within each of the groups than by the remaining dyads i n the groups. This evidence of heterogeneity can be put into some perspective by pointing out that the repair behavior of the mixed group was p a r a l l e l e d by that of the homogeneous group (the sources 98 Table 2 Level of Homogeneity Within Groups by Dependent Variable Chi-square by group Variable Mixed Homogeneous Major source of rank order differences (mixed/homogeneous) C l a r i f i c a t i o n request 8.89 Comprehension check 11.87* Confirmation check 8.74 D e f i n i t i o n 8.12 Display question 16.72* Echo 6.62 Lexi c a l uncertainty 12.73* Other-expansion 12.91* Other-r e p e t i t i o n 4.13 Ref e r e n t i a l question 15.14* S e l f -expansion 8.89 S e l f -r e p e t i t i o n 5.28 Anaphora 3.06 Exophora 17.20* 8.44 14.96* LEG1 > DIS/LEG1 > DIS 17.64* 8.55 .67 LEG2 > DIS 14.07* C0M1 > LEG1/COM1 > LEG2 8.14 12.96* DIS > LEG2/DIS > LEG2 LEG1 > COM2 DIS > COM1 4.85 8.79 4.21 14.31* — COM1 > DIS 8.17 17.52* LEG2 > LEG1/LEG2 > LEG1 Note, df = 4 i n a l l cases * p_ < .025 99 of s i g n i f i c a n t differences were the same) and noting, prospectively, that v i r t u a l l y a l l sources of within-group differences indicated during preliminary treatment of the data were r e p l i c a t e d during the analysis of variance (see Chapter 4, Table 22). One way of looking at these p a r a l l e l r e s u l t s i s that p r o f i c i e n c y i n English was apparently not responsible f o r within-group differences i n the homogeneous (Japanese-Japanese) group since these differences were also found i n a group with native speakers of English. Another i s that i n d i v i d u a l or i d i o s y n c r a t i c differences are also u n l i k e l y to have been responsible for these p a r t i c u l a r differences since they occurred i n both groups. Moreover, i n two cases (use of comprehension checks and display questions, i . e . , behaviors t y p i c a l l y associated with teachers conducting instruction) the p a r a l l e l outcomes suggest that some of the teachers i n both groups may have been performing i n a s i m i l a r "teacherly" fashion. Without developing t h i s discussion beyond a f a i r l y simple l e v e l of explanation, i t may be useful to mention that these preliminary r e s u l t s a n t i c i p a t e those obtained i n the analysis of variance i n terms of fundamental between-group s i m i l a r i t i e s and sources of variance within tasks. Although the homogeneity of the groups i s c l e a r l y a mixed a f f a i r (about h a l f of the dependent variables showed some degree of within-group difference) a number of areas i n which homogeneity was not demonstrated turn out to be recurrent patterns which expand 100 to become c h a r a c t e r i s t i c of both groups during performance of s p e c i f i c tasks. Assessing the use of Transformations Table 3 l i s t s seven of the 14 dependent variables ranging from the l e a s t to the greatest degree of skewness. Table 3 Comparison of Selected Transformed and Untransformed Variables by Significance of ANOVA E f f e c t s Treatment of dependent v a r i a b l e by transformation and • l e v e l of s i g n i f i c a n c e (group-task) Dep. Normality var. Skew Test p_ Sq. root Log. Untrans. Comp. check 1. 851 .272 .018 .792- . 000 .768- .000 .850- .000 Conf. check 1. 817 .188 .073 .582- .000 .430- .000 .683- .000 Def. 1. 364 .315 .007 .179- .011 .167- .008 . 193-.017 Exo. 1. 323 .217 .046 .733- . 000 .797- .000 .976- .000 Other-rep. • 980 .183 .078 .578- . 061 .752- .044* . 485-.085 S e l f -rep. • 565 .103 .212 .585- .000 .494- . 000 . 688-.000 Echo • 271 . 109 .200 .619- .004 . 620-. 006 .599- .005 * movement from a non-significant to a s i g n i f i c a n t value at p < .05 The c o e f f i c i e n t of normality and associated l e v e l of 101 s i g n i f i c a n c e (via StatView 512+) indicate that d e f i n i t i o n s and comprehension checks are p o s i t i v e l y skewed to a severe degree (p_ < .025), although a l l of the remaining variables are p o s i t i v e l y skewed to l e s s e r degrees, including several which show what might be termed moderate skewness. Among the possible remedies for skewness of t h i s range, Tabachnick and F i d e l l (1983) recommend a square root transformation for moderate p o s i t i v e skewness and a logarithmic transformation for severe p o s i t i v e skewness, noting, at the same time the value of r e t a i n i n g the data i n i t s o r i g i n a l form. Both transformed and untransformed d i s t r i b u t i o n s for the l i s t e d variables were then used i n a s e r i e s of ANOVAs. S i g n i f i c a n t e f f e c t s for group (df = 1) and task (df = 4) were compared i n order to e s t a b l i s h the extent to which transforming the d i s t r i b u t i o n s f o r r e p a i r and reference affected s i g n i f i c a n c e l e v e l s . (A table of F values f o r Table 3 w i l l be found i n Appendix G.) In 13 of 14 cases transformation al t e r e d neither the s i g n i f i c a n c e of the r e s u l t nor the o v e r a l l r e l a t i o n s h i p of mean scores across tasks for a given v a r i a b l e . In one case, other-repetition, the logarithmic transformation moved the p r o b a b i l i t y value for the e f f e c t s for task from s l i g h t l y above a t y p i c a l c r i t i c a l value of .05 (square root and untransformed) to s l i g h t l y below .05. Since transformation appeared to have v i r t u a l l y no e f f e c t on the s i g n i f i c a n c e of the r e s u l t s , a l l analyses of variance were conducted and evaluated using the o r i g i n a l (untransformed) d i s t r i b u t i o n s . 102 Summary T h i s c h a p t e r has f o c u s e d on the d e s i g n o f t h e s t u d y and the p r o c e d u r e s used t o c a r r y i t o u t . Bases f o r t h e f a c t o r i a l d e s i g n were o u t l i n e d and a s t r a t e g y f o r i m p l e m e n t a t i o n o f the d e s i g n t h r o u g h a n a l y s i s o f v a r i a n c e was l i n k e d t o a p r o c e d u r e f o r h y p o t h e s i s t e s t i n g . A c o m b i n a t i o n o f c o n s e r v a t i v e , l a r g e l y e x p l o r a t o r y p r o c e d u r e s was o u t l i n e d f o r the i n i t i a l phase o f t h e r e s e a r c h s t r a t e g y , w h i l e more a g g r e s s i v e , d i r e c t i o n a l h y p o t h e s i s t e s t i n g was o u t l i n e d f o r t h e s e c o n d a r y phase o f t h e s t u d y . Beyond the a n a l y s i s o f v a r i a n c e , a q u a l i t a t i v e e v a l u a t i o n o f s e l e c t e d t r a n s c r i p t s was d e s c r i b e d f o r t h e l a s t s t a g e o f t h e r e s e a r c h s t r a t e g y . A t t h e c e n t e r o f t h e r e s e a r c h methodology i s t h e c o m p a r i s o n o f f r e q u e n c i e s f o r c o n v e r s a t i o n a l r e p a i r o c c u r r i n g i n f i v e t a s k s i t u a t i o n s under t aken by two k i n d s o f t e a c h e r - l e d d y a d s : mixed , c o n s i s t i n g o f a n a t i v e E n g l i s h -s p e a k i n g t e a c h e r and a Japanese l e a r n e r o f E n g l i s h , and homogeneous. c o n s i s t i n g o f a Japanese t e a c h e r o f E n g l i s h and a Japanese l e a r n e r o f E n g l i s h . The r e s u l t s o f t h e a n a l y s i s o f v a r i a n c e a r e p r e s e n t e d i n C h a p t e r 4. These f i n d i n g s w i l l then be i n t e r p r e t e d and r e l a t e d t o t h e r e s e a r c h d e s i g n i n C h a p t e r 5. 103 CHAPTER 4: QUANTITATIVE ANALYSIS OF THE DATA This chapter begins with a summary of r e s u l t s f o r the nine d e s c r i p t i v e features which were not calculated on the basis of i n t r a - t e x t u a l codings. Although the summary i s not part of the hypothesis-testing strategy, i t does follow the pattern established i n the research design f o r comparison of tasks and groups. Next, r e s u l t s from the analysis of variance are presented i n two sections, the f i r s t reporting r e s u l t s f o r comparisons between i n d i v i d u a l tasks and the second f o r comparisons between tasks combined or selected on the basis of the d i s t i n c t i o n between expository and e x p e r i e n t i a l a c t i v i t y . Each of these sections i s further divided into r e s u l t s for the two sets of variables coded within the transcribed t e x t s — r e f e r e n c e and repair. The Descriptive Features of Talk by Group and Task Tables 4, 5, 6, 7, 8, and 9 report the means and standard deviations f o r three categories of d e s c r i p t i v e features by group and task: words ( t o t a l words, words per minute (WPM), unique words, type-token r a t i o (TTR)), utterances ( t o t a l utterances, words per utterance (WPU) and turns ( t o t a l turns, words per turn (WPT), utterances per turn (UPT)). Since each table summarizes r e s u l t s f o r one h a l f of the design (one group at a time), the number of dyads i n each i s 6, i . e . , n = 6. A l l decimal f r a c t i o n s have 104 been rounded to the nearest hundredth. Table 4 shows a f a i r l y symmetrical r e l a t i o n s h i p between the t o t a l word and WPM averages attained by the teacher and by the learner. In general, as teachers spoke more, learners spoke le s s , the greatest gap of t h i s sort occurring during the face-to-face Lego task (LEG2); t o t a l word count, f o r example, averaged about 712 words for teachers and 152 words for learners. Learners were most l i k e l y , however, to speak during ordinary conversation (DIS) whereas teachers were more l i k e l y to reduce the rate of t h e i r own speech during t h i s task to accommodate the learners. Talk during DIS i n the mixed-group dyads was more evenly balanced i n terms of t o t a l words, WPM, unique words and TTR (respectively, T = 445.33/L = 370.17, T = 74.22/L = 61.70, T = 178.67/L = 147.00, T = .40/L =. 41) than during other tasks. Highest average TTRs for learners were achieved during e x p e r i e n t i a l a c t i v i t y , during the two tasks which entailed observing and manipulating objects, COM2 (.47) and LEG2 (.42). The teachers' highest average TTR was also achieved during DIS (.4 0); the lowest average for the e n t i r e mixed group (.24), however, was attained by teachers during LEG2. Whereas DIS and LEG2 were generally responsible for producing the highest and lowest means for the word-based measures (depending on the role of the p a r t i c i p a n t ) , at the dyadic l e v e l COMl, the most " l e c t u r e - l i k e " task, was 105 T a b l e 4 Means and S t a n d a r d D e v i a t i o n s f o r Word-based Measures o f C o n v e r s a t i o n a l A c t i v i t y by M i x e d - g r o u p Task Task C0M1 COM2 DIS LEG1 LEG2 V a r i a b l e Mean SD Mean SD Mean SD Mean SD Mean SD T o t a l words dyad t e a c h e r l e a r n e r WPM 864.33 71.30 670.67 77.11 192.50 93.30 822.33 127.95 665.83 129.33 156.50 45.79 815.50 89.64 445 .33* 108.35 3 7 0 . 1 7 * * 96.71 746 .67* 121.52 523.83 87.49 222.83 39.98 8 6 4 . 6 7 * * 103.60 7 1 2 . 1 7 * * 82.04 152 .50* 39.13 dyad t e a c h e r l e a r n e r Un ique words 144.03 11.86 111.92 13.03 32 . 09 15.55 137.06 21.32 110.97 21.55 26. 09 7.63 135.92 14.94 74 .22 * 18.06 6 1 . 7 0 * * 16.12 124 .45* 20.25 87.31 14.58 37.14 6. 66 144 .08 * * 17.26 1 1 8 . 7 0 * * 13.67 25 .42* 6.52 dyad t e a c h e r l e a r n e r TTR 247.50 16.95 179.67 22.40 67.67 24.70 251.17 17.58 1 8 0 . 3 3 * * 20.25 71. 00 10.00 3 2 5 . 8 3 * * 30.02 178.67 34.14 1 4 7 . 0 0 * * 26.58 219 .50* 13.17 140 .33* 15.48 79.17 3 .19 235.00 23.37 165.33 11.98 61 .33 * 6.35 dyad t e a c h e r l e a r n e r .29 . 03 .27 . 02 .38 . 08 .31 .04 . 28 . 04 . 4 7 * * . 10 . 4 0 * * .01 . 4 0 * * . 05 .41 . 05 .30 . 04 .27 . 04 . 36* .06 . 2 7 * .02 .24* . 02 .42 .08 * * h i g h e s t mean among f i v e t a s k s ; * lowest mean among f i v e t a s k s . 106 v i r t u a l l y i d e n t i c a l to LEG2 i n terms of t o t a l number of words and WPM generated. Moreover, comparisons of means at the dyadic l e v e l across the f i v e tasks indicate a narrow range for each of the word-based measures, except for TTR. When teacher or learner constituents of the dyads are considered i n d i v i d u a l l y , however, much larger gaps between the means become evident. Summary figures for the homogeneous group (Table 5) are s i m i l a r to those for the mixed group. Although homogeneous dyads, taken as a group, used a larger number of words and WPM during COMl than during the four other tasks, the symmetrical, teacher-learner production of words i s most c l e a r l y indicated i n DIS and LEG2: Compared to t h e i r performance on other tasks, teachers used t h e i r lowest average number of words (about 412) and WPM (about 69) i n ordinary discussion while learners produced t h e i r highest averages during discussion (about 347 and 58, r e s p e c t i v e l y ) . Just the reverse was true for word production ( t o t a l words and WPM) during the face-to-face Lego task; teachers spoke the most and learners the least during LEG2. The central p o s i t i o n of DIS and LEG2 as f a r as word-based measures i s concerned was also indicated by both the dyads 1 and the i n d i v i d u a l participants* use of unique words, and by TTR. On the average, dyads, teachers and learners used the largest number of unique words—and achieved t h e i r largest TTRs—during undirected discussion. With the 107 T a b l e 5 Means and S t a n d a r d D e v i a t i o n s f o r Word-based Measures o f C o n v e r s a t i o n a l A c t i v i t y by Homogeneous-group Task Task V a r i a b l e COMl Mean SD COM2 DIS LEG1 LEG2 Mean SD Mean SD Mean SD Mean SD T o t a l words dyad t e a c h e r l e a r n e r WPM dyad t e a c h e r l e a r n e r Un ique words dyad t e a c h e r l e a r n e r TTR dyad t e a c h e r l e a r n e r 7 9 4 . 5 0 * * 714.50 64.68 75.03 545.50 132.58 249.00 84.08 540.83 79.55 173.67 66.13 132 .42 * * 119.08 10.78 12.50 90.92 22 .10 41.50 14.01 204.50 24.80 154.83 40. 68 92.33 17.89 .26 .03 .29 . 04 . 39* . 08 90.14 13.26 29.61 12.23 191.00 24.02 150.83 13.09 72.50 25.70 .27 .04 .30 .05 .44 . 10 758.67 704 .33* 51.87 50.24 411 .67* 534.83 148.82 89.22 3 4 6 . 6 7 * * 169.17 109.78 54.70 126.44 8.65 68 .61 * 24.80 5 7 . 7 8 * * 18.30 116 .56* 9.78 89.14 14.87 28.20 9. 12 2 7 5 . 0 0 * * 203.17 35.74 59.37 1 6 8 . 5 0 * * 147.67 62.91 28.74 1 4 8 . 5 0 * * 71.83 38.72 17.49 . 3 6 * * .04 . 4 1 * * . 02 . 44* * .04 .29 . 08 . 28 . 02 .43 . 04 744.67 104.88 6 2 7 . 3 3 * * 111.49 117 .33* 18.42 124.14 17.45 1 0 4 . 5 6 * * 18.58 19 .56* 3 . 07 175 .50* 16.83 145 .33* 21.48 5 6 . 8 3 * 5.19 .24* .02 .24* .03 .49 .09 * * h i g h e s t mean among f i v e t a s k s ; * lowest mean among f i v e t a s k s . 108 exception of learners (whose lowest average TTR i s associated with C0M1), the lowest TTRs for the dyads i n general and teachers i n p a r t i c u l a r are associated with LEG1. These r e s u l t s c l o s e l y p a r a l l e l those for the mixed group. Two further p a r a l l e l s between the two groups are, f i r s t , the p o s i t i o n of C0M1 as a strong generator of words at the dyadic l e v e l , e c l i p s i n g LEG1 i n terms of t o t a l words and WPM and, second, the r e l a t i v e l y narrow range of averages among the f i v e tasks at the dyadic l e v e l as compared with the much greater range for teachers and learners. At the l e v e l of t o t a l words, for example, the range f o r the dyad i s about 90 words (between C0M1 and LEG1), on the average; the teachers' range i s about 216 words, while the learners' i s about 229, between DIS and LEG2. When utterance-based measures for the mixed group are considered (Table 6), a p o l a r i t y between COM1 and LEG2 emerges. With one exception—that of the learner generating the highest number of WPU during DIS—COM1 was the source of the smallest, and LEG2 the largest, number of utterances. The largest gap between these two tasks was found i n the teachers' t a l k (roughly 65 utterances as compared with the learners' gap, about 15 utterances). WPU were at t h e i r highest average l e v e l , on the other hand, during COM1 and at t h e i r lowest during LEG2. For teachers, t h i s lengthening of utterances during i n s t r u c t i o n about use of the computer, as compared with face-to-face construction of Lego, averaged 109 about f o u r words ( 3 . 6 8 ) . A l t h o u g h l e a r n e r s , l i k e t e a c h e r s , u s e d the g r e a t e s t number o f u t t e r a n c e s d u r i n g LEG2, t h e i r T a b l e 6 Means and S t a n d a r d D e v i a t i o n s f o r U t t e r a n c e - b a s e d Measures o f C o n v e r s a t i o n a l A c t i v i t y by M i x e d - g r o u p Task Task COMl COM2 DIS LEG1 LEG2 Mean Mean Mean Mean Mean V a r i a b l e SD SD SD SD SD U t t e r a n c e s dyad 133. 00* 166. 50 133. 67 140. 50 180. 67* 17. 39 24. 18 22. 49 32 . 62 24. 57 t e a c h e r 72 . 33* 95. 50 79. 33 78. 00 136. 83* 9. 87 17. 01 42. 20 14. 64 34. 44 l e a r n e r 63 . 50* 66. 00 68. 17 71. 00 76. 50* r 12 . 99 11. 98 12. 54 13 . 96 10. 05 dyad 7. 00 * * 5. 22 5. 84 5. 49 4. 63* 1. 17 1. 12 1. 03 94 * 76 t e a c h e r 9 . 4 8 * * 7. 15 7. 05 8. 01 5. 80* 1. 23 1. 95 2 . 28 1. 86 1. 03 l e a r n e r 2. 93 2 . 38 5. 4 1 * * 3 . 19 1. 98* • 97 • 62 • 71 • 51 • 38 * * h i g h e s t mean among f i v e t a s k s ; * lowest mean among f i v e t a s k s . l a r g e s t i n c r e a s e i n WPU (3.43) o c c u r r e d when t h e y were f r e e t o d i s c u s s whatever they l i k e d . U t t e r a n c e - b a s e d measures f o r the homogeneous group (Tab le 7) were v e r y s i m i l a r t o t h o s e o b t a i n e d f o r t h e mixed 110 g r o u p . Both t h e p a t t e r n o f extreme average v a l u e s between C0M1 and LEG1, and the l e n g t h e n i n g o f l e a r n e r s ' u t t e r a n c e s between LEG1 and DIS a r e r e f l e c t e d i n the t a b l e s . L e a r n e r s i n t h e homogeneous g r o u p , however, used more u t t e r a n c e s on t h e average d u r i n g b a c k - t o - b a c k Lego c o n s t r u c t i o n t h a n T a b l e 7 Means and S t a n d a r d D e v i a t i o n s f o r U t t e r a n c e - b a s e d Measures o f C o n v e r s a t i o n a l A c t i v i t y by Homogeneous-group Task Task C0M1 COM2 DIS LEG1 LEG2 Mean Mean Mean Mean Mean V a r i a b l e SD SD SD SD SD U t t e r a n c e s WPU dyad 122. 50* 163.00 146. 67 154. 00 187 .83* 21. 14 31. 01 28. 66 12 . 02 24.92 t e a c h e r 63 . 33* 94.83 79. 67 83. 50 122 .50* 11. 24 10.53 23. 28 6. 72 21.30 l e a r n e r 59. 17* 68.17 67. 00 70. 5 0 * * 65.33 r 15. 47 23 .78 9. 90 8. 17 9.81 dyad 6. 72 * * 4.49 5. 30 4. 61 3 .83* 1. 69 .80 • 87 . 59 .54 t e a c h e r 8. 82 * * 5.73 5. 78 6. 45 5 .18* 2 . 63 . 82 1. 80 1. 24 .83 l e a r n e r 4. 19 2 . 61 5. 4 6 * * 2 . 38 1 .82* • 71 .87 2. 43 • 65 . 32 * * h i g h e s t mean among f i v e t a s k s ; * l owes t mean among f i v e t a s k s . d u r i n g f a c e - t o - f a c e c o n s t r u c t i o n . Some d i f f e r e n c e s between the mixed and homogeneous groups on t u r n - b a s e d measures w i l l now be c o n s i d e r e d . T a b l e 8 shows t h a t most t u r n s among the mixed group were t a k e n d u r i n g per formance o f the f a c e - t o - f a c e Lego t a s k ; T a b l e 8 Means and S t a n d a r d D e v i a t i o n s f o r T u r n - b a s e d Measures o f C o n v e r s a t i o n a l A c t i v i t y by M i x e d - g r o u p Task Task COMl COM2 DIS LEG1 LEG2 Mean Mean Mean Mean Mean V a r i a b l e SD SD SD SD SD T o t a l t u r n s dyad 75.83 81.17 7 3 . 3 3 * 82.67 9 1 . 5 0 * * 15.87 15.37 21.06 9.83 29.66 t e a c h e r 39.00 42.00 36 .67 * 40.67 4 8 . 0 0 * * 7.72 8.07 10.63 3.27 14.31 l e a r n e r 36.83 39.17 36 .67 * 40.33 4 3 . 3 3 * * 8.23 7.36 10.48 4.27 15.85 WPT dyad 1 1 . 9 1 * * 10.49 11.90 9 .14 * 10.55 3.35 2.87 3.78 1.48 4.60 t e a c h e r 1 8 . 0 9 * * 16.47 13.47 12 .92* 16.21 5.51 5.15 6.05 2.17 6.37 l e a r n e r 5.26 4.02 1 0 . 3 7 * * 5.55 3 .98* 2.28 1. 03 2.31 1.02 1.74 UPT dyad 1.86 2.00 1.90 1 .68* 2 . 2 8 * * .49 . 16 .40 .28 .83 t e a c h e r 1.93 2.29 1.86 1 .68* 2 . 7 4 * * .53 .26 . 32 .42 .63 l e a r n e r 1.78 1. 69* 1.95 1.76 2 . 0 1 * * .47 .14 . 53 .29 .90 * * h i g h e s t mean among f i v e t a s k s ; * lowest mean among f i v e t a s k s . 112 fewest were taken during the discussion task (Dyad: 91.50 vs. 73.33; Teacher: 48.00 vs. 36.67; Learner: 43.33 vs. 3 6.67). Learners took t h e i r longest turns (WPT) during discussion, however, while teachers took t h e i r s during performance of C0M1. Overall, very small differences were found among the tasks at the dyadic l e v e l . UPT, another measure of turn length, was generally at i t s highest l e v e l during LEG2 (Lego constructed face-to-face) and at i t s lowest l e v e l during LEG1 (Lego constructed back-to-back), although learners were more l i k e l y to make the fewest utterances per turn when undertaking the computer demonstration (COM2). The homogeneous group generally d i f f e r e d from the mixed group on the turn-based measures (Table 9). The c l e a r e s t differences between groups were found i n averages for t o t a l turns and WPT. The homogeneous group, for example, took i t s greatest number of turns during LEG1 (with LEG2 the source of most turns for the mixed group). Although teachers i n both groups demonstrated a common low average frequency for UPT i n the back-to-back Lego task (1.68 for the mixed group and 1.98 f o r the homogeneous group), learners i n the two groups d i f f e r e d as to t h e i r production of utterances. UPT means for the mixed group of learners were highest during COM1. Homogeneous group learners, on the other hand produced t h e i r highest average number of utterances per turn during C0M1. Inspection of the tables reveals one point of 113 s i m i l a r i t y between groups which occurred i n many of the other measures of conversational a c t i v i t y , namely the teachers' preference for t a l k during COMl and the learners' Table 9 Means and Standard Deviations for Turn-based Measures of Conversational A c t i v i t y by Homogeneous-group Task Task COMl COM2 DIS LEG1 LEG2 Mean Mean Mean Mean Mean Variable SD SD SD SD SD Total turns WPT UPT dyad 54.17* 72. 67 77.83 87.33** 71.17 24.98 13.92 23.76 14.42 10.81 teacher 27.83* 37.00S 39.83 43.00** 37.33 12.38 7.40 12 .48 6.78 5.68 learner 26.33* 35.33 37.83 43.67** 33 .83 12.61 6.74 11.27 8.36 5.34 dyad 19.15 10.23 11.12** 8.34* 10.56 13.08 2.77 5.52 1.56 1.43 teacher 25.40** 15.49 10.75* 12.67 16.99 18.98 5.68 2.90 2.75 3.10 learner 11.35 4.83 11.38** 4.07 3.55' i 5.94 1.21 9.27 1.72 .90 dyad 2.71 2.29 2.13 1.81** 2.69 1.22 .49 1.01 .31 .51 teacher 2.73 2 . 68 2.31 1.98* 3.32 1.37 .78 1.28 .33 .60 learner 2.69** 1.92 1.94 1.66* 2.00 1.27 .44 .78 .34 .53 highest mean among f i v e tasks; * lowest mean among f i v e tasks. 114 preference f o r talk, during DIS, Tables 8 and 9 show these r e l a t i v e l y high l e v e l s of t a l k ( r e l a t i v e to t a l k i n other tasks) i n the form of words uttered per turn. Learners i n both groups produced t h e i r largest average number of WPT on the discussion task, DIS, whereas teachers i n both groups produced t h e i r s during C0M1. The scope and s i g n i f i c a n c e of task and group differences w i l l now be summarized. Unlike the preceding d e s c r i p t i v e section, the following sections report r e s u l t s obtained from analyses of variance on values for r e p a i r and reference and evaluate the extent to which the r e s u l t s support the hypotheses. No d e t a i l e d examination of the means and standard deviations w i l l be presented here (see Appendix K f o r the complete l i s t i n g by dyad, teacher and learner). Although occasional reference w i l l be made to teacher-learner differences, the fundamental unit of analysis w i l l be the dyad. The Analysis of Variance: Repair by Group and Individual Task (H1-H2) HI: The frequency of repair i n dyads does not vary s i g n i f i c a n t l y by group membership or type of task performed. Among the 12 repair exponents (REs) tested, 10 showed s i g n i f i c a n t e f f e c t s for task: c l a r i f i c a t i o n request, comprehension check, confirmation check, d e f i n i t i o n , display question, echo, l e x i c a l uncertainty, r e f e r e n t i a l question, self-expansion and other-expansion. Two (other-expansion and other-repetition) showed no e f f e c t s for group or task. 115 No s i g n i f i c a n t interactions between group and task were noted. The n u l l hypothesis for task i s thus rejected for 10 of 12 REs at E < .025 and accepted for the remaining two at p_ > .025. The n u l l hypothesis i s accepted f o r the group factor at p_ > .025. The tables and figures which follow describe these r e s u l t s i n greater d e t a i l . C l a r i f i c a t i o n Request; CCLAR A trend towards a s i g n i f i c a n t e f f e c t f o r group was noted for CCLAR (F = 5.172, df = 1, p_ = .046, > .025). At Table 10 E f f e c t s of Group Membership and Task on C l a r i f i c a t i o n Requests Source of Sum of Mean E p s i l . v a r i a t i o n df squares square F p_ corr. Group 1 40.017 40.017 5.172 .046 Error 10 77.367 7.737 Task 4 101.900 25.475 5.646 .001* G x T 4 15.233 3 .808 .844 .506 Error 40 180.467 4.512 * p_ < . 025 . the same time cl e a r differences between tasks were indicated (F = 5.646, df = 4, p_ = .001), with s i g n i f i c a n t differences (p_ < .05) noted s p e c i f i c a l l y between discussion on the one hand and the two Lego tasks on the other, and between the f i r s t computer task (COM1) and back-to-back Lego. The Lego tasks comprise the central source for c l a r i f i c a t i o n requests among the f i v e tasks. Figure 4 depicts these differences i n 116 7 6 Range 5 of Means 4 for DV: CCLAR 3 2 1 0 Task: * = Mixed o = Homogeneous COMl COM2 DIS LEG1 l _ LEG2 Figure 4. Plot of means by task for CCLAR. Comprehension Check: CCOM Table 11 indicates a s i g n i f i c a n t e f f e c t f o r task Table 11 E f f e c t s of Group Membership and Task on Comprehension Checks Source of v a r i a t i o n df Sum of squares Mean square E p s i l , p_ corr. Group Error Task G x T Error 1 10 4 4 40 .26 70.87 184.43 17.567 638.867 .26 7.09 46.11 4.39 15.972 .038 .850 11.191 .001* 1.066 .386 .57 * p < .025. (E = .001) during comprehension checks, a rep a i r type 117 employed almost exclusively by the teachers i n both groups. Contrasts between a l l task means showed that CCOM was used s i g n i f i c a n t l y more frequently (p_ < .05) during the back-to-back Lego task than i t was during any of the four other tasks, as i s i l l u s t r a t e d i n Figure 5. 6 5 Range 4 of Means 3 for DV: CCOM 2 1 0 -1 * = Mixed o = Homogeneous I. .1. .1. Task: C0M1 COM2 DIS Figure 5. Plot of means by task for CCOM. LEG1 l _ LEG 2 Confirmation Check: CCON Although the magnitude of differences between tasks i s v i r t u a l l y the same for both comprehension checks and confirmation requests (respectively, F = 11.191, df = 4, p_ = .001 and F = 11.680, df = 4, p = .001), the major source of variance within CCON (Table 12) i s the face-to-face (rather than back-to-back) Lego task. LEG2 i s s i g n i f i c a n t l y d i f f e r e n t from a l l other tasks at p_ < .05 and from a l l tasks except LEG1 at p < .01. I t i s also i n t e r e s t i n g to note that group means were v i r t u a l l y 118 i n t e r e s t i n g to note that group means were v i r t u a l l y Table 12 E f f e c t of Group Membership and Task on Confirmation Checks Source of Sum of Mean E p s i l . v a r i a t i o n df squares square F Group 1 6. 017 6. 017 .177 . 683 Error 10 339.633 33.963 Task 4 746.167 186.542 11.680 .001* G x T 4 32.567 8.142 .510 .729 Error 40 638.867 15.972 * p_ < . 025. i d e n t i c a l for each of the two Lego tasks f o r both CCOM and CCON. Figure 6 shows the r e l a t i v e l y high l e v e l of confirmation checks during performance of LEG2. Task: COMl COM2 DIS LEG1 LEG2 Figure 6. Plot of means by task for CCON. 119 Table 13 shows that within-group ( i . e . , task) differences f o r d e f i n i t i o n also reached s i g n i f i c a n t l e v e l s (F = 3.434, df = 4, p_ = .017), l a r g e l y through the Table 13 Source of Sum of Mean E p s i l v a r i a t i o n df squares square F p_ corr. Group 1 2.817 2.817 1.899 . 198 Error 10 14.833 1.483 Task 4 15.567 3.892 3.434 .017* G x T 4 3.100 .775 .684 .607 Error 40 45.333 1.133 .61 * p_ < .025. contribution of teachers i n both groups during the des c r i p t i o n of the computer's s t r i n g search function (COMl). As Figure 7, following, indicates, the differences between Task: COMl COM2 DIS LEG1 LEG2 Figure 7. Plot of means by task for DDEF. t h i s task and, respectively, discussion and back-to-back 120 t h i s task and, respectively, discussion and back-to-back Lego were r e l a t i v e l y large and, following pairwise comparisons of means, s i g n i f i c a n t (p < .05). Although d e f i n i t i o n s were employed infrequently, they occurred s i g n i f i c a n t l y more frequently during C0M1 than during either DIS or LEG2. Display Question: DDQ Tasks are c l e a r l y distinguished i n Table 14 by the frequency of display questions (F = 23.220, df = 4, p_ = .001). Overall, DDQs were most l i k e l y to occur during COM2, the face-to-face demonstration of the computer's Table 14 Ef f e c t s of Group Membership and Task on Display Questions Source of Sum of Mean E p s i l . v a r i a t i o n df squares square F P Group 1 11.267 11.267 1 .130 . 313 Error 10 99.733 9.973 Task 4 834.233 208.558 23 .220 .001* G x T 4 44.900 11.225 1 .250 . 306 Error 40 359.267 8.982 * E < .025. s t r i n g search function. As Figure 8 shows, COM2 served as the p i v o t a l source of s i g n i f i c a n t differences between tasks, although C0M1 was found to be a secondary source of differences between tasks. These two tasks were s i g n i f i c a n t l y d i f f e r e n t from the others (p < .01) i n terms of the frequency of display questions. 121 E c h o : EECH A s i g n i f i c a n t e f f e c t f o r t a s k was a l s o found when examin ing the f r e q u e n c y o f echoes (F = 4 .455 , d f = 4, p_ = . 0 0 5 ) . F o r b o t h groups EECH was l e a s t f r e q u e n t d u r i n g T a b l e 15 E f f e c t s o f Group Membership and Task on Echoes Source o f Sum o f Mean v a r i a t i o n d f s q u a r e s square F E Group 1 5.400 5.400 .295 .599 E r r o r 10 183.000 18.300 Task 4 224.667 56.167 4.455 . 005* G x T 4 30.600 7. 650 . 607 . 660 E r r o r 40 504.333 12.608 = p_ < . 025. f r e e d i s c u s s i o n and most f r e q u e n t d u r i n g the computer t a s k s . 122 However, as Figure 9 indicates i t was the difference between COM1 and COM2 on the one hand and DIS on the other which served as the major sources of variance ( s i g n i f i c a n t at £ < Lexical Uncertainty: LLEX Lex i c a l uncertainty was most frequently expressed by pa r t i c i p a n t s i n both groups during free discussion; i t was l e a s t l i k e l y to be expressed during the computer demonstration (COM2) and face-to-face Lego (LEG2) tasks. A l l task differences taken together (Table 16) indicated a s i g n i f i c a n t main e f f e c t for task (F = 8.016, df = 4, p_ = .001) . 123 Table 16 Ef f e c t s of Group Membership and Task on Lexical Uncertainty Source of Sum of Mean E p s i l . v a r i a t i o n df squares square F p_ corr. Group 1 2.817 2.817 .910 .363 Error 10 30.967 3.097 Task 4 71.767 17.942 8.016 .001* G X T 4 9.100 2.275 1.016 .411 Error 40 89.533 2.238 .51 * p_ < .025. Post hoc pairwise comparisons between means (Figure 10) indicated that the e f f e c t was at t r i b u t a b l e to the contrasts between DIS and the two tasks which enta i l e d face-to-face manipulation of objects: LEG2 and COM2. These differences Task: COMl COM2 DIS LEG1 LEG2 Figure 10. Plot of means by task f o r LLEX. were s i g n i f i c a n t at p < .01. S i g n i f i c a n t differences at 124 p_ < .01 were also found between C0M1 and LEG2. Other-expansion: OOEXP Other-repetition: OOREP Neither task nor group e f f e c t s were noted for OOEXP (Table 17) and OOREP (Table 18) following the c r i t e r i o n established for s i g n i f i c a n c e (p_ < .025), although r e s u l t s f o r OOREP suggested a possible trend towards s i g n i f i c a n c e f o r task (F = 2.208, df = 4, p_ = .085). Since no s i g n i f i c a n t differences were found for these RE's, no s p e c i f i c examination w i l l be made of the means. Table 17 Ef f e c t s of Group Membership and Task on Other-expansion Source of Sum of Mean E p s i l . v a r i a t i o n df squares square F p_ corr. Group 1 18.150 18.150 .767 .402 Error 10 236.700 23.670 Task 4 44.400 11.100 1.210 .322 G x T 4 27.600 6.900 .752 .562 Error 40 366.800 9.170 .70 Table 18 Ef f e c t s of Group Membership and Task on Other-repetiton Source of Sum of Mean E p s i l , v a r i a t i o n df squares square F p_ corr. Group 1 20.41 20.417 .525 .485 Error 10 388.83 Task 4 100.90 25.225 2.208 .085 G X T 4 28.500 7.125 .624 .648 Error 40 457.000 11.425 .66 125 Referential Question: RRQ Table 19 shows an e f f e c t for task (F = 11.920, df = 4 , p_ = .001) which i s r e f l e c t e d i n the difference between the r e l a t i v e l y small number of RRQ's during computer i n s t r u c t i o n (COMl, COM2) and the r e l a t i v e l y high number Table 19 Ef f e c t s of Group Membership and Task on Referential Questions Source of v a r i a t i o n df Sum of squares Mean square F E Group 1 12.150 12.150 .594 .459 Error 10 204.700 20.470 Task 4 1347.567 336.892 11.920 . 001* G x T 4 60.767 15.192 .538 .709 Error 40 1130.467 28.262 * E < .025. during non-instructional tasks (DIS, LEG1, LEG2). Comparison among the means indicates that the s i g n i f i c a n t differences (p_ < .05) l i e between each of the computer tasks and each of DIS, LEG2 and LEG1 (the sequence here i n d i c a t i n g s i g n i f i c a n t differences between means from highest to lowest). A s i m i l a r pattern of s i g n i f i c a n t differences appeared at p_ < .01, although s p e c i f i c comparisons between COM2 and LEG1/LEG2 were not s i g n i f i c a n t at t h i s l e v e l . A graph of the means (Figure 11) reveals that the pattern for RRQ was a v i r t u a l mirror image of the pattern for DDQ ( i . e . , a s i g n i f i c a n t concentration of r e f e r e n t i a l questions during 126 the non-teaching tasks). Self-expansion: SSEXP The rate of self-expansion varied s i g n i f i c a n t l y by task (F = 3.167, df = 4, p_ = .024) but not by group, as Table 20 E f f e c t s of Group Membership and Task on Self-expansion Source of Sum of Mean E p s i l . v a r i a t i o n df squares square F p_ corr. Group 1 570. 417 570.417 2. 004 .187 Error 10 2846. 433 284.643 Task 4 599. 567 149.892 3. 167 .024* G x T 4 406. 167 101.542 2. 146 .093 Error 40 1893. 067 47.327 * p_ < . 025. 127 Table 20 shows. A trend towards the i n t e r a c t i o n of group and task which did not reach s i g n i f i c a n c e was also observed. Closer inspection of the means (Figure 12) indicates a Task: COMl COM2 DIS LEG1 LEG2 Figure 12. Plot of means by task f o r SSEXP. f a i r l y low frequency for SSEXP by task except for LEG2, and, i n p a r t i c u l a r , the response of the mixed group to LEG2. Although the performance of the mixed dyads produced a s i g n i f i c a n t difference (p < .05) between DIS and LEG2, no other s i g n i f i c a n t differences between the means for s e l f -r e p e t i t i o n by task were found. S e l f - r e p e t i t i o n : SSREP Table 21 shows that the occurrence of s e l f - r e p e t i t i o n d i f f e r e d s i g n i f i c a n t l y by task (F = 7.829, df = 4, p = .001) but not by group or the i n t e r a c t i o n of group and task. Unlike the s i g n i f i c a n t l y frequent use of SSEXP during face-to-face Lego, however, i t was the r e l a t i v e l y infrequent use of SSREP during ordinary discussion which served as the 128 source of the e f f e c t for task. A l l dyads employed Table 21 Ef f e c t s of Group Membership and Task on S e l f - r e p e t i t i o n Source of Sura of Mean E p s i l . v a r i a t i o n df squares square F P. Group 1 26.667 26.777 .171 .688 Error 10 1562.867 156.287 Task 4 1964.667 491.167 7.829 . 001* G x T 4 488.667 122.167 1.947 . 121 Error 40 2509.467 62.737 * p_ < .025. s e l f - r e p e t i t i o n s i g n i f i c a n t l y less frequently during DIS than during LEG2, followed by COM1 and then COM2 (p_ < .01). Figure 13 i l l u s t r a t e s t h i s r e l a t i o n s h i p among the tasks. Range of Means for DV: SSREP 40 30 20 10 * = Mixed o = Homogeneous Task: COM1 COM2 DIS LEG1 Figure 13. Plot of means by task for SSREP. LEG2 Table 22 summarizes a l l s i g n i f i c a n t e f f e c t s ( i . e . , those associated with task) by RE and d e t a i l s the major 129 sources of variance within each e f f e c t . Sources are l i s t e d by the magnitude of difference between means, from large to small. Thus, for each RE, the largest difference between two means i s l i s t e d f i r s t , followed by the next largest, and so on u n t i l the smallest s i g n i f i c a n t difference i s encountered. As the table indicates, d e f i n i t i o n s , display questions and echoes were associated with the dayds• performance of the two teaching tasks, COMl and COM2. The main sources for production of c l a r i f i c a t i o n requests, comprehension checks, confirmation checks, expressions of l e x i c a l uncertainty, r e f e r e n t i a l questions and s e l f - and other-expansions, on the other hand, were the non-teaching tasks, DIS, LEG1 and LEG2. DIS was the primary se t t i n g for production of r e f e r e n t i a l Table 22 S i g n i f i c a n t Repair Exponents and Sources of Variance for A l l Tasks Main sources of variance Repair exponent F r a t i o E E < .05 p_ < .01 C l a r i f i c a t i o n request 5 . 646 . 001 LEG1 > DIS LEG2 > DIS LEG1 > DIS Comprehension 11.191 check . 001 LEG1 > DIS LEG1 > LEG2 LEG1 > COM2 LEG1 > COMl LEG1 > DIS LEG1 > LEG2 LEG1 > COM2 LEG1 > COMl table continues 130 T a b l e 22 ( c o n t ' d . ) S i g n i f i c a n t R e p a i r Exponents and S o u r c e s o f V a r i a n c e f o r A l l T a s k s Main s o u r c e s o f v a r i a n c e R e p a i r exponent F r a t i o E E < . 05 E < .01 C o n f i r m a t i o n 11 . 680 . 001 LEG2 > DIS LEG2 > DIS check LEG2 > COM2 LEG2 > COM2 LEG2 > COM1 LEG2 > COM1 LEG2 > LEG1 — D e f i n i t i o n 3 .434 . 017 C0M1 > LEG2 — C0M1 > DIS — D i s p l a y 23 .220 . 001 COM2 > LEG1 COM2 > LEG1 q u e s t i o n COM2 > LEG2 COM2 > LEG2 COM2 > DIS COM2 > DIS COM2 > COM1 COM2 > COM1 COM1 > LEG1 COM1 > LEG1 Echo 4 .455 . 005 COM2 > DIS COM2 > DIS C0M1 > DIS COM1 > DIS L e x i c a l 8 . 016 . 001 DIS : > : LEG2 DIS > LEG2 u n c e r t a i n t y DIS : > i COM2 DIS > COM2 COM1 > LEG2 COM1 > LEG2 C0M1 > COM2 COM1 > COM2 R e f e r e n t i a l 11 .920 '. 001 DIS : COM1 DIS > COM1 q u e s t i o n DIS : > COM2 DIS > COM2 LEG2 > COM1 LEG2 > COM1 LEG 2 > COM2 — LEG1 > COM1 LEG1 > COM1 LEG1 > COM2 — S e l f - 3 . 167 . 024 LEG 2 > DIS — e x p a n s i o n S e l f - 7 .829 . 001 LEG2 > DIS LEG2 > DIS r e p e t i t i o n COM1 > DIS COM1 > DIS COM2 > DIS COM2 > DIS q u e s t i o n s and e x p r e s s i o n s o f l e x i c a l u n c e r t a i n t y ( f o l l o w e d by LEG1 and L E G 2 ) , whereas LEG1 and LEG2, i n p a r t i c u l a r , 131 were primary sources for c l a r i f i c a t i o n requests, comprehension checks, confirmation checks, and s e l f - and other-repetition. These r e s u l t s apply to the task performances of both groups of dyads: Repair of t a l k during the tasks was not s i g n i f i c a n t l y related to the f a c t that h a l f of the groups were led by native speakers of English and h a l f by Japanese. H2: The frequency of reference i n dyads does not vary s i g n i f i c a n t l y by either group membership or task performed. Both types of reference, anaphora and exophora, showed s i g n i f i c a n t e f f e c t s for task. As i n the case of repair, no e f f e c t s for group and none for the i n t e r a c t i o n of group and task were s i g n i f i c a n t . With respect to task, the n u l l hypothesis i s thus rejected for reference at p_ < .025 and accepted f o r group (at p > .025). Tabular r e s u l t s for each ANOVA, one for anaphora (Table 23) and one for exohpora (Table 24) as the dependent variables, are presented below. A singl e graphic representation of task means for both of these variables however, i s also presented (Figure 14) and i s intended to i l l u s t r a t e the complementary a l l o c a t i o n of both forms of reference during each of the tasks. Anaphora: AANA Although a trend towards group differences i n the use of anaphora i s indicated i n Table 23, only the e f f e c t for task was s i g n i f i c a n t (F = 3.178, df = 4, p_ = .023, < .025). COMl was the source of l e a s t anaphora, LEG1 the 132 T a b l e 2 3 E f f e c t s o f Group Membership and Task on A n a p h o r i c R e f e r e n c e Source o f Sum o f Mean E p s i l . v a r i a t i o n d f squares square Z E c o r r . Group 1 735.000 735.000 4.102 .070 E r r o r 10 1791.933 179.193 Task 4 2529.167 632.292 3 .178 .023* G x T 4 518.833 129.708 . 654 . 628 E r r o r 40 7936.400 198.410 . 69 * p_ < .025 . g r e a t e s t ; d i f f e r e n c e s between the means f o r t h e s e t a s k s were s i g n i f i c a n t (p_ < . 0 5 ) . Exophora : EEXO Whereas AANA showed a t r e n d towards group d i f f e r e n c e s , EEXO was employed by dyads i n bo th groups t o a n e a r l y i d e n t i c a l degree (F = .001) d u r i n g the f i v e t a s k s . T a b l e 24 E f f e c t s o f Group Membership and Task on E x o p h o r i c R e f e r e n c e Source o f Sum o f Mean E p s i l . v a r i a t i o n d f squares square F p_ c o r r . Group 1 . 067 . 067 .001 .976 E r r o r 10 710 .533 71.053 Task 4 23539 .667 5884.917 48.108 .001* G x T 4 393 . 600 98.400 .804 .530 E r r o r 40 4893 . 133 122.328 .51 * E < . 025 . S i g n i f i c a n t d i f f e r e n c e s were found between t a s k s , however (F - 48 .108 , d f = 4, p = . 0 0 1 ) . Moreover , extreme v a l u e s f o r 133 exophora were achieved i n the two Lego tasks, the highest mean f o r EEXO during face-to-face Lego and the lowest during back-to-back Lego. These extremes served as poles for the other tasks: S i g n i f i c a n t differences between means for LEG2 and a l l other tasks, and between LEG1 and the other tasks, were noted (p < .01). Relationships between means for AANA and EEXO are presented i n graphic form together i n Figure 14. Figure 14 indicates that as one form of reference increased i n frequency, the other tended to decrease. (But see LEG2, a major source of both forms of reference, for the exception to t h i s trend.) Thus, anaphora i s at a high l e v e l r e l a t i v e to exophora during COMl, the lecture about the computer Range of Means for DV: AANA, EEXO 60 50 40 30 20 10 = Mixed/AANA = Homogeneous/AANA = Mixed/EEXO = Homogeneous/EEXO .1. Task: COMl COM2 DIS LEG1 LEG2 Figure 14. Plot of means by group and task for AANA and EEXO 134 search function. These positions change during COM2, the computer demonstration, so that anaphora becomes slightly-l e s s frequent, on the average during COM2. LEG2, the only other task besides COM2 which enta i l e d seeing and pointing out objects, boosted the frequency of both exophora and anaphora. Exophora was generated s i g n i f i c a n t l y more frequently during face-to-face Lego than during any of the other tasks. Anaphora, moreover, occurred at roughly the same l e v e l s i n both LEG2 and LEG1, although homogeneous dyads working through the back-to-back Lego task used anaphoric reference at a somewhat greater frequency than t h e i r mixed-dyad counterparts, thus making LEGl a r e l a t i v e l y more frequent source of anaphoric reference than COM2 Table 25 summarizes a l l s i g n i f i c a n t e f f e c t s by category Table 25 S i g n i f i c a n t Reference Categories and Sources of Variance f o r A l l Tasks Main sources of variance Reference category F_ r a t i o p_ p. < .05 p_ < .01 Anaphora Exophora 3.187 48.108 .023 . 001 LEGl > COM2 LEG2 LEG2 LEG2 LEG2 COM2 COM2 COM2 LEGl DIS COMl COM2 LEGl DIS COMl LEG2 > LEGl LEG2 > DIS LEG2 > COMl LEG2 > COM2 COM2 > LEGl COM2 > DIS COM2 > COMl 135 of reference and l i s t s the major sources of variance within each e f f e c t . As i n the case of repair, these sources are l i s t e d by the magnitude of difference between means, from large to small. The Analysis of Variance: Repair and Reference i n Groups During Combined and Selected Tasks The following r e s u l t s are based on two treatments of the data designed to compare tasks on the basis of t h e i r e x p e r i e n t i a l and expository c h a r a c t e r i s t i c s . The f i r s t treatment combined and then averaged LEG2 and COM2, the two tasks which allowed part i c i p a n t s to observe and point out objects i n t h e i r immediate environment, into a new independent v a r i a b l e designated, for purposes of analysis, as EXPER1—the experi e n t i a l task category. EXPER1 was compared with C0M1, the only task category which did not allow p a r t i c i p a n t s to observe and point out objects i n the conversational environment and which required the teaching of functional procedures abstractly, that i s , without d i r e c t experience with the object of i n s t r u c t i o n . C0M1 was re-designated as EXPOS1—the expository t a s k — f o r purposes of the analysis. This treatment of the data w i l l be referred to below as EXPER-EXP0S1. DIS, the free discussion task, and LEG1, a non-teaching information-exchange task, were dropped from the analysis. Although LEG1 could have been used as the expository stem i n the EXPER-EXPOS treatment of the data outlined above, i t was 136 a r e l a t i v e l y u n i m p r e s s i v e s o u r c e o f a n a p h o r i c r e f e r e n c e (Tab le 25) and remained c o n c e p t u a l l y l e s s i n t e r e s t i n g as a s o u r c e o f e x p o s i t o r y a c t i v i t y than COMl, a t a s k l a r g e l y d i r e c t e d towards the t e a c h i n g o f c o g n i t i v e knowledge. N e v e r t h e l e s s , L E G l was used i n a supp lementary e x p l o r a t i o n o f t h e EXPER-EXPOS d i m e n s i o n , the r e s u l t s o f which a r e summarized i n Appendix H. The second t rea tment o f the d a t a s i m p l y s e l e c t e d LEG2 and COMl ( r e - d e s i g n a t e d , r e s p e c t i v e l y , as EXPER2 and EXPOS2) f o r c o m p a r i s o n w i t h o u t t h e i n f l u e n c e o f the r e m a i n i n g t h r e e t a s k v a r i a b l e s . T h i s t rea tment o f the d a t a w i l l be r e f e r r e d t o as EXPER-EXPOS2 One o b j e c t o f combin ing and s e l e c t i n g t a s k s i n t h e s e ways was t o examine how e x p e r i e n t i a l and e x p o s i t o r y t a s k s d i f f e r when a t e a c h i n g o r i e n t a t i o n i s i n t e g r a t e d i n t o the e x p e r i e n t i a l t a s k , th rough c r e a t i o n o f EXPER1, and when i t i s n o t , t h r o u g h use o f EXPER2. The r e s u l t s o f ANOVAs employ ing each s e t o f t a s k c a t e g o r i e s (EXPER-EXPOS1 and EXPER-EXPOS2) a r e r e p o r t e d t o g e t h e r inasmuch as t h e y were d e s i g n e d t o t e s t the same d i r e c t i o n a l h y p o t h e s e s . R e p a i r D u r i n g E x p e r i e n t i a l and E x p o s i t o r y A c t i v i t y (H3) H3: R e p a i r o c c u r s s i g n i f i c a n t l y more f r e q u e n t l y d u r i n g t a s k s which emphasize e x p e r i e n t i a l a c t i v i t y than d u r i n g t a s k s which emphasize e x p o s i t o r y a c t i v i t y . In g e n e r a l , each s e t o f ANOVAs produced v e r y s i m i l a r r e s u l t s . S i g n i f i c a n t d i f f e r e n c e s (p < .05) between e x p e r i e n t i a l and e x p o s i t o r y t a s k s were s h a r e d by EXPER-137 EXP0S1 and EXPER-EXPOS2 on four of the twelve repair exponents (confirmation checks, d e f i n i t i o n s , l e x i c a l uncertainty and r e f e r e n t i a l questions). EXPER-EXP0S1 and EXPER-EXPOS2 also shared non-significant task differences fo r the remaining eight REs ( c l a r i f i c a t i o n requests, comprehension checks, display questions, echoes, other-expansions, other-repetitions, self-expansions and s e l f r e p e t i t i o n s ) . H3 thus finds only l i m i t e d support, i . e . , with respect to four of the twelve repair exponents. The only cases of s i g n i f i c a n t between-group differences ( c l a r i f i c a t i o n requests) and s i g n i f i c a n t group/task i n t e r a c t i o n ( s e l f - r e p e t i t i o n ) found thus f a r i n the analysis of variance also occurred i n both EXPER-EXP0S1 and EXPER-EXP0S2. Although t h i s f inding does not s p e c i f i c a l l y r e l a t e to H3, i t does suggest some i n t e r e s t i n g differences between the mixed and homogeneous groups which w i l l be discussed i n the next chapter. Results for the two sets of ANOVAs are summarized i n d e t a i l i n Table 26. The ANOVA tables which document these r e s u l t s are contained i n Appendix I (for EXPER-EXP0S1) and Appendix J (for EXPER-EXPOS2). Among the four REs which showed s i g n i f i c a n t e f f e c t s for task, two were more frequent during the ex p e r i e n t i a l tasks (confirmation checks and r e f e r e n t i a l questions) and two were more frequent during the expository tasks ( d e f i n i t i o n s and expressions of l e x i c a l uncertainty). Thus, when the point 138 T a b l e 26 S i g n i f i c a n t R e p a i r C a t e g o r i e s and Sources o f V a r i a n c e f o r E x p e r i e n t i a l and E x p o s i t o r y Tasks U s i n g COMl as t h e E x p o s i t o r y Stem ANOVA Main s o u r c e s o f EXPER-EXP0S1 EXPER-EXPOS2 v a r i a n c e (p <) R e p a i r exponent F r a t i o E z r a t i o E .05 .01 C l a r . 11. , 125 . 008 M > H M > H r e q u e s t 7, .895 . 019 M > H -— C o n f i r m . 8. ,360 .016 EL > EY check 14. .426 . 004 E L > EY EL > EY D e f i n i t i o n 5. ,768 . 033 EY > E L — 14. .426 . 004 E L > EY EL > EY L e x i c a l 14. ,989 .003 EY > E L EY > EL u n c e r t . 12, .707 . 005 EY > E L EY > EL R e f e r . 28. ,324 . 003 E L > EY E L > EY q u e s t i o n 44, .471 .001 EL > EY E L > EY S e l f - 6. ,805 . 026 a t EXPOS1 r e p . H > M a t EXPER1 M > H 8, .282 . 016 a t EXPOS2 • H > M a t EXPER2 • M > H N o t e . G r o u p s : M = M i x e d , H = Homogeneous; T a s k s : EL = E x p e r i e n t i a l , EY = E x p o s i t o r y ; d f = 1 i n a l l c a s e s o f compar ison was e n t i r e l y r e s t r i c t e d t o e x p e r i e n t i a l and e x p o s i t o r y t a s k s , the number o f r e p a i r exponents which d i f f e r e d on the b a s i s o f t a s k was c o n s i d e r a b l y narrowed from 139 the i n i t i a l l i s t of 12. This was j u s t as much the case when the e x p e r i e n t i a l task was oriented towards teaching as i t was when the e x p e r i e n t i a l task contained no such ori e n t a t i o n , that i s , when i t was e n t i r e l y concerned with constructing a Lego toy. The i n t e r a c t i o n between group and task ( i l l u s t r a t e d i n Figure 15) i s somewhat more complex than the simple main e f f e c t s otherwise noted i n Table 26. Range of Means for DV: SSREP 37 34 31 28 25 * = Mixed/EXPER-EXPOSl o = Homogeneous/EXPER-EXPOSl + = Mixed/EXPER-EXP0S2 . = Homogeneous/EXPER-EXP0S2 Task: Ex p e r i e n t i a l Expository Figure 15. Plot of means by group and e x p e r i e n t i a l -expository tasks for SSREP One c h a r a c t e r i s t i c of t h i s p l o t i s the s i m i l a r i t y of group patterns for s e l f - r e p e t i t i o n during the e x p e r i e n t i a l tasks. Regardless of whether the e x p e r i e n t i a l task was oriented towards teaching or not, dyads led by a Japanese teacher produced s i g n i f i c a n t l y more s e l f - r e p e t i t i o n than dyads le d by a native speaker of English. Conversely, dyads led by native English speakers were responsible f o r 140 producing s i g n i f i c a n t l y more s e l f - r e p e t i t i o n during C0M1, the teaching-oriented expository task, than dyads led by the Japanese speakers of English. Anaphoric Reference During; E x p e r i e n t i a l and Expository A c t i v i t y (H4a) H4a: Anaphoric reference occurs s i g n i f i c a n t l y more frequently during tasks which emphasize expository a c t i v i t y than during tasks which emphasize ex p e r i e n t i a l a c t i v i t y . As noted i n the previous section, r e s u l t s from the analysis of anaphora rejected the assumption that there were no differences between task frequencies. Given a five-task within-subjects factor, AANA was found to occur more frequently during LEG1 than COM2. A question remains, however, whether t h i s finding also extends to the contrast between independent variables s p e c i f i c a l l y constructed (EXPER-EXP0S1 and EXPER-EXP0S2) to r e f l e c t the d i s t i n c t i o n between e x p e r i e n t i a l and expository tasks. These two kinds of task have been associated with the conceptual framework fo r the study and can be examined further with the use of reference i n order to d i s t i n g u i s h between them. Comparison of EXPER1 and EXP0S1 revealed no s i g n i f i c a n t differences i n the dyads 1 use of anaphoric reference, although a trend for between-group differences was noted (F = 4.167, df = 1, p_ = .068). When means for EXPER2 (the uncombined, non-training-oriented Lego task) and EXPOS2 are compared (Table 27), however, a s i g n i f i c a n t e f f e c t f o r task i s observed: F = 5.558, df = 1, p_ < .040. (This difference 141 T a b l e 27 E f f e c t s o f Group Membership and S e l e c t e d T a s k s on A n a p h o r i c R e f e r e n c e : C o n t r a s t i n g : LEG2 w i t h COMl (EXPER-EXPOS2) S o u r c e o f Sum o f Mean E p s i l . v a r i a t i o n d f s q u a r e s square F E Group 1 345.042 345.042 1.294 .282 E r r o r 10 2667.083 266.708 Task 1 392.042 392.042 5.558 .040* G x T 1 77.042 77.042 1.092 .321 E r r o r 10 705.417 70.542 * p < .05 r a t h e r c l o s e l y resembles the one r e p o r t e d above (see T a b l e 25) f o r t h e d i f f e r e n c e i n use o f a n a p h o r i c r e f e r e n c e d u r i n g t h e b a c k - t o - b a c k Lego and computer d e m o n s t r a t i o n t a s k s ) . 46 43 Range 40 o f Means 37 f o r DV: AANA 34 31 28 25 * = M ixed /EXPER-EXPOSl o = Homogeneous/EXPER-EXPOSl +<— — — ,__ :r_=-*+ Mixed/EXPER-EXPOS2 Homogeneous/EXPER-EXPOS 2 .1. T a s k : E x p e r i e n t i a l E x p o s i t o r y F i g u r e 16. P l o t o f means by group and e x p e r i e n t i a l -e x p o s i t o r y t a s k s f o r AANA. 142 Anaphora was found to be even more frequent during the e x p e r i e n t i a l task than during the expository task, as i s indicated i n Figure 16. The r e s u l t s are consistent with those obtained during comparisons among the f u l l range of f i v e tasks. H4a thus finds no support from the data reported here. Exophoric Reference During E x p e r i e n t i a l and Expository A c t i v i t y (H4b) H4b: Exophoric reference occurs s i g n i f i c a n t l y more frequently during tasks which emphasize e x p e r i e n t i a l a c t i v i t y than during tasks which emphasize expository a c t i v i t y . R e c a l l i n g r e s u l t s for exophora reported during the examination of H2, s i g n i f i c a n t differences were found between LEG2 and COM2 on the one hand and the remaining three tasks on the other. Indeed, although anaphora was frequent during the back-to-back Lego task (LEG1), exophora was s i g n i f i c a n t l y more frequent than anaphora during t h i s task. As Table 28 and 29 indicate, s i m i l a r r e s u l t s were obtained f o r both sets of comparisons between the e x p e r i e n t i a l and expository tasks: For EXPER-EXP0S1, F = 49.258, df = 1, p = .001; for EXPER-EXPOS2, F = 57.794, df = 1, E ~ •001. 143 Table 28 E f f e c t s of Group Membership and Selected Tasks on Exophoric Reference: Contrasting LEG2 and COM2 with C0M1 (EXPER-EXPOS1) Source of Sum of Mean E p s i l . v a r i a t i o n df squares square F p_ corr. Group 1 38.760 38.760 * 353 .566 Error 10 1099.104 109.910 Task 1 6224.260 6224.260 49. 258 .001* G x T 1 213.010 213.010 1. 686 .223 Error 10 1263.604 126.360 1.00 * p_ < .05 Table 29 E f f e c t s of Group Membership and Selected Tasks on Exophoric Reference: Contrasting LEG2 with C0M1 (EXPER-EXPOS2) Source of Sum of Mean E p s i l . v a r i a t i o n df squares square F p_ corr. Group 1 5. 042 5. 042 . 021 .889 Error 10 2451. 417 245. 142 Task 1 14162. 042 345. 042 57. 794 . 001 G x T 1 345. 042 345. 042 1. 408 .263 Error 10 2450. 417 245. 042 * p_ < .05 These r e s u l t s uniformly support Hypothesis 4b. The consistency with which the groups produced exophora under both forms of exp e r i e n t i a l task i s indicated i n Figure 17. 144 60 Range 50 of Means 40 for DV: EEXO 30 20 10 0 Mixed/EXPER-EXPOSl Homogeneous/EXPER-EXPOS1 Mixed/EXPER-EXPOS2 Homogeneous/EXPER-EXPOS2 Task: Exp e r i e n t i a l Expository Figure 17. Plot of means by group and E x p e r i e n t i a l -Expository tasks f o r EEXO. Summary Analysis of variance was used to t e s t f i v e hypotheses about the e f f e c t of task or group on repair and reference. At the i n i t i a l l e v e l of analysis, i n which a l l tasks were compared on each of the 12 repair exponents and two forms of reference, s i g n i f i c a n t differences between tasks were found f o r 10 of the 12 REs (except other-expansion and other - r e p e t i t i o n ) , and for reference. The r e s u l t s also indicated an a l l o c a t i o n of RE to task: c l a r i f i c a t i o n requests, comprehension checks, confirmation checks, expressions of l e x i c a l uncertainty, r e f e r e n t i a l questions, self-expansions and s e l f - r e p e t i t i o n s to the non-teaching tasks (DIS, LEGl, LEG2), d e f i n i t i o n s , display questions and echoes to the teaching tasks (COMl, COM2). Both forms of reference, anaphora and exophora, occurred with greatest 145 frequency i n the non-teaching tasks. At the same time, no s i g n i f i c a n t differences between the mixed and homogeneous groups were found. The n u l l hypotheses, that there are no s i g n i f i c a n t differences between tasks i n the frequency of re p a i r (HI) and reference (H2), were thus rejected. The second l e v e l of analysis focused on the e f f e c t of two theorized task constructs, e x p e r i e n t i a l tasks and expository tasks, on the production of repair and reference. Regardless of whether the e x p e r i e n t i a l task was oriented to teaching (LEG2 + COM2) or not (LEG2 alone), only four of the i n i t i a l l i s t of REs d i f f e r e d s i g n i f i c a n t l y on the basis of task: confirmation checks and r e f e r e n t i a l questions (most c l o s e l y associated with the e x p e r i e n t i a l tasks), d e f i n i t i o n s and i n d i c a t i o n s of l e x i c a l uncertainty ( t y p i c a l l y associated with the expository task). S i g n i f i c a n t differences between the native speaker-led and Japanese-led groups were found only for c l a r i f i c a t i o n requests, whereas the frequency of s e l f - r e p e t i t i o n was found to d i f f e r on the basis of both group membership and task. The homogeneous dyads produced the greatest amount of s e l f -r e p e t i t i o n during the e x p e r i e n t i a l tasks and the l e a s t during the expository task; j u s t the reverse obtained for the mixed group. On the basis of these r e s u l t s one of three d i r e c t i o n a l hypotheses (H3) gained p a r t i a l support, that i s , a l e v e l of support l i m i t e d to four of the twelve REs. One of the d i r e c t i o n a l hypotheses (4a) was not supported by the data; anaphora was, i n fact, used at a 146 higher mean frequency during one of the exp e r i e n t i a l tasks (contrary to the hypothesis) and at about the same mean frequency during the other. The f i n a l hypothesis (H4b), however, was strongly supported by r e s u l t s obtained for both groups i n both sets of tasks. Exophora was generated s i g n i f i c a n t l y more often during e x p e r i e n t i a l tasks than during the expository task. Discussion and int e r p r e t a t i o n of these r e s u l t s follows i n Chapter 5. 147 CHAPTER 5: DISCUSSION OF THE ANALYSIS OF VARIANCE T h i s c h a p t e r c o n c l u d e s t h e a n a l y s i s o f v a r i a n c e th rough a d i s c u s s i o n o f r e s u l t s p r e s e n t e d i n t h e p r e v i o u s c h a p t e r . The d i s c u s s i o n i s o r g a n i z e d i n t o s e c t i o n s keyed t o t h e major r e s e a r c h q u e s t i o n s . A l t h o u g h t h e s e q u e s t i o n s have been "answered" i n the l i m i t e d c o n t e x t o f the h y p o t h e s e s , t h a t i s , w i t h r e f e r e n c e t o a p a r t i c u l a r s e t o f d a t a and r e s e a r c h c o n d i t i o n s , t h i s d i s c u s s i o n i s i n t e n d e d t o i n t e r p r e t and s u g g e s t e x p l a n a t i o n s f o r the r e s u l t s w i t h r e f e r e n c e t o the b r o a d e r background o f r e s e a r c h i n SLA and second language e d u c a t i o n . I t s h o u l d be s t r e s s e d t h a t t h e s e i n t e r p r e t a t i o n s a r e b a s e d on the q u a n t i t a t i v e t r e a t m e n t o f r e p a i r and r e f e r e n c e and n o t , f o r the moment, on an e x a m i n a t i o n o f r e p a i r and r e f e r e n c e as q u a l i t i e s o f spoken t e x t s . What has been l e a r n e d so f a r ? The Use o f R e p a i r by Group and Task The G e n e r a l Lack o f Group D i f f e r e n c e s One o f t h e c l e a r e s t f i n d i n g s i n t h i s s t u d y i s t h e b r o a d l a c k o f d i f f e r e n c e s between groups i n the use o f r e p a i r : The f o r e i g n - and J a p a n e s e - l e d dyads were l a r g e l y i n d i s t i n g u i s h a b l e i n t h e i r approaches t o overcoming o r a v o i d i n g m i s u n d e r s t a n d i n g . T h i s i s not the same as s a y i n g t h a t t h e groups a c t e d i d e n t i c a l l y o r t h a t t h e r e a r e no d i f f e r e n c e s between them. Indeed, p a r t o f t h e d i s c u s s i o n wh ich f o l l o w s examines the few a r e a s which do seem t o 148 d i s t i n g u i s h the groups. However, the general p i c t u r e of dyadic i n t e r a c t i o n i s that c u l t u r a l and language background do not constrain conversational repair. S p e c i f i c Group Differences Some apparent group differences within s p e c i f i c tasks, however, are worth examining. As noted previously, a trend fo r o v e r a l l group differences was found for c l a r i f i c a t i o n requests (F = 5.172, df = 1, p = .0462). Although the groups produced i d e n t i c a l frequencies f o r t h i s form of repa i r during discussion, and very s i m i l a r frequencies during the back-to-back Lego task, members of the mixed dyads requested c l a r i f i c a t i o n much more frequently during the two tasks, taken together, which entailed s e t t i n g objectives and teaching use of the laptop computer (COMl and COM2), and to a le s s e r degree during the face-to-face problem-solving task (LEG2). More pr e c i s e l y , i t was l a r g e l y the learners i n the dyads led by native speakers of English who requested c l a r i f i c a t i o n , and they tended to do so when they f e l t themselves the subjects of i n s t r u c t i o n . This rather l i m i t e d area of group difference suggests that Japanese learners may f e e l somewhat freer to ask t h e i r American and B r i t i s h partners f o r help during i n s t r u c t i o n a l s e s s i o n s — s p e c i f i c a l l y , f o r t h e i r partners to c l a r i f y something not immediately understood—than to ask f o r s i m i l a r help from t h e i r Japanese partners. One addit i o n a l trend towards group differences was associated with self-expansions during the face-to-face Lego 149 task. Members of mixed-group dyads were evidently much more w i l l i n g to expand t h e i r own conversational utterances i n order to c l a r i f y meaning than were members of the Japanese-led dyads (see Figure 12, above). Others have also noted the impact of e t h n i c a l l y mixed dyads on conversational performance (e.g., Long & Porter, 1985; Varonis & Gass, 1985). Because the task i n which t h i s kind of difference occurred allowed both part i c i p a n t s to view the construction, i t may be surmised that teachers i n both groups were able to monitor the learner's progress accurately, but that the native teachers were readier than t h e i r Japanese counterparts to head o f f p o t e n t i a l misunderstanding by increasing the redundancy of t h e i r utterances. To put the issue of group differences i n perspective, however, i t should be pointed out that group differences did not prove s t a t i s t i c a l l y s i g n i f i c a n t o v e r a l l , as predicted by the null-hypothesis, and that the more important differences are based i n the performance of tasks without regard to group membership. As noted i n the following section, these differences were almost wholly task-based and t y p i c a l l y emerged as repairs allocated to e i t h e r teaching or non-teaching s i t u a t i o n s . The Relationship of Repair and Task: A l l o c a t i o n of Repair to Teaching and Non-teaching tasks The a l l o c a t i o n of repairs to task (see Table 22) i s the second major fi n d i n g of the study. Repairs are not equally 150 useful during performance of various tasks, nor are they equally appropriate means of making the discourse more comprehensible. I t i s not enough simply to say that repair happens more often when members are cooperating to solve a motivating problem. Although t h i s i s generally true, i t i s more accurate to say, f i r s t , that p a r t i c u l a r repairs work best i n p a r t i c u l a r environments and, second, that groups with d i f f e r e n t c u l t u r a l and language backgrounds appear competent at knowing which repairs serve which tasks. This kind of competence i s one of the more appealing features of studies i n IT which have noted that NNSs can negotiate and rep a i r e f f e c t i v e l y i n t h e i r target language (see, es p e c i a l l y , Porter, 1983; Doughty & Pica, 1986). In addition to r e j e c t i o n of the n u l l hypothesis f o r task, then, i t i s necessary to specify task/repair r e l a t i o n s h i p s : d e f i n i t i o n s , echoes and display questions to teaching tasks (COMl and COM2); c l a r i f i c a t i o n requests, comprehension checks, confirmation checks, expressions of l e x i c a l uncertainty, r e f e r e n t i a l questions, self-expansion and s e l f - r e p e t i t i o n to non-teaching tasks (DIS, LEGl and LEG2). Some explanation for t h i s a l l o c a t i o n of tasks i s possible by p o s i t i n g what might be c a l l e d "rules of t a l k " fo r each of the task groupings. Ord i n a r i l y , teaching s i t u a t i o n s require teachers to organize and sequence i n s t r u c t i o n a l content which they already possess as experts r e l a t i v e to the learner. The teacher i s expected to 151 demonstrate a degree of pr e c i s i o n i n conveying the material, or at l e a s t to show sp e c i a l concern f o r the comprehensibility of the material with respect to the learner. Learners are expected to be attentive to what the teacher presents and to demonstrate t h i s attentiveness i n various ways, although, e s p e c i a l l y i n Japan, learners are not normally responsible for i n i t i a t i n g communication on the comprehensibility of the i n s t r u c t i o n a l presentation. Competent performance on non-teaching tasks i s somewhat more dependent on pa r t i c i p a n t s ' requesting and o f f e r i n g evidence of comprehensibility than on planning or attentiveness per se. Problem-solving tasks, i n p a r t i c u l a r , revolve around mutual exchange of information and frequent checks to determine the present status of the problem with respect to a solut i o n . Planning and execution of t a c t i c a l moves occur within the task i t s e l f and depend on the qu a l i t y of feedback p a r t i c i p a n t s generate for each other from moment to moment. In general, members' ri g h t s to manipulate the discourse are d i s t r i b u t e d r e l a t i v e l y evenly (as Pica, 1987 has also noted) although they are not necessarily exercised when openings f o r such manipulation become av a i l a b l e . Examples of A l l o c a t i o n to Teaching; Tasks: D e f i n i t i o n and Echo. Given these characterizations, the a l l o c a t i o n of d e f i n i t i o n s and echoes to the teaching tasks makes some sense. D e f i n i t i o n happens when speakers are concerned with the nature of an object and the importance of communicating 152 i t to a l i s t e n e r . This i s fundamentally a concern with knowledge of concepts rather than, or, perhaps i n addition to, information of immediate use i n the solu t i o n of a problem. Echoes unambiguously demonstrate learner attentiveness to the stream of teaching t a l k . Echoes appear to have much less to do with evincing comprehension of the teacher than with reassuring the teacher that the l i s t e n e r i s aware of a topic as i t i s conveyed. One of the learner's major r e s p o n s i b i l i t i e s during cooperative problem-solving, on the other hand, i s to volunteer frequent evidence of understanding or to request enough information to provide such evidence. This i s p r e c i s e l y what occurred during the Lego tasks. Ordinary discussion, as a form of s o c i a l discourse, moreover, i s c l e a r l y and frequently punctuated by evidence of attentiveness ( t y p i c a l l y i n the form of i n a r t i c u l a t e "mm's"), although there seems to be no pressing need to expressly and continually re-nominate the speaker's topics—something which learners i n both groups accomplished s i g n i f i c a n t l y more frequently i n COMl and COM2 than i n any of the other tasks. Support for t h i s way of i n t e r p r e t i n g echoic behavior i s found i n Gass and Varonis (1986) who also found that echoing evinces s o l i d a r i t y with an i n t e r l o c u t o r . A l l o c a t i o n of Question Types to Teaching and Non-teaching Tasks: Display and Referential Questions. Interestingly, the " p i v o t a l " forms of re p a i r i n each of the two task categories (repairs with the largest F r a t i o s 153 and smallest alpha levels) are the two basic question t y p e s — d i s p l a y and r e f e r e n t i a l questions. R e c a l l i n g the l i t e r a t u r e reviewed e a r l i e r (in p a r t i c u l a r Long & Sato, 1983; Duff, 1986), display questions are q u i n t e s s e n t i a l l y the technique of teachers and others determined to t e s t knowledge, simp l i f y putatively complex i n s t r u c t i o n a l material and point out s a l i e n t features i n a presumably confusing f i e l d . Although the asking of display questions can q u a l i f y as repair behavior i n nearly any se t t i n g , the high frequency of display questions i n teaching s i t u a t i o n s ( r e l a t i v e to other situations) suggests that they are a fundamental and recognizably appropriate class of behavior fo r anyone engaged i n a teaching task. Re f e r e n t i a l questions are on the other side of the equation. The r e f e r e n t i a l question i s fundamentally part of non-teaching s i t u a t i o n s , p a r t i c u l a r l y those which allow freedom to structure the discourse content (as i n DIS) and make language the servant of the pa r t i c i p a n t s , as i s the case i n tasks requiring the cooperative exchange of information f o r successful completion (LEG1 and LEG2). Participants who know, as teachers, f o r example, that display questions are appropriate forms of behavior when knowledge i s to be taught, know equally well that r e f e r e n t i a l questions are appropriate ways to i n i t i a t e r e p a i r when information i s to be exchanged. I t appears, then, that task i s a kind of frame f o r repai r i n which conversational roles and r e s p o n s i b i l i t i e s 154 can and do s h i f t depending on the task. Pica's (1987) observation about the nature of the task influencing the production of modified i n t e r a c t i o n i s of some explanatory value here. Given the requirement of a more-or-less equal exchange of information, even the normally dominant p o s i t i o n of the teacher can be altered to permit greater negotiation with a learner during task-based t a l k . This p o s i t i o n helps to explain why ordinary discussion, which c l e a r l y does not control for e i t h e r one-way or two-way exchange of information, was most frequently the s e t t i n g f o r expressions of l e x i c a l uncertainty and r e f e r e n t i a l questions, and why teachers i n both groups reduced t h e i r t a l k to accommodate unaccustomedly voluble learners and thus to help bring the r e l a t i v e contribution of both partners into balance. When cooperation was imposed by the demands of the task, as was the case i n both Lego tasks, somewhat d i f f e r e n t r e p a i r s , those involving constant checks of comprehension or c l a r i f i c a t i o n of v i t a l information, for example, predominated. Whereas free discussion permits p a r t i c i p a n t s who have accepted reasonably equal statuses ( i f only temporarily) to pass up unclear material without comment, information-gap tasks undertaken by p a r t i c i p a n t s of equal status require moment-to-moment monitoring and r e p a i r of the discourse i n order to keep i t on track. I t may be r e c a l l e d that s i m i l a r findings were reported i n Pica (1987) who pointed to the the "equalizing nature of an information exchange task" (p. 16). 155 A l l o c a t i o n of repair to tasks can, of course, be elaborated beyond the f a i r l y simple d i v i s i o n between teaching and non-teaching tasks. For example, the r e s u l t s also suggest that one non-teaching task, discussion, i s p a r t i c u l a r l y associated with expressions of l e x i c a l uncertainty and more or less equal use of r e f e r e n t i a l questions by both teachers and learners. There also appears to be a d i s t i n c t i o n between LEG1 and LEG2—between a two-way task and a one-way t a s k — s o that comprehension checks are, given the data gathered i n t h i s study, c h a r a c t e r i s t i c a l l y associated with the two-way task, whereas confirmation checks are c h a r a c t e r i s t i c a l l y associated with the one-way task. This p a r t i c u l a r d i s t i n c t i o n gains some explanatory force from the nature of the tasks: When pa r t i c i p a n t s are working towards a solution, s i t t i n g back-to-back, the p a r t i c i p a n t who supplies d i r e c t i o n s has a spe c i a l stake i n knowing immediately i f the di r e c t i o n s were understood. A confirmation check, on the other hand, i s a l i s t e n e r ' s r e l a t i v e l y confident response (as opposed to a c l a r i f i c a t i o n request) to a d i r e c t i o n offered by a speaker who can, l i k e the l i s t e n e r , unambiguously observe the e f f e c t of the d i r e c t i o n . Whatever the explanatory q u a l i t y of these ways of looking at the r e s u l t s , they remain minor streams i n comparison with the primary d i s t i n c t i o n between tasks which are conducted through teaching and tasks which are c a r r i e d out through negotiation. Considering the p o t e n t i a l 156 applications of t h i s research to second and foreign language i n s t r u c t i o n a l methodology, i t may seem odd to suggest that non-instructional tasks could play a useful r o l e i n the design of i n s t r u c t i o n . However, the r e s u l t s do indicate that teachers' p a r t i c i p a t i o n i n tasks which r e f l e c t normal small group conversational processes produces s i g n i f i c a n t l y d i f f e r e n t forms of repair than tasks which r e f l e c t small group i n s t r u c t i o n a l processes. A related f i n d i n g i s reported i n Pica and Long's (1986) study which compared i n s t r u c t i o n a l and non-instructional discourse. While both conversational and i n s t r u c t i o n a l processes may be useful to learning a second or foreign language, the bulk of evidence reported i n the l i t e r a t u r e (see Pica & Long, 1986) supports the view taken here, namely that the kind of t a l k involved i n teacher-led conversational and problem-solving tasks more frequently produces negotiation and repair believed to be of value i n acquiring an additional language than the kind of t a l k which t y p i f i e s teacher-led i n s t r u c t i o n a l tasks. The Use of Reference by Group and Task Anaphora and exophora were found to d i f f e r by task, contrary to the n u l l hypothesis, but not by group. In these respects, reference and repair share a general l e v e l of s i m i l a r i t y . Beyond t h i s general l e v e l , however, reference has i t s own pattern of association with task. I t should be r e c a l l e d from the r e s u l t s that anaphora and exophora are t y p i c a l l y mixed within the same task. I t i s the r e l a t i v e d i s t r i b u t i o n 157 of these two forms of repair within and between tasks, however, which suggests a d i s t i n c t i o n between tasks emphasizing the existence of shared experience (including perception of the speech situation) and those emphasizing the need to b u i l d shared experience through language i n order to accomplish the goals of the task. The need to create a set of shared experiences and perceptions i s an underlying feature of C0M1, DIS and LEG1, so i t was not s u r p r i s i n g to f i n d that anaphora, a form of endophoric reference which helps to l i n k p r i o r with current elements of the discourse, was a great deal more frequent than exophora. Tasks i n which shared perception predominated (COM2 and LEG2) showed a rough equivalence of frequencies f o r anaphora and exophora (see Figure 14, above). Thus, while anaphora appears to be a ubiquitous feature of task-based t a l k i n general, the r e l a t i v e frequencies of anaphora and exophora change i n response to the experiential l e v e l of a given task. Exophora appears to serve as the exper i e n t i a l barometer i n spoken texts, i n d i c a t i n g how extensively p a r t i c i p a n t s are able to presume on each other's understanding of the speech s i t u a t i o n . This characterization of reference i n task i s quite d i f f e r e n t from the one presented here previously with respect to r e p a i r — i . e . , teaching/non-teaching—but does accord with the way Gaies (1983), Gass and Varonis (1985a) and Wagner (1983) have discussed t h e i r studies of task-based discourse. The point of departure with previous studies, 158 however, i s the focus on how textual cohesion i s achieved i n tasks which emphasize various l e v e l s of shared experience and not s p e c i f i c a l l y on how meanings are negotiated. Given t h i s focus, i t may be possible to add some empirical refinement to the speculations about the r o l e of "experience" reported i n the l i t e r a t u r e . P a r t i c i p a n t s ' shared experience of a culture, of knowledge of the world and of the p a r t i c u l a r conversational s e t t i n g i n which they f i n d themselves eventually f i n d expression i n textual reference. While i t i s possible, of course, to argue that sharing of assumptions increases negotiation i n general (see, i n p a r t i c u l a r , Gass & Varonis, 1985a) i t i s perhaps more accurate to say that shared experience simply increases reference to the here-and-now (through exophora). At a c e r t a i n point p a r t i c i p a n t s ' common experience renders negotiation superfluous, so i t i s not u n i v e r s a l l y the case that t a l k w i l l be negotiated when inte r l o c u t o r s have almost everything i n common. I t i s possible to observe t h i s more complex r e l a t i o n s h i p among task, negotiation and experience when the tasks to be studied range beyond the one-way/two-way d i s t i n c t i o n , as i s the case here, when i t was found, for example, that more exophoric reference (but l e s s d e f i n i t i o n ) occurred during COMl than during either the one-way or two-way problem-solving tasks (LEG2 and LEGl, r e s p e c t i v e l y ) . S p e c i f i c A l l o c a t i o n s of Reference to Task The balance between anaphora and exophora during task-based discourse appears very much influenced by the kind of 159 task i n which p a r t i c i p a n t s f i n d themselves (Table 25). Anaphora occurred s i g n i f i c a n t l y more frequently during LEG1 (a two-way information exchange task which l i m i t e d the a b i l i t y of p a r t i c i p a n t s to share perception of the task materials) than during COM2 (the face-to-face demonstration of the computer's s t r i n g search function). Exophora occurred s i g n i f i c a n t l y more frequently during the two tasks which allowed p a r t i c i p a n t s to observe, examine and manipulate the objects of t h e i r discussion (COM2 and LEG2) than during a l l of the remaining tasks, including ordinary discussion. These r e s u l t s suggest the emergence of an ex p e r i e n t i a l factor which depends i n large part on the sharing of experience and perception between the two p a r t i c i p a n t s . Without such sharing, without the t a n g i b i l i t y and d i r e c t experience of task processes, tasks become much more dependent on the cohesive function of anaphora to make them i n t e l l i g i b l e . With t a n g i b i l i t y and d i r e c t experience b u i l t into task processes, p a r t i c i p a n t s use exophora to point out the e s s e n t i a l features of the task environment; negotiation of the t a l k i n some of these tasks may well contribute to t h e i r comprehensibility, although negotiation i s apparently not the only resource p a r t i c i p a n t s employ to improve the comprehensibility of t h e i r t a l k . Thus, an addit i o n a l major finding of t h i s study relates to understanding the f i v e tasks i n terms of textual q u a l i t i e s and the r e l a t i o n of these to an educational 160 perspective on language use. Based on r e s u l t s reported e a r l i e r , i t i s now possible to think of tasks as having e x p e r i e n t i a l or expository c h a r a c t e r i s t i c s based on the r e l a t i v e contributions of anaphora and exophora to the spoken texts. The empirical dimensions of t h i s perspective have only been suggested, however, and then only with respect to two exponents (anaphora and exophora) of a theory of reference. The remainder of t h i s chapter w i l l be devoted f i r s t to an examination of repair and then to a further examination of reference occurring i n the l i m i t e d group of tasks found to be most c l o s e l y associated with exophoric reference (LEG2 and COM2) i n comparison with COMl, the l e c t u r e - l i k e task which best embodies expository behavior. Repair During Combined and Selected Tasks: Towards a Framework for Complementary Task Structures When tasks are combined or selected to r e f l e c t the reference-based concept of e x p e r i e n t i a l and expository tasks, the number of s i g n i f i c a n t r e s u l t s dwindles and one r e s u l t i n p a r t i c u l a r , the r e s u l t f o r c l a r i f i c a t i o n requests, shows s i g n i f i c a n t group differences. I t should be r e c a l l e d that when the f i v e tasks were treated as repeated measures, the mixed and homogeneous groups tended to d i f f e r on the basis of c l a r i f i c a t i o n requests, a trend which reached s i g n i f i c a n c e when c l a r i f i c a t i o n requests occurring i n the expository task were compared with those occurring i n both forms of the e x p e r i e n t i a l task ( i . e . , EXP0S-EXPER1 and EXP0S-EXPER2). Since c l a r i f i c a t i o n requests are l a r g e l y 161 based on the contributions of Japanese learners i n both groups, the s i g n i f i c a n t l y larger number of such requests i n the mixed group suggests greater conversational a c t i v i t y among learners when a native speaker i s a v a i l a b l e f o r feedback. One possible way of looking at t h i s e f f e c t f o r group i s to consider the learner's perception of the native speaker as more open to a c l a r i f i c a t i o n request than a Japanese teacher would be—more w i l l i n g to respond u s e f u l l y or informatively to the request. Since t h i s perception cannot be measured d i r e c t l y , i t should be h e l p f u l to examine the t r a n s c r i p t s i n order to compare the actual responses of the teachers. Q u a l i t a t i v e differences i n the responsiveness of teachers from the two groups can be inspected d i r e c t l y and reported. For the time being i t i s i n t e r e s t i n g to note that when the quantitative analysis i s focused on differences between the e x p e r i e n t i a l and expository tasks, a si n g l e , s i g n i f i c a n t group difference emerges. Beyond t h i s difference between the groups, the experiential-expository contrast also severely l i m i t s the scope of s i g n i f i c a n t differences while, at the same time, repeating the rather c l e a r a l l o c a t i o n of r e p a i r to a bifurcated task structure. The smaller number of s i g n i f i c a n t repair exponents, the appearance of a group difference for c l a r i f i c a t i o n requests and the i n t e r a c t i o n between group and task f o r s e l f -r e p e t i t i o n , have f a i r l y straightforward explanations. The experiential-expository contrasts were made through the 162 elimination of other tasks. Because of the nearly i d e n t i c a l r e s u l t s obtained f o r both EXPER-EXP0S1 and EXPER-EXPOS2, i t appears that the tasks eliminated from the analyses of variance ( f i r s t DIS and LEG1, and then DIS, LEG1 and COM2) were the main sources of variance for comprehension checks, display questions, echoes and self-expansions—none of which attained s i g n i f i c a n t group or task differences. Put i n somewhat d i f f e r e n t terms, DIS, LEG1 and COM2 served to suppress group differences f o r c l a r i f i c a t i o n requests and in t e r a c t i o n between group and task for s e l f - r e p e t i t i o n . Once these tasks were removed from the analysis, the non-s i g n i f i c a n t trends for c l a r i f i c a t i o n requests and s e l f -r e p e t i t i o n which developed during the i n i t i a l , f i v e - t a s k repeated measures ANOVA developed into, respectively, a s i g n i f i c a n t group difference and a s i g n i f i c a n t i n t e r a c t i o n . REs which retained t h e i r s i g n i f i c a n t l e v e l s of differe n c e f o r task i n the experiential-expository contrasts (confirmation checks, d e f i n i t i o n s , expressions of l e x i c a l uncertainty and r e f e r e n t i a l questions) can be viewed as s t r a t e g i c behaviors of use i n two d i f f e r e n t , although complementary, task structures. Where Table 22 indicates that confirmation checks and r e f e r e n t i a l questions are all o c a t e d to non-teaching tasks and that d e f i n i t i o n s and indicati o n s of l e x i c a l uncertainty are all o c a t e d to teaching tasks, Table 26 shows that these tasks also f a l l , r e spectively, within the experie n t i a l and expository task structures. These r e s u l t s reinforce the suggestion made 163 e a r l i e r that teaching and expository behavior o v e r l a p — t h a t non-teaching (or conventionally negotiated) and ex p e r i e n t i a l a c t i v i t y overlap, and thus that some forms of rep a i r at l e a s t can be expected to appear i n what may be termed complementary task structures. Although there i s c l e a r l y no i d e n t i t y between teaching and expository tasks on the one hand and non-teaching or exper i e n t i a l tasks on the other, i t does seem possible to draw a rough approximation of how a l i m i t e d number of REs f a l l within the complementary task structures, as follows: TASK PROCESSES Expe r i e n t i a l TASK GOALS Educational + confirm, check + referen. ques. So c i a l + d e f i n i t i o n + l e x i c a l uncert. Expository Figure 18. A l l o c a t i o n of four repair exponents to complementary task structures. Extending the research design somewhat i n order to accommodate the r e s u l t s , Figure 18 emphasizes the goal and process of t a l k i n various contexts. I t thus broadens 164 the notion of task to an e x p l i c i t concern with the reasons people have for t a l k i n g with each other and the ways by which they accomplish t h e i r goals through language. This framework f o r tasks appears more complex that the one-dimensional versions of use i n empirical research, although i t i s intended to r e f l e c t more of the world i n which conversational exchanges a c t u a l l y occur. R e c a l l i n g the discussion of dimensions of task and interlanguage t a l k i n Chapter 2, f a i r l y straightforward comparisons can be made between the two-dimensional framework outlined here and others reported i n the l i t e r a t u r e : Long (1980) and Gass and Varonis (1985a) with regard to the one-way/two-way task d i s t i n c t i o n , Duff (1986) on convergent and divergent tasks, Doughty and Pica (1986) and Pica (1987) on required vs. optional information exchange tasks, and Pica (1987) on information exchange vs. decision-making tasks. Figure 18 should be interpreted broadly. Since the goals and processes of s o c i a l t a l k i n ordinary circumstances can s h i f t over time, i t may be more useful to view the figure as a kind of snapshot taken during one phase of a s o c i a l i n t e r a c t i o n . From t h i s point of view, a task constitutes the frame i n which i n s t r u c t i o n a l planners and researchers may decide to organize t a l k and thus to capture something of the complexity of communicative behavior. The u n f i l l e d quadrants i n the figure suggest additional s l o t s f o r REs under task conditions which were not encountered i n t h i s study. In p r i n c i p l e , for example, a 165 non-teaching task with expository c h a r a c t e r i s t i c s — t h e exchange of anecdotes or report of a t r i p abroad, f o r example—could produce a d i s t i n c t i v e d i s t i l l a t i o n of REs given opportunities for in t e r a c t i o n by the p a r t i c i p a n t s . Implications f o r t h i s way of looking at complementary task structures w i l l be explored again i n Chapter 7. For the present, i t i s perhaps s u f f i c i e n t to indicate that p a r t i c u l a r sets of repair behavior may be expected to occur given the i n t e r s e c t i o n of task factors and that tasks may be evaluated f o r t h e i r o r i e n t a t i o n to goal as well as process. Reference During Combined and Selected Tasks This l a s t section re-examines reference from the viewpoint of an experiential-expository contrast. Before discussing the r e s u l t s f o r reference i n d i v i d u a l l y , some general points w i l l be raised which apply to both forms of reference. F i r s t , i t may be r e c a l l e d that the groups were separated by a large, i f not s i g n i f i c a n t , gap i n the use of both forms of rep a i r during C0M1, the expository task. The homogeneous (Japanese-led) group of dyads employed anaphora demonstrably more often during t h i s task then the mixed (native speaker-led) group (see Figure 16). Just the reverse obtained for exophoric reference: The mixed group preferred exophoric reference over the homogeneous group (Figure 17). These r e s u l t s reinforce the impression reported e a r l i e r that exophora and anaphora are e s s e n t i a l l y complementary i n spoken texts, e s p e c i a l l y so when a range of 166 speech contexts can be examined for the r e l a t i v e d i s t r i b u t i o n s . The re s u l t s also suggest that d i f f e r e n t groups can perceive the same context, or task, i n d i f f e r e n t ways, as indicated by t h e i r preference f o r a given form of reference. The point should not be overstressed, however, p a r t i c u l a r l y since the group differences across both sets of tasks indicated only a trend i n the d i r e c t i o n of group differences. Anaphora During E x p e r i e n t i a l and Expository Tasks Rejection of the n u l l hypothesis that anaphora i s unrelated to task, i n addition to the fin d i n g that anaphora did not occur more frequently during expository tasks than during e x p e r i e n t i a l tasks, require some explanation. Anaphora appears to be a ubiquitous feature of task-based t a l k . Even though s i g n i f i c a n t l y more anaphora was found i n LEGl (nominally, a non-teaching task with expository c h a r a c t e r i s t i c s ) than i n COM2 (a teaching task with e x p e r i e n t i a l c h a r a c t e r i s t i c s ) , anaphora was also found at nearly equivalent l e v e l s during the three remaining tasks. When tasks were s p e c i f i c a l l y l i m i t e d to e x p e r i e n t i a l -expository comparisons, anaphora was found to be even more frequent during the experi e n t i a l task. Thus, anaphora cannot alone be expected to serve as the textual signature f o r teaching and exposition, a conclusion c l e a r l y at odds with the view of anaphoric reference presented i n Chapter 2. This point i s worth re s t a t i n g with reference to the Cummins* conceptual framework (1983). Although anaphora 167 demonstrably serves to l i n k elements i n a text, i t does not thus automatically work as evidence of the cognitive complexity or l e v e l of contextual support i n the text. Other factors must be implicated i n such useful concepts as cognitive complexity and contextual support, and i t i s here that another finding of the study can be restated to incorporate these concepts: I t i s the balance between anaphora and exophora which provides some evidence f o r the degree to which p a r t i c i p a n t s t r e a t t h e i r t a l k as more or les s complex, as o f f e r i n g more or le s s contextual support. Exophora During Ex p e r i e n t i a l and Expository Tasks Exophoric reference i s c l e a r l y the more active factor i n the anaphora-exophora balance. Exophoric reference occurred s i g n i f i c a n t l y more frequently during the e x p e r i e n t i a l tasks than during the expository tasks (in which i t was v i r t u a l l y absent). Regardless of the number of tasks used i n the a n a l y s i s — f i v e or two—and regardless of group membership, the two nominally e x p e r i e n t i a l tasks con s i s t e n t l y influenced the speakers to produce a high and sustained l e v e l of exophoric reference. The p o s i t i o n of exophoric reference i n the exper i e n t i a l tasks c l e a r l y accords with Halliday and Hasan's (1976) view of i t s function i n spoken and written texts and supports Mohan's (1986) view that e x p e r i e n t i a l a c t i v i t y plays a key r o l e i n communicating p r a c t i c a l knowledge. This strong r e l a t i o n s h i p between c e r t a i n kinds of tasks and exophora suggests something of the transparency and 168 immediacy of e x p e r i e n t i a l tasks. Even though e x p e r i e n t i a l tasks are not uniformly the source of frequent negotiation and repair, and allow participants to achieve c l a r i t y of meaning at l e a s t as much through shared perception as negotiation, e x p e r i e n t i a l tasks elevate the comprehensibility of things and events i n the s i t u a t i o n to a p o s i t i o n of t a c t i c a l importance. Comprehensibility can be an outcome of negotiation as well as a common understanding of the s i t u a t i o n i n which parti c i p a n t s f i n d themselves. I t i s the e x p e r i e n t i a l task, i n p a r t i c u l a r the non-teaching e x p e r i e n t i a l task, which seems to have emerged as a kind of compromise between tasks which p r o h i b i t negotiation and those which enforce i t . Summary Several interpretations of the r e s u l t s have been highlighted i n t h i s chapter. F i r s t , the general lack of group differences suggests that Japanese teachers of English are not working under a s t r a t e g i c handicap i n comparison with t h e i r native English speaking counterparts. Teachers i n both groups were equally competent i n recognizing and responding to the learners' c a l l s for assistance. The r e s u l t s further suggest that the a b i l i t y to repair t a l k i n a foreign language may become a conversational resource of learners at a r e l a t i v e l y early stage i n t h e i r a c q u i s i t i o n of the foreign language. Both groups of intermediate l e v e l learners i n t h i s study, for example, used t h e i r i n t e r l o c u t o r s with equal competence to c l a r i f y information, 169 confirm understanding and otherwise ask questions to e l i c i t new information. Second, the major differences i n the frequency of r e p a i r are found i n q u a l i t a t i v e l y d i s t i n c t tasks. D i f f e r e n t kinds of repair are allocated to d i f f e r e n t tasks, the basic d i v i s i o n l y i n g between teaching and non-teaching tasks. Whereas teaching tasks are associated with r e l a t i v e l y passive or formulaic behavior, such as echoing and asking display questions, non-teaching tasks seem to generate a wider range of repairs and negotiation of meaning between the p a r t i c i p a n t s . The negotiation i s characterized by frequent contributions from learners and a sharing of r e s p o n s i b i l i t y for giving and getting information by both members of the teacher-learner group. Third, j u s t as repair i s distinguished by and allocated to task, reference i s d i s t r i b u t e d to task i n a d i s t i n c t i v e way, anaphora to tasks emphasizing the need to b u i l d l i n k s across the discourse through language and exophora to tasks emphasizing the existence of shared perceptions of the speech s i t u a t i o n . Although anaphoric reference was f r e e l y used i n a l l of the tasks, exophoric reference was found l a r g e l y i n tasks i n which pa r t i c i p a n t s shared perception and experience. The notion of an e x p e r i e n t i a l task was developed to encompass these properties of t a n g i b i l i t y and d i r e c t experience i n the task discourse, and contrasted with the concept of an expository task i n which one p a r t i c i p a n t i s mainly responsible for conveying information and keeping 170 p a r t s o f t h e d i s c o u r s e i n t a c t . When t a s k s a r e v iewed i n t h i s c o n t r a s t i v e way, the a l l o c a t i o n o f r e p a i r t o t a s k narrows c o n s i d e r a b l y , but a t the same t ime r e i n f o r c e s t h e i m p r e s s i o n t h a t e x p e r i e n c e and e x p o s i t i o n a re f u n d a m e n t a l l y d i f f e r e n t approaches t o a c c o m p l i s h i n g t a s k s t h r o u g h l a n g u a g e . A f o u r t h , and r e l a t e d , i s s u e c o n c e r n s the i n t e r -r e l a t i o n s h i p o f the two s e t s o f t a s k s t r u c t u r e s , t e a c h i n g and n o n - t e a c h i n g , e x p e r i e n t i a l and e x p o s i t o r y . Based on the o v e r l a p o f r e p a i r w i t h i n t h e s e two s e t s , i t i s p o s s i b l e t o e n v i s i o n a complementary t a s k s t r u c t u r e which v e r y c l e a r l y d i s t i n g u i s h e s a v e r y few r e p a i r b e h a v i o r s . The i n t e r s e c t i o n o f t e a c h i n g and e x p o s i t o r y b e h a v i o r d u r i n g t a s k s p r o v i d e s pe rhaps t h e most c o n s e r v a t i v e d i s c o u r s e env i ronment , so i t i s no t s u r p r i s i n g t h a t d e f i n i t i o n s and e x p r e s s i o n s o f l e x i c a l u n c e r t a i n t y t y p i f y t h i s p o i n t o f i n t e r s e c t i o n . When n o n - t e a c h i n g and e x p e r i e n t i a l t a s k s a re t aken t o g e t h e r , i t i s pe rhaps a l s o not s u r p r i s i n g t h a t c o n f i r m a t i o n checks and r e f e r e n t i a l q u e s t i o n s p r e d o m i n a t e — j u s t the s o r t o f b e h a v i o r which might be expec ted d u r i n g c o o p e r a t i v e d i s c o u r s e o u t s i d e o f most c l a s s r o o m s . The p o t e n t i a l v a l u e o f t h i s way o f l o o k i n g a t t a s k s t r u c t u r e s , however , may l i e i n the f a c t t h a t t e a c h e r s and o t h e r p l a n n e r s have a c h o i c e i n the way they s t r u c t u r e i n s t r u c t i o n , c h o o s i n g groups and t a s k s t o a s s i s t second and f o r e i g n language l e a r n i n g . P a r t i c i p a t i o n by t e a c h e r s i n s m a l l group t a s k s p e r se n e i t h e r i n h i b i t s nor encourages n e g o t i a t i o n and r e p a i r o f the g r o u p ' s t a l k . What does seem 171 to matter i s the kind of task employed, the e x p e r i e n t i a l q u a l i t i e s of the task and the degree to which language i s used to accomplish educational purposes. 172 CHAPTER 6 : QUALITATIVE ANALYSIS OF THE TASK TRANSCRIPTS Unlike the previous two chapters, t h i s chapter i s an analysis of the t r a n s c r i p t s themselves; the data are thus expressed d i r e c t l y i n the words of the p a r t i c i p a n t s . Although the coding scheme has been retained f o r d e s c r i p t i v e purposes, i t i s not the frequency of coded behavior which i s of i n t e r e s t here but rather the ways i n which p a r t i c i p a n t s accomplish re p a i r and reference through discourse. Two topics have been selected for examination: d i s t i n g u i s h i n g between display and r e f e r e n t i a l questions and o u t l i n i n g the c h a r a c t e r i s t i c s of repair occurring i n what has been referre d to i n the previous chapter as complementary task structures, namely the confluence of expository behavior i n teaching tasks and e x p e r i e n t i a l a c t i v i t y i n non-teaching tasks. Both topics have been broached i n the quantitative analysis. Their further, q u a l i t a t i v e treatment here i s intended to r e i n f o r c e findings reached e a r l i e r with examples from the spoken texts, to s h i f t the focus of analysis from s t a t i s t i c a l averages to s p e c i f i c cases and thereby to a s s i s t second and foreign language teaching professionals to recognize key forms of r e p a i r i n context. Beyond providing s p e c i f i c cases of i n d i v i d u a l REs i n context, however, the q u a l i t a t i v e analysis w i l l attempt to r e l a t e the complementary task structures outlined i n Chapter 173 5 to r e p a i r processes i n selected t r a n s c r i p t s . Thus, co-occurring sets of REs w i l l be examined at a l e v e l of r e a l i t y which could not otherwise be achieved through quantitative methods. F i n a l l y , i t should be stressed that the analysis of variance reported i n Chapter 4 has provided a l i s t of REs f o r the q u a l i t a t i v e analysis which are c l e a r l y a l located to p a r t i c u l a r tasks i n Table 22. This means that d i r e c t i o n s f o r the textual analysis to follow are provided by s t a t i s t i c a l reasoning rather than by an observer's preliminary induction of s a l i e n t categories. Before beginning the analysis, i t should also be pointed out that interpretations of how various forms of r e p a i r are accomplished through discourse are always subject to a l t e r n a t i v e explanations. Although i n t e r p r e t a t i o n of the conversational data obtained i n t h i s study requires some understanding of the c u l t u r a l and s o c i a l background of the p a r t i c i p a n t s , that i s , a l e v e l of understanding beyond what might be c a l l e d a common-sense e x p l i c a t i o n of a t a l k between two people, i t would s t i l l be unreasonable to claim that other perceptions cannot be applied at l e a s t as a r t f u l l y , i f not as v a l i d l y , to the data. This aspect of q u a l i t a t i v e analysis i s b r i e f l y reviewed i n the following chapter during discussion about the g e n e r a l i z a b i l i t y of conclusions achieved through textual analysis. Display and Referential Questions The d i s t i n c t i o n between display and r e f e r e n t i a l questions was one of the clearest achieved during the 174 a n a l y s i s o f v a r i a n c e . How does t h i s d i s t i n c t i o n appear i n t h e t r a n s c r i p t s and under what c i r c u m s t a n c e s i s one form o f q u e s t i o n p r e f e r r e d o v e r the o t h e r ? One way o f d e a l i n g w i t h t h e s e i s s u e s i s t o examine t h o s e t r a n s c r i p t s i n wh ich the h i g h e s t average f r e q u e n c i e s were r e c o r d e d f o r each o f the q u e s t i o n t y p e s , t h a t i s , t r a n s c r i p t s f o r COM2 ( d i s p l a y q u e s t i o n s ) on t h e one hand and DIS and LEG2 ( r e f e r e n t i a l q u e s t i o n s ) on t h e o t h e r . COM2: The I n s t r u c t i o n a l Demonst ra t ion Task I t may be c o n v e n i e n t t o r e c a l l t h a t COM2 e n t a i l e d t h e t e a c h i n g o f how t h e s t r i n g s e a r c h f u n c t i o n on a l a p t o p p o r t a b l e computer o p e r a t e s . An i m p o r t a n t c h a r a c t e r i s t i c o f t h e t a s k was t h e p r e s e n t a t i o n o f i n f o r m a t i o n about t h e f u n c t i o n w i t h the computer a c t u a l l y a v a i l a b l e f o r i n s p e c t i o n and m a n i p u l a t i o n . T h i s s o r t o f e x p o s i t o r y p r e s e n t a t i o n o f i n f o r m a t i o n i s a f a i r l y common i n s t r u c t i o n a l p r o c e d u r e ; i n d e e d i t may r e p r e s e n t the most s a l i e n t p o p u l a r image o f what t e a c h i n g and l e a r n i n g a r e a l l a b o u t . I t may no t be s u r p r i s i n g , t h e n , t o f i n d t h a t the l e a r n e r ' s b e h a v i o r d u r i n g t h e such a p r o c e d u r e i s l a r g e l y p a s s i v e and d i r e c t e d by the t e a c h e r . D i s p l a y q u e s t i o n s a r e t h u s c o n s i s t e n t w i t h t h e t e a c h e r ' s r o l e as the s o u r c e o f knowledge and r e s p o n s i b i l i t y f o r examin ing t h e l e a r n e r ' s u n d e r s t a n d i n g , i . e . , f o r h a v i n g the l e a r n e r d i s p l a y u n d e r s t a n d i n g . Based on an e x a m i n a t i o n o f COM2 t r a n s c r i p t s , t h e r e seem t o be many forms o f d i s p l a y q u e s t i o n , however , each o f which s e r v e s a f u n c t i o n a p p r o p r i a t e t o communicat ion i n a t e a c h e r -175 directed i n s t r u c t i o n a l environment. A frequent a c t i v i t y i n t h i s sort of environment, of course, i s the teacher getting a demonstration of information which i s supposed to have been passed to the learner during a period of i n s t r u c t i o n . The most t y p i c a l sort of question during COM2 was, i n fact, s p e c i f i c a l l y intended by the teacher to display the learner's understanding, as the following two examples i l l u s t r a t e ( t r a n s c r i p t reference i n square brackets, teacher i n c a p i t a l l e t t e r s , learner i n lower case, display questions underlined): (1) THAT'S RIGHT. - NOW - UH - PRESS THE BUTTON. DO /Hm YOU REMEMBER THE NAME OF THE - KEY - THAT WE HAVE TO PRESS - TO FIN Ah, F - l . CAN YOU FIND F-l? F - l . Ah-hah! [2COM2] (2) CAN YOU PUSH - PUSH THE ARROW? WHAT HAPPENS WHEN YOU PUSH THE ARROW? Uh, I must f i n d i t ? OKAY. NOW WHAT DID YOU TYPE? ++ WHAT DID YOU TYPE JUST NOW? "The". OKAY. WHA. WHY - WHY DID YOU TYPE "THE"? I must f i n d out the - theory. The - f i r s t of. OKAY, WELL, IT ISN'T THAT YOU'RE LOOKING, YOU'RE REALLY NOT LOOKING FOR THE WORD, "THEORY". YOU'RE LOOKING FOR THE WORD "THE". [4COM2] 176 As b o t h examples i l l u s t r a t e , i t i s no t enough f o r t h e l e a r n e r t o m e r e l y push the the r i g h t k e y , a l t h o u g h d o i n g so i s c l e a r l y a p a r t i a l d e m o n s t r a t i o n o f u n d e r s t a n d i n g . What seems t o be even more i m p o r t a n t , however, i s t h a t t h e l e a r n e r v e r b a l i z e t h e a c t i o n o r even j u s t i f y i t . P a r t o f what c o n s t i t u t e s s u c c e s s f u l work on t h i s k i n d o f t a s k i s a c t i o n , bu t p a r t i s e x p l i c a t i o n o r (as i n o t h e r p a r t s o f the t r a n s c r i p t ) summary o f a s e r i e s o f p r o c e d u r e s . V e r y l i t t l e o f what the t e a c h e r i s a s k i n g t h e l e a r n e r t o say i s l i k e l y t o be s a i d d u r i n g what might be c o n s i d e r e d more o r d i n a r y c o n v e r s a t i o n a l a c t i v i t i e s . C o n v e r s a t i o n a l p a r t n e r s do no t o r d i n a r i l y query each o t h e r about t h e o b v i o u s o r about t h i n g s which a r e a l r e a d y known t o the q u e s t i o n e r . But i t i s p r e c i s e l y t h e o b v i o u s and a l r e a d y -known which s e r v e s as t h e c e n t e r o f t a l k i n s i t u a t i o n s d e v o t e d t o i n s t r u c t i o n o f s u b j e c t m a t t e r , p a r t i c u l a r l y t h o s e , such as COM2, which o f f e r a m o t i v a t i n g c o n t e x t f o r t a l k about the a l r e a d y - k n o w n . I t a p p e a r s , t h e n , t h a t l e a r n e r s ' o p p o r t u n i t i e s t o use language a r e c o n s t r a i n e d by t h e t e a c h e r ' s p e r c e p t i o n o f how w e l l a l e a r n e r a r t i c u l a t e s u n d e r s t a n d i n g o f i n s t r u c t i o n a l c o n t e n t . I s s u e s beyond c o n t e n t , such as t h e mutual i n t e l l i g i b i l i t y o f t h e d i s c o u r s e and i n f o r m a t i o n exchange, a r e r e l a t i v e l y u n i m p o r t a n t s o u r c e s o f c o n v e r s a t i o n a l a c t i v i t y . A s e c o n d a r y f u n c t i o n of t e a c h e r - l e a r n e r t a l k which emerged d u r i n g i n s p e c t i o n o f the COM2 t r a n s c r i p t s was v e r b a l p o i n t i n g - o u t , f o c u s i n g on something the t e a c h e r c o n s i d e r e d 177 useful f o r moving smoothly through the i n s t r u c t i o n a l procedure. The "display" q u a l i t i e s of pointing-out can be observed i n the following examples: (3) OKAY, UH, CAN YOU FIND THE - BUTTON SAYING "SHIFT"? Yes, I can. /OKAY, OH, PLEASE - DO NOT PRESS THAT SHIFT. [6COM2] (4) OKAY. NOW IT'S IDENTIFIED. YOU SEE WHERE THE CURSOR IS THERE? ++ SEE IT BLINKING? + Blink. THERE, YOU SEE IT BLINK? Yeah, the - f i r s t , the f i r s , t h i s one? Kore? /CURSOR RIGHT, OKAY. DON'T TOUCH THAT, THOUGH. OKAY. + NOW, IT'S FOUND THE FIRST LETTER, "A". [7COM2] The teacher has temporarily stopped the action and brought the learner to the point of an important i n s t r u c t i o n . The teacher c l e a r l y has a motive beyond confirming a common perception of the s i t u a t i o n . Here the teacher breaks the in s t r u c t i o n into two components, one which establishes a s p e c i f i c t o p i c (the s h i f t button, the b l i n k i n g cursor) and another which d i r e c t s behavior ("do not press that", "do not touch t h a t " ) . This approach to i n s t r u c t i o n resembles such forms of language s i m p l i f i c a t i o n as t o p i c a l i z a t i o n (Hatch, 1983) and decomposition of lengthy propositions (Long, 1980). However, i t i s perhaps easier to inte r p r e t the pointing-out 178 q u e s t i o n s i n (3) and (4) above as t h e t e a c h e r ' s a t tempt t o o p e r a t i o n a l i z e a c c u r a c y and s e q u e n c e — v a l u e s l i k e l y t o be promoted i n any f o r m a l i n s t r u c t i o n a l s e t t i n g . Q u e s t i o n s wh ich a r e i n t e n d e d t o r i v e t the l e a r n e r ' s a t t e n t i o n , t h e n , seem not so much o r i e n t e d toward c l a r i f i c a t i o n o f meaning as s e t t i n g up the c o n d i t i o n s f o r c o n v e y i n g p r o c e d u r a l i n f o r m a t i o n e f f i c i e n t l y . Q u e s t i o n s which s e t up key b e h a v i o r a r e c l o s e l y r e l a t e d t o q u e s t i o n s which immedia te ly d i r e c t per fo rmance but which t a k e t h e form o f a p o l i t e s u g g e s t i o n : (5) YOU WANT TO MOVE THE CURSOR AROUND A L ITTLE BIT TO SEE WHAT IT DOES? Y e s . YEAH. SO IT JUST GOES SPACE BY SPACE IN THIS CASE. [12COM2] (6) YEAH, CAN YOU MOVE THAT? ++ J - CAN YOU JUST TRY. PLAY WITH IT? + YEAH, UP AND DOWN OR L E F T , - RIGHT - ALL RIGHT. NOT THE TOP, YEAH, DOWN. OKAY. / A h no t / T o p ? - R i g h t , l e f t . [9COM2] (7) OKAY. CAN YOU PUSH " A " ? ++ RIGHT. Y e s . [4COM2] 179 (8) AND. HOW ABOUT TRYING AGAIN? SEE HOW MANY- WE'LL SEE HOW MANY WE HAVE. (4) How ma + Ha How many /MAYBE YOU CAN PUSH F-2 AGAIN. WE'LL SEE HOW MANY WE HAVE HERE. THERE'S ANOTHER ONE. Aaah. WHY DON'T YOU KEEP PUSHING UNTIL YOU GET TO THE END? KEEP PUSHING F-2. [3COM2] A strategy which emphasized language s i m p l i f i c a t i o n would not normally employ the kinds of p o l i t e suggestions c i t e d i n (5-8), above. As the examples indicate, p o l i t e suggestions are t y p i c a l l y much less d i r e c t than simple imperative forms ("Move the cursor around . . .") and thus employ more complex syntax. This property of such requests indicates that comprehensibility (as might be achieved through s i m p l i f i c a t i o n of the syntax) i s not nearly as important as the teacher's manipulation of the learner's behavior i n aid of moving through the lesson content. C l e a r l y the teacher i s not deaf to the learner's request for help i n (8), since the teacher repeats and expands on the i n i t i a l j u s t i f i c a t i o n f o r " t r y i n g again". The disingenuous q u a l i t y of the question, however, puts i t squarely into the display category and the o v e r a l l impression i s that the teacher i s exercising a r e g i s t e r expressly employed f o r t r a i n i n g s i t u a t i o n s — a kind of teacher t a l k . Similar r e g i s t e r s can be imagined for a number of other settings, such as asylums and h o s p i t a l s , i n which rel a t i o n s h i p s are characterized by dependence and r e l a t i v e incompetence on the part of one of 180 the i n t e r l o c u t o r s . A v a r i a t i o n on the use of questions as d i r e c t i v e s i s a kind of prompt or reminder i n which the teacher asks a question and then proceeds to supply part of the answer. Once again an impression of the learner's r e l a t i v e ignorance and dependence i s conveyed by the prompting form of display question: (9) WHAT'S THE SPELLING OF "TOGETHER"? T-O-G-E- -t-h-e-r. T-H-E-R. AND YOU WANT TO FIND A WORD? t-h-e. T-H-E. THEREFORE, THIS - CURSOR IS SHOWING. [10COM2] Since t h i s form of the question i s designed to produce a response from the learner, i t thus might seem to function as a check of the learner's a b i l i t y to s p e l l . Because the question seems so naive, however, because the response which the teacher wants to produce i s v i r t u a l l y assured even as the question i s posed, an al t e r n a t i v e motive may be at work. As i t turns out, the d i s t i n c t i o n between T-H-E produced with leading and following spaces, and without them, i s c r u c i a l to the computer's a b i l i t y to locate a s t r i n g accurately; spaces are important i n s t r i n g searching. The teacher apparently wanted to te s t the learner's understanding of t h i s q u a l i t y of s t r i n g searching even though the i n s t r u c t i o n might seem to be caught up with a s p e l l i n g problem. This 181 b i t of t a l k has helped to explain why the cursor didn't f i n d the, the word, and so i t was probably worth the teacher's investment i n having the learner focus on a r e l a t i v e l y simple, known piece of information i n order to make an important inference. A general feature of t h i s kind of i n s t r u c t i o n a l discourse, then, may well be the teacher's use of display questions to lead the learner ever-so-mincingly to the point of an inescapable conclusion which had not previously been made e x p l i c i t . The t r a n s c r i p t s show numerous long patches of dialogue i n which the teacher has the learner move gradually towards an inescapable conclusion. The teacher's tolerance f o r engaging i n t h i s l e v e l of somewhat tedious discourse may, i n fa c t , run out at times, even though a piece of information may s t i l l be worth bringing into the discussion. In such cases the teacher may f i n d i t more e f f i c i e n t to simply ask the question and then supply the whole answer without ever r e a l l y intending that the learner respond. The following examples i l l u s t r a t e t h i s pre-emptive approach: (10) WHEN WE WERE GOING THROUGH, I SHOULD HAVE STOPPED YOU AT TWO POINTS. WE HAD THE WORD "THEORY", T-H-E-O-R -I-E-S. WHY DID IT STOP AT THE WORD "THEORY"? BECAUSE WE TYPED IN T-H-E, BUT WE DIDN'T LEAVE A SPACE IN /Hnn FRONT OR A SPACE IN BACK... [4COM2] 182 (11) YES, AND WHAT HAPPENS? WE'VE COME UP TO ANOTHER + ANOTHER VERSION, ANOTHER USE OF THE WORD "NEEDS". THAT'S TWO TIMES. [5COM2] (12) OKAY. + SO. WHAT HAVE YOU GOT? NOW- THERE ARE - FIVE THINGS AT THE, UH, BOTTOM OF THE SCREEN, RIGHT? [9COM2] The pre-empting question e f f e c t i v e l y turns dialogue into monologue, i f only temporarily, and reduces the opportunities for negotiation. I t i s , however, a f a i r l y f a m i l i a r i n s t r u c t i o n a l resource which the teacher may f i n d of value even when the learner may be prepared to attempt a response. This small group of display question types by no means exhausts the v a r i e t y of display questions found throughout the 12 t r a n s c r i p t s . I t does, however, account f o r the great majority of display questions found i n the f a i r l y t r a d i t i o n a l i n s t r u c t i o n a l context represented by COM2. When the s e t t i n g changes r a d i c a l l y from one which i s l a r g e l y oriented to serving educational aims to ones which revolve around information exchange or problem-solving, a very d i f f e r e n t kind of question predominates. DIS and LEG2: Discussion and Cooperative Problem-solving DIS had the fewest formal constraints of any task. This i s not to say, of course, that discussion which has been given no e x p l i c i t objectives i s without very powerful controls on structure and development. The term "free discussion" i s reserved for the discussion task, however, 183 l a r g e l y t o i n d i c a t e t h a t the p a r t i c i p a n t s and no t t h e r e s e a r c h e r were r e s p o n s i b l e f o r the c o n t e n t and d i r e c t i o n o f t h e t a l k . By c o n t r a s t the c o n t e n t and d i r e c t i o n o f LEG2 was v e r y much i n f l u e n c e d by the i n i t i a l i n s t r u c t i o n s t o t h e p a r t i c i p a n t s t o b u i l d the model t o g e t h e r and by t h e m a t e r i a l s o f the model i t s e l f , i n c l u d i n g the g r a p h i c i n s t r u c t i o n s wh ich r e q u i r e d each p i e c e t o be p l a c e d i n a p a r t i c u l a r way. T h i s t a s k i s c h a r a c t e r i z e d by i t s f o c u s on c o o p e r a t i v e l y c o n s t r u c t i n g a model which can be v iewed by b o t h p a r t i c i p a n t s . I t i s thus perhaps more a c c u r a t e ( a l t h o u g h l e s s c o n s i s t e n t ) t o speak o f the dyad as c o n s i s t i n g o f an " i n f o r m a t i o n p r o v i d e r " and an " i n f o r m a t i o n consumer" r a t h e r than o f a t e a c h e r and a l e a r n e r . Compared t o t a s k s w i t h p u r e l y i n s t r u c t i o n a l p u r p o s e s , DIS and LEG2 r e d u c e d t h e impor tance o f the s t a t u s gap between t e a c h e r and l e a r n e r and r e q u i r e d an a c t i v e exchange o f i n f o r m a t i o n . Taken t o g e t h e r , t h e d i s c u s s i o n , f a c e - t o - f a c e Lego a n d , t o a s m a l l e r d e g r e e , b a c k - t o - b a c k Lego t a s k s p r o v i d e a c l e a r c o n t r a s t t o b o t h o f the t e a c h i n g t a s k s i n terms o f t h e o c c u r r e n c e o f r e f e r e n t i a l q u e s t i o n s . R e f e r e n t i a l q u e s t i o n s have g r e a t c u r r e n c y when p a r t i c i p a n t s a r e p r e p a r e d t o s u p p l y i n f o r m a t i o n which i s known (or a t l e a s t b e l i e v e d ) t o be o f i n t e r e s t t o a c o n v e r s a t i o n p a r t n e r . The f o l l o w i n g examples s u g g e s t a range o f f u n c t i o n s f o r r e f e r e n t i a l q u e s t i o n s i n d i s c u s s i o n and p r o b l e m - s o l v i n g s e t t i n g s . Most o f t h e e x c e r p t s have been taken from t h e d i s c u s s i o n t r a n s c r i p t s ( the g r e a t e s t number and v a r i e t y o f r e f e r e n t i a l q u e s t i o n s 184 occurred during free discussion) although problem-solving of the sort enforced during performance of the face-to-face Lego task produced a small and i n t e r e s t i n g set of r e f e r e n t i a l questions which d i d not occur during discussion. This set w i l l be examined b r i e f l y following examination of the discussion t r a n s c r i p t s . DIS: Free Discussion Within the broad range of questions c l a s s i f i e d as " r e f e r e n t i a l " which occurred during discussion, the most common question i s a response to the previous speaker's extension of the current t o p i c . Responsiveness here means acknowledgement of the previous speaker's utterance and encouragement to expand on i t . The responsive question o r d i n a r i l y aims at e l i c i t i n g fresh conversational material, although i t may well be the case that the material i t s e l f i s of l e s s importance than the fac t of conversation being extended cooperatively. (13) YEAH. I LIKED KOCHI AND I WAS A LITTLE BIT -SURPRISED. Ah! On what point? WELL, BEFORE I WENT TO SHIKOKU, I TOLD SOME OF MY FRIENDS THAT I WAS GOING TO SHIKOKU. AND THEY SAID, "SHIKOKU! AH!" IT'S REALLY IN THE STICKS. IT'S - UH - THERE ARE ONLY FARMERS. [2DIS] There i s no externally imposed goal or d i r e c t i o n f o r the conversation although the par t i c i p a n t s do seem to share an in t e r e s t i n drawing each other out. This increases the 185 chances that turns w i l l be linked to a t o p i c or that candidate topics can be examined for i n c l u s i o n i n the conversation. Allowing a speaker to elaborate on the speaker 1s t o p i c of i n t e r e s t i s t y p i c a l l y the i n i t i a l e f f e c t of a responsive question; the eventual e f f e c t of such generosity, however, i s to seed the conversation with points which can be taken up by a l i s t e n e r . (14) YEAH. + AND + WELL, MY HOMETOWN IS QUITE NEAR. Where? UH, SANDA. Sanda? HAVE YOU HEARD OF SANDA? No. IT'S IN THE MIDDLE OF, UH, HYOGO - PREFECTURE. [9DIS] The a l t e r n a t i o n of r e f e r e n t i a l questions from one p a r t i c i p a n t to the other as exemplified i n (14) i s the basis of information exchange about a t o p i c which i s so c h a r a c t e r i s t i c of the discussion task. Depending on the l e v e l of proficiency, such responsive questions may be a learner's major contribution towards development of a t o p i c and a useful means of getting more out of the conversation than i s put into i t . Frequently, however, responsive questions do not produce balanced exchanges of information but instead serve e i t h e r to tease out a topic piece by piece or to r a i s e 186 candidate topics for acceptance or r e j e c t i o n by the next speaker. Such cases t y p i c a l l y produce r e l a t i v e l y long responsive st r i n g s with one member of the dyad asking and the other member answering questions. (15) FUNA? Funa. Yes + umm ++ that i s one ++ urn - that i s a kind of + g o l d f i s h . UH HUH. Oh. AND YOU CAUGHT THAT IN THE POND? Urn yes. I t i s t y p i c a l Japanese f i s h which i s i n r i v e r ++ r i v e r or pond. AND THEN YOU. YOU. DID YOU YOU TAKE THAT HOME AND EAT IT THEN? Ha ha so ha ha ohhh no fe umm so few people eat i t but almost doesn't eat. AH. SO YOU THROW IT BACK? Yeah. YOU CATCH IT AND THEN THROW IT BACK IN? Yeah. WHEN YOU CAME TO FISH IN THE OCEAN. WAS THAT DIFFERENT? [3DIS] This way of constructing a topic c l e a r l y puts a burden on the teacher, the more p r o f i c i e n t member of the dyad, but i t also makes i t easier f o r the learner to exercise some r e s p o n s i b i l i t y f o r d i r e c t i n g the t a l k . The teacher's questions are l a r g e l y responsive to the learner, even though 187 the l e v e l of question s i m p l i f i c a t i o n i s not e s p e c i a l l y responsive to the learner's demonstrated a b i l i t y to handle unsimplified yes/no questions. An even greater conversational burden i s taken on when topics are thrown out for consideration by the l i s t e n e r . This a d d i t i o n a l general category of r e f e r e n t i a l q u e s t i o n s — r e f e r e n t i a l questions which help to bring new material into the conversation—lends a degree of u n p r e d i c t a b i l i t y to the conversation and enriches i t with opportunities f o r p a r t i c i p a n t s to insinuate personally i n t e r e s t i n g material into the ongoing t a l k . Sometimes the t o p i c i s pursued immediately; sometimes, however, topics must be raised one a f t e r the other u n t i l one i s found to be worth blending into the conversation. This method of examining prospective topics i s i l l u s t r a t e d i n (16), below. (16) UH HUH. (2) MMM. DO YOU HAVE ANY UH (2) CLUB. CLUB ACTIVITIES CONNECTED WITH OTHER UH ++ THE ESS. UH ESS OF OTHER UNIVERSITIES? Mmm. Yes we have. So c a l l e d K-I- K-I-E-F... AH HAH. /mmm. AH HAH. DO YOU HAVE A PART TIME JOB? UH Yes. Ahh as a tutor. YOU TEACH ENGLISH. Yes. /UH HUH. Or h i s t o r y . UH HUH. WHERE ARE YOU FROM? FROM OSAKA OR [10DIS] 188 The learner i s responsive, and even volunteers information, but i s somehow unable to help the teacher f i n d a qu a l i f y i n g t o p i c . Even though a stable topic i s not yet avail a b l e , the pa r t i c i p a n t s s t i l l t r e a t each other's contributions as worthy of response. Eventually, however, new topics are introduced into the discussion (17-18) and form stable resources f o r exchanging information. The following example shows one common way i n which t h i s i s accomplished through a r e f e r e n t i a l question. (17) WHERE- WHERE IN KANSAI., UH, IS YOUR HOMETOWN? Mmmm. Kobe City . AH, KOBE CITY? HAH HA HA HA. + WELL, AAAND. UHH. DO YOU- HAVE YOU EVER SORT OF, UM. + EXPERIENCED NEW YEAR'S IN KOBE? No. N000. ++ UHHH , + RIGHT. THEN WHAT- WHAT SORT OF THINGS YOU'RE GOING TO DO? ++ DURING - DURING THE NEW YEAR'S? During New Year's? YEAH. /Mmmm. To t e l l the truth, my- um, my brother s t i l l -i s studying YEAH. For- entrance examination? + And I would l i k e to /AH, HAH-HAH-HAH. help him. YOU WOULD LIKE TO HELP HIM, YEAH. /Hmmm. And also, I wanted to meet my friends. [9DIS] The teacher's i n i t i a l question i s the end of a s t r i n g of 189 responsive questions which are not t o p i c a l l y developed. The next question, however, begins a period of development i n which both p a r t i c i p a n t s share more or less equally. This question contains a group of markers ("WELL, AAAND, UHH") which function to hold a turn and indicate the imminence of a fresh t o p i c to the l i s t e n e r . Although an in t e r p r e t a t i o n of such markers i s quite speculative at t h i s point, i t appears that the speaker intends to signal the exhaustion of one l i n e of questioning and the beginning of another; the l i s t e n e r i s also n o t i f i e d that what i s about to be mentioned i s probably worth consideration as the next t o p i c . Markers of the sort employed i n (17) appear throughout the t r a n s c r i p t s as l i k e l y topic boundaries. Frequently, however, topics are "pushed" much more aggressively than simply marking t h e i r desirable s t a r t i n g point. As (18) indicates, p a r t i c i p a n t s may preface a r e f e r e n t i a l question with material which i n v i t e s a p a r t i c u l a r response from the l i s t e n e r and which thus has the e f f e c t of d i r e c t i n g the course of the conversation. (18) Uh, so - some my friends l i k e s - wearing kimono, but many of - of - most of my friends don't l i k e i t . How about your - wife? Does your wife wear kimono? MMM, WELL, OF COURSE MY WIFE IS ENGLISH, SO SHE THINKS IT LOOKS VERY - SHE DOESN'T FEEL UH COMFORTABLE + WEARING KIMONO. SHE FEEL IT'S - LOOKS A BIT + STRANGE. /Comfortable! Ha, ha. Yes, I understand. [5DIS] 190 This s e t t i n g up of the conditions f o r the teacher's response i s apparently e f f e c t i v e . The teacher's response prompts the learner to re-assert an in t e r e s t i n the topic, although evidence of the learner's misunderstanding ("Comfortable!") i n i t i a t e s some corrective action i n the form of a restatement. The negotiation i s resolved s u c c e s s f u l l y i n t h i s case, but even without a successful r e s o l u t i o n the fact remains that the learner was able to set up a str e t c h of the conversation and cooperate i n i t s development. The t r a n s c r i p t s for DIS also demonstrate how par t i c i p a n t s set up t h e i r own prospective contribution across more than one turn through use of r e f e r e n t i a l questions. Although the l i s t e n e r i s t y p i c a l l y i n v i t e d to negotiate the i d e n t i t y of a t o p i c a l focus (a person, a thing, a place), i t i s not so much the l i s t e n e r ' s responses that matter as i t i s the i n i t i a l speaker's intention to expand a to p i c of personal i n t e r e s t with the formal cooperation of a conversation partner: (19) DO YOU KNOW. YOU KNOW WOODY ALLEN. DON'T YOU? /Hmm. I - Allen? WOODY. WOODY ALLEN. Yeah, I know. YEAH. WOODY ALLEN AS A PERSON - IS - MISERABLE. /Hmm, oh. [4DIS] 191 "Formal cooperation" i n (19) means that the l i s t e n e r i s given opportunities to answer the question, to take o f f i c i a l turns i n order to continue cooperative work on the t o p i c . The negotiation done here i s s i g n i f i c a n t i n terms of the kinds of repairs the learner experiences (including a request f o r c l a r i f i c a t i o n , r e p e t i t i o n and s e l f - r e p e t i t i o n ) . I t i s also s i g n i f i c a n t , however, for what i t demonstrates about the ways i n which r e f e r e n t i a l questions can be employed to e f f e c t the construction of discourse beyond the next conversational turn. LEG2: Cooperative Problem-solving The two problem-solving tasks (LEG1 and LEG2) were, l i k e free-discussion, important sources for the production of r e f e r e n t i a l questions. LEG2 w i l l be considered here, however, because i t was a somewhat r i c h e r source of r e f e r e n t i a l questions than LEG1 (see Table 22). In addition, LEG2 extends the range of questions found i n the discussion task, the only other non-teaching task which involved the p a r t i c i p a n t s i n face-to-face t a l k . The t r a n s c r i p t s for both Lego tasks indicate an intense concern with f i n d i n g and placing pieces i n accordance with the graphic i n s t r u c t i o n s . I t i s not surpr i s i n g , then, that an e s s e n t i a l feature of the discourse was negotiation over the precise l o c a t i o n and appearance of objects, and the r e l a t i o n s h i p of one object to another. What p a r t i c u l a r l y distinguished LEG2, however, was the p a r t i c i p a n t s ' a b i l i t y to see what they were t a l k i n g about and the influence of 192 t h i s q u a l i t y o f p e r c e p t i o n had on t h e d i s c o u r s e . One e f f e c t a l r e a d y n o t e d i s the r e l a t i v e l y h i g h l e v e l o f e x o p h o r i c r e f e r e n c e . A n o t h e r , no ted h e r e , i s the p a c i n g o f t u r n s based on t h e s e a r c h f o r o r p lacement o f an o b j e c t . One p a r t i c i p a n t , f o r example, may t a k e a t u r n t o ask how s e v e r a l o b j e c t s might be d i s t i n g u i s h e d from each o t h e r ; t h e o t h e r p a r t i c i p a n t may t a k e the next t u r n t o respond w i t h what i s i n t e n d e d t o be i n f o r m a t i o n o f use i n making t h e d i s t i n c t i o n . Among t h e most common forms o f r e f e r e n t i a l q u e s t i o n d u r i n g t h i s k i n d o f c o o p e r a t i v e exchange a r e q u e s t i o n s wh ich h e l p the p a r t n e r t o p r o v i d e e f f i c i e n t ( i . e . , t i m e - and l a b o r -s a v i n g ) i n f o r m a t i o n , q u e s t i o n s which r e q u e s t a f o c u s ; (20) UH, NO, I GUESS THAT - i s n ' t - THAT ISN 'T IT . AH . / N o ? Which - wh ich one? /CAN YOU FIND ANOTHER ONE? WHERE IS IT? UH Mmm. THINK IT MIGHT BE CLOSE + IS THAT IT? OR IS THAT THE ONE YOU T h i s i s not + t h i s one? UH, UH, TRY IT . SEE IF YOU CAN PUT IT IN THERE - IN THE SAME WAY. [2LEG2] The t e a c h e r i n t h i s c a s e , o f c o u r s e , i s t r y i n g t o p r o v i d e u s e f u l i n f o r m a t i o n t o the c o n v e r s a t i o n p a r t n e r and i s c o n t i n u a l l y c h e c k i n g the p o s i t i o n o f the p i e c e s on the t a b l e a g a i n s t the p o s i t i o n o f p i e c e s on the i n s t r u c t i o n s . What s e r v e s w e l l as a t e a c h e r ' s d i s p l a y q u e s t i o n i n t h e 193 i n s t r u c t i o n a l context becomes a r e f e r e n t i a l question i n the problem-solving context. Exophoric reference i s an i n t e g r a l part of v i r t u a l l y a l l turns and i s t y p i c a l l y expressed through the medium of the r e f e r e n t i a l question. The transformation of one kind of question into another depends v i r t u a l l y not at a l l on who i s t a l k i n g to whom, but rather on the underlying intention of one of the p a r t i c i p a n t s , the "teacher", to eit h e r teach or to exchange information with an equal i n the business of moving a problem towards solut i o n . In (20), above, both p a r t i c i p a n t s must ask and answer questions which focus the partner's attention; both p a r t i c i p a n t s thus e f f e c t i v e l y commit each other to supplying the needed information. During construction of the Lego model p a r t i c i p a n t s would frequently request each other to assess performance or perception, or to otherwise provide guidance i n the placement of pieces. In a cooperative s i t u a t i o n , such requests i n the form of r e f e r e n t i a l questions are powerful influences on the speed with which the problem i s solved and frequently constitute useful a l t e r n a t i v e s to a simple direct-and-respond strategy. Indeed, the negotiation which often follows the request, f a r from wasting time i n roundabout discussion, i s a central feature of e f f i c i e n t information exchange during problem-solving. The following excerpts i l l u s t r a t e three commonly employed requests f o r assistance: request for assessment (21), request f o r d i r e c t i o n (22) and request for explanation (23). 194 (21) ON THE TOP. THEY'RE BLACK, THEY'RE SMALL. Umituti. I s n ' t t h a t r i g h t ? Here? YEAH, T H A T ' S , THAT'S ONE. YEAH. AND THE OTHER ONE ALSO + ALSO IS - NO, THAT'S NOT IT [4LEG2] (22) THE TRUCK. BUT YOU MUST TURN THE WHEELS SO THAT THE, THE PIECE WITH BUMPS IS UP. I, I can connec t? YEAH. Um hut. The bumps. YOU WANT TO CONNECT THE WHEELS TO THE MAIN BLACK P I E C E . THERE YOU GO. [12LEG2] (23) I t ' s i m p o s s i b l e , maybe. / I T ' S IMPOSSIBLE? + IT IS? + UH + DON'T WE FIND - SIMILAR - SQUARE ONE WHICH - CAN -/ T h i s - Huh? PUT IN THERE? + YOU CAN'T (4) WOW. WHAT HAPPENED? I d o n ' t know. [10LEG2] A l l o f t h e s e c a n be d i s t i n g u i s h e d from r e q u e s t s f o r c l a r i f i c a t i o n by v i r t u e o f t h e i r f o c u s on t h e t a s k r a t h e r t h a n t h e language by which the t a s k i s moved f o r w a r d . A c c o r d i n g l y , t h e n e g o t i a t i o n i s p r e c i p i t a t e d by t h e r e f e r e n t i a l q u e s t i o n s — q u e s t i o n s about c o n t e n t and a c t i o n r a t h e r t h a n l a n g u a g e — w h i c h i n v o l v e t h e l i s t e n e r i n c o n s t r u c t i o n o f an a p p r o p r i a t e r e s p o n s e . What c o n s t i t u t e s a p p r o p r i a t e c o n t e n t f o r the t u r n which f o l l o w s the q u e s t i o n i s sometimes o n l y b a r e l y s u g g e s t e d by t h e form i n which the 195 question i s put. Given a task i n which sequenced information i s v i t a l to successful communication, such as LEG2, e f f i c i e n t requests for d i r e c t i o n can be made with a sin g l e word and can also signal completion of one step and readiness f o r the next: (24) WE HAVE THREE WHEELS. TAKE, TAKE THEM. Yeah. ++ And? ALL RIGHT. PICK UP THE PIECE AND PUT THE WHEELS ON THE BOTTOM. [4LEG2] Sometimes, however, a partner's perception must be assessed e x p l i c i t l y before the next step can be taken. This would o r d i n a r i l y be accomplished by the person g i v i n g d i r e c t i o n s , although nothing i n p r i n c i p l e p r o h i b i t s the re c i p i e n t of di r e c t i o n s from checking on the partner's view of objects i n the task environment. Assessment of perception may s i g n i f y that the task has reached a turning point or that the partner has demonstrated uncertainty, or an unconvincing degree of certainty, about what to do next. Functions of t h i s sort are i l l u s t r a t e d i n (25), below. (25) YEAH. AND UH, THEN PUT THE LONG BLACK PIECE ON THEM. + DO YOU SEE HOW THE WHEELS HAVE BUMPS + TO HOLD PIECES? Mm, I /BETWEEN THE TWO WHEELS. Between the two wheels. YEAH. LOOK WHAT'S BETWEEN THE TWO WHEELS. LOOK AT THE PART BETWEEN THE TWO WHEELS. PICK UP A PIECE OF 196 WHEEL. YEAH, OK. SEE THE PART? IT HAS A BUMP. IT HAS A THING WITH BUMPS TO + HOLD SOMETHING. OK, YEAH. /two, uh four, four bumps. Mm /OK. [12LEG2] What seems to d i s t i n g u i s h questions framed to check a partner's perception of the s i t u a t i o n from questions intended to check learning (as i n teaching tasks) i s the emphasis on getting to the next step. A partner's perception becomes e n t i r e l y i r r e l e v a n t once t h i s has been accomplished and there i s no spec i a l value placed on the information beyond f a c i l i t a t i n g the task i t s e l f . Display and Referential Questions: Summarizing the Contrasts Between the Teaching and Non-teaching Tasks The analysis of variance demonstrated s i g n i f i c a n t differences between tasks on the basis of question types, most p a r t i c u l a r l y on the basis of display and r e f e r e n t i a l questions. The textual analysis has further examined the contexts i n which p a r t i c u l a r kinds of display and r e f e r e n t i a l questions occur and outlined various discourse functions which are accomplished by these questions. This analysis has c l a r i f i e d the intensive use of display questions during the two teaching t a s k s — t h a t display questions are ess e n t i a l features of a teacher's i n s t r u c t i o n a l behavior—and underscored the r o l e of r e f e r e n t i a l questions as fundamental structures of ordinary s o c i a l exchange. 197 The textual analysis has also suggested several differences between tasks based on the predominance of e i t h e r display or r e f e r e n t i a l questions. These differences, i t should be emphasized, are based on the observation of dyadic discourse i n an experimental rather than a n a t u r a l i s t i c s e t t i n g , although the q u a l i t a t i v e method of analysis has treated the discourse as natural texts. 1) Whereas r e f e r e n t i a l questions serve a broad v a r i e t y of functions r e l a t e d to information exchange i n non-teaching tasks, display questions focus r e l a t i v e l y narrowly on the extent and q u a l i t y of learning associated with a teacher's i n s t r u c t i o n a l purposes. 2) Following from 1), the scope of r e f e r e n t i a l questions i n a given non-teaching text appears much wider than i n a given teaching text. Opportunities f o r learners to negotiate the language by which the task i s accomplished with a w i l l i n g partner are thus considerably increased over the opportunities a v a i l a b l e during formal i n s t r u c t i o n . 3) By d e f i n i t i o n , display questions have only one correct answer; the a r b i t e r of correctness i s i n v a r i a b l y the teacher. This feature of display questions contributes to t h e i r r e l a t i v e e f f i c i e n c y i n accomplishing educational purposes, although they would appear to be r e l a t i v e l y i n e f f i c i e n t i n creating the conditions f o r f r e e l y exploring topics which have not been planned p r i o r to i n s t r u c t i o n . R e f e r e n t i a l questions are often open-ended; while they may 198 influence behavior, they do not necessarily require a s p e c i f i c response. 4) Following from 3 ), r e f e r e n t i a l questions are es p e c i a l l y useful i n opening up conversational topics which occur as each speaker takes a turn. Indeed, although r e f e r e n t i a l questions are not necessary f o r information exchange to occur, i t i s d i f f i c u l t to imagine conversations i n which they do not play an important part i n making exchange of information easier. Although the point i s speculative and requires further assessment, non-teaching tasks which emphasize the solution of problems may be conducted with greater e f f i c i e n c y — w i t h greater speed and d i r e c t movement towards the solution—when r e f e r e n t i a l questions are applied to the task than when they are not. This would seem to be an unintended, although f e l i c i t o u s , e f f e c t of use of r e f e r e n t i a l questions. In contrast, display questions are c h a r a c t e r i s t i c a l l y intended to improve the e f f i c i e n c y of ins t r u c t i o n . I t i s s t i l l very much an open question, however, whether i n s t r u c t i o n which i s conducted with the use of display questions r e s u l t s i n a fas t e r or higher l e v e l of achievement than without t h e i r use. 5) The negotiated character of non-teaching t a l k i s marked by recourse to r e f e r e n t i a l questions (and such other re p a i r behaviors as confirmation checks, c l a r i f i c a t i o n requests and s e l f - r e p e t i t i o n s ) . Although r e f e r e n t i a l questions are o r d i n a r i l y ostensibly targeted on the content 199 o f t h e t a l k — w a y s t o spend New Y e a r ' s , c o n s t r u c t i o n o f a t o y s p a c e c r a f t — t h e y a r e a l s o v e r y much i n v o l v e d i n t h e moment-to-moment c o m p r e h e n s i b i l i t y o f the t a l k t o each o f the p a r t i c i p a n t s . J u s t as a c l a r i f i c a t i o n r e q u e s t may l e a d a p a r t n e r t o r e p e a t o r r e p h r a s e , a r e f e r e n t i a l q u e s t i o n which p r e s s e s t h e l i s t e n e r t o expand o r e x e m p l i f y a s ta tement a l s o t h u s i n i t i a t e s a p e r i o d o f n e g o t i a t i o n . S i n c e t h e language o f d i s p l a y q u e s t i o n s , by c o n t r a s t , i s f u n d a m e n t a l l y c o n c e r n e d w i t h t e s t i n g knowledge o r u n d e r s t a n d i n g , t h e k i n d o f n e g o t i a t i o n which does o c c u r d u r i n g a t e a c h i n g t a s k i s c h a r a c t e r i z e d by such moves as l e a r n e r - p r o d u c e d e x p r e s s i o n s o f l e x i c a l u n c e r t a i n t y o r t e a c h e r - p r o d u c e d d e f i n i t i o n s . Under t h e s e i n s t r u c t i o n a l c o n d i t i o n s , i t i s no t e s p e c i a l l y l i k e l y t h a t t e a c h e r s w i l l e n t e r t a i n q u e s t i o n s from l e a r n e r s wh ich e f f e c t i v e l y s h i f t t h e power t o nominate and c o n t r o l t o p i c s . 6) The f i n a l d i s t i n c t i o n t o be drawn between d i s p l a y and r e f e r e n t i a l q u e s t i o n s i s e x p l i c i t l y c o n c e r n e d w i t h the d i s t r i b u t i o n o f power and r i g h t s o v e r t a l k d u r i n g v a r i o u s t a s k s . D i s p l a y q u e s t i o n s can be v iewed as o p e r a t i o n a l i z a t i o n o f the t e a c h e r ' s r e s p o n s i b i l i t y f o r o r g a n i z i n g and c a r r y i n g out i n s t r u c t i o n . D i s p l a y q u e s t i o n s c o n s t i t u t e a k i n d o f ready e v i d e n c e t h a t the t e a c h e r i s t h e c e n t e r o f c o n t r o l i n the i n s t r u c t i o n a l p r o c e s s and t h a t t h e l e a r n e r , i d e a l l y , i s w i l l i n g t o demonst ra te the e x t e n t o f l e a r n i n g w i t h o u t engag ing the t e a c h e r i n an exchange o f i n f o r m a t i o n . Beyond a f a i r l y n a r r o w l y c o n s t r u e d l e v e l o f 200 exchange, perhaps i n the form of a ro l e i n a textbook d r i l l or an expression of l e x i c a l uncertainty which the teacher chooses to deal with, the learner i s u n l i k e l y to exercise much control over eit h e r the teacher's or the learner's p a r t i c i p a t i o n . Referential questions i n ordinary conversational or problem-solving tasks, by way of contrast, s i g n i f y a s h i f t i n the balance of r e s p o n s i b i l i t y f o r re p a i r i n g and elaborating t a l k . In the t r a n s i t i o n from i n s t r u c t i o n a l to non-instructional contexts, teachers give up some of t h e i r control over the discourse while learners take much of i t i n . Repair i n Complementary Task Structures The discussion next turns from the forms of display and r e f e r e n t i a l questions i n various task settings to the functions of two small groups of repairs found i n complementary task structures (Figure 18). These structures are the repairs found at the i n t e r s e c t i o n of the teaching and expository tasks on the one hand (hereafter Group 1. including d e f i n i t i o n s and expressions of l e x i c a l uncertainty), and the non-teaching and ex p e r i e n t i a l tasks on the other (Group 2, including confirmation checks and r e f e r e n t i a l questions). I t should be stressed that Group 1 and Group 2 structures are merely convenient ways of summarizing the r e s u l t s of a quantitative analysis. Further, Group 1 and Group 2 structures are only suggested by the analysis and not demonstrated by i t . The b r i e f examination which follows thus has a foundation, although i t 201 must s t i l l be described as a f a i r l y speculative way of d i s t i l l i n g the discourse into c l e a r l y contrasting sets. The general questions to be raised during the analysis are, "What are the l i n k s , i f any, between the repairs i n each of the task structures? What does the discourse look l i k e when the repairs i n each category co-occur?". This view of the t r a n s c r i p t s i s intended to o f f e r some ins i g h t into the "short l i s t s " , the d i s t i l l a t i o n , of repair exponents as they are found i n two, apparently very d i f f e r e n t , task environments. Group 1: Defin i t i o n s and Expressions of Lexical Uncertainty More than any of the other tasks, COMl was concerned with communication of abstract knowledge from teacher to learner e n t i r e l y through a verbal medium. By comparison, the back-to-back lego task, LEGl, permitted reference to graphic i n s t r u c t i o n s and required one of the p a r t i c i p a n t s to move objects around on the table. As an i n s t r u c t i o n a l task, moreover, COMl involved the teacher i n frequent, short digressions over b i t s of knowledge related to, but not es s e n t i a l for, proper operation of the computer. These digressions t y p i c a l l y took the form of d e f i n i t i o n s , some of which were e l i c i t e d by the learner, some of which seemed to anti c i p a t e a question from the learner over a just-mentioned step i n the procedure. Although d e f i n i t i o n s which were apparently unrelated to l e x i c a l uncertainty were found i n the t r a n s c r i p t s , evidence of l e x i c a l uncertainty more 202 t y p i c a l l y helped the teacher to weave abstract knowledge i n the form of a d e f i n i t i o n into the t a l k . This f a i r l y s ophisticated kind of verbal performance, i t should be noted, i s the province of the teacher during i n s t r u c t i o n a l t a l k . Although teachers may e l i c i t d e f i n i t i o n s from t h e i r learners i n order to t e s t knowledge, i t i s more t y p i c a l l y the case that teachers t r e a t l e x i c a l uncertainty, both t h e i r own and the learner's, as a kind of tri p w i r e f o r production of a d e f i n i t i o n . This r e l a t i o n s h i p of l e x i c a l uncertainty and d e f i n i t i o n , s p e c i f i c a l l y of l e x i c a l uncertainty occasioning d e f i n i t i o n (LLEX -> DDEF), i s i l l u s t r a t e d i n the next excerpt. The excerpt begins with a display question designed to t e s t cognitive knowledge (understanding of the concept s t r i n g ) . LLEX and DDEF are indicated by underlining and marginal notation; r e p a i r exponents f o r which the learner i s responsible are shown with an S added (e.g., LLEXS). (2 6) DO YOU KNOW WHAT A - PIECE OF STRING IS? St r i n g IN OTHER WORDS, THREAD? Thread? Ah. thread. Uh -THAT'S IT. A STRING IS JUST A THICK PIECE OF THREAD BUT IN, IN COMPU UH, COMPUTER / r i b - ribbon, ribbon. No. DOESN'T HAVE A RIBBON. TH - THERE'S A SPECIAL MEANING OF THE WORD STRING IN THE COMPUTER. IT JUST MEANS A WORD. A PHRASE OR SENTENCE. LLEXS DDEF LLEXS DDEF [2C0M1] 203 The learner's l e x i c a l uncertainty i s taken as evidence that more i n s t r u c t i o n i s required, which the teacher supplies immediately i n the form of d e f i n i t i o n s — o n e following each i n d i c a t i o n of l e x i c a l uncertainty. An open-and-shut r e l a t i o n s h i p between l e x i c a l uncertainty and d e f i n i t i o n i s a f a i r l y common pattern, although i t i s not always the case that opportunities f o r negotiation are so abbreviated by the teacher's i n t e r e s t i n i n s t r u c t i n g . For example, the impression of a learner's l e x i c a l uncertainty may be demonstrated over several turns and through the use of several devices: echoes, fumbling over words or phrases (as i n excerpt 26, above), or, simply, d i r e c t requests for information about something the teacher has recently inserted into the conversation. Because such devices serve to recycle conversational material, or to r e d i r e c t the discourse over several turns, negotiation i s c l e a r l y i n evidence when they are employed. The following i l l u s t r a t e s how learners can r i v e t the teacher's attention to the learner's l e x i c a l p r i o r i t i e s . LLEX i n the form of a fumbling search for the r i g h t word does not s p e c i f i c a l l y p r e c i p i t a t e a d e f i n i t i o n , although the o v e r a l l impression i s that l e x i c a l uncertainty i s the basis of the learner's claim on the teacher's attention. (27) AND I THINK THIS IS THE UH + THE SWITCH FOR - ADJUSTING REFRECTION - OF THE UH -LIGHTS Refraction means 204 MEANS What? AH, REFRECTION + OKAY, UM + HE HERE WE HAVE LIGHT OKAY. - AND ON THE SURFACE - OF THE GLASS. DDEF Yes. OKAY, THE LIGHT REFRECTS. Ah hnn - Yes, yes. /RIGHT? - THAT THE REFRECTION. THE NOUN FORM OF REFRECT. /Uh ahhh, ahh, ah, I understand, yes. /OKAY? [6COM1] The teacher's discussion of " r e f l e c t i o n " (including the i n i t i a l attempt at a d e f i n i t i o n ) i s constructed across several turns and c l o s e l y follows the learner's interventions. Even though the task as a whole i s devoted to i n s t r u c t i o n , more s p e c i f i c a l l y the teacher's intention to cover a l i m i t e d set of objectives established before the task begins, i t i s the learner who manages to channel part of the i n s t r u c t i o n a l process towards resolution of trouble created during the course of the conversation. This may not be e s p e c i a l l y what the teacher had intended, although extension of the d e f i n i t i o n across turns and through negotiation does indicate that Group 1 settings are not driven only by author i t a t i v e monologues. Further v a r i a t i o n from the simple LLEX -> DDEF pattern appears i n the t r a n s c r i p t s . Sometimes, f o r example, d e f i n i t i o n precedes or occurs v i r t u a l l y simultaneously with 205 an e x p r e s s i o n o f l e x i c a l u n c e r t a i n t y (DDEF -> LLEX o r DDEF + L L E X ) . These a r e c a s e s i n which t h e d e f i n i t i o n i t s e l f c o m p r i s e s the s o u r c e o f n e g o t i a b l e t r o u b l e . I t t h u s m a t t e r s r e l a t i v e l y l i t t l e whether a d e f i n i t i o n i s p repackaged o r o t h e r w i s e made d i f f i c u l t t o b reak down, i f i t i s g o i n g t o be " c h a l l e n g e d " by one o f the p a r t i c i p a n t s . As t h e f o l l o w i n g e x c e r p t s u g g e s t s , d e f i n i t i o n s may w e l l be a u s e f u l s o u r c e o f n e g o t i a b l e m a t e r i a l d u r i n g t e a c h i n g t a s k s . (28) SO, UH, WHEN YOU WANT TO MOVE THE MARKER, WE CALL THAT MARKER A CURSOR. DDEF C u r s o r . CURSOR. IT COMES FROM THE WORD MEANING "RUN". OR MOVE. SO IT . IT SHOWS WHICH / c u r s o r WORD THE COMPUTER IS WORKING ON. + SO THE DDEF /Mm hm BUTTONS AT THE RIGHT ++ CORNER CAN BE /Mm USED TO MOVE THE CURSOR UNTIL YOU FIND THE WORD YOU WANT. Mm. T h e , u h , + l e t t e r s , uh f i l e s ? LLEXS THE WORDS ON THE SCREEN RIGHT NOW. AFTER / w o r d s YOU TURN THE MACHINE ON, UH /moves + l e f t o r r i g h t ? MOVE, YEAH, THE CURSOR MOVES LEFT OR RIGHT. THE WORDS STAND S T I L L ON THE SCREEN AND THE CURSOR MOVES FROM ONE WORD TO THE NEXT WORD. / A h h a h . C u r s o r moves! [12COMl] The d e f i n i t i o n poses a l e x i c a l p rob lem f o r the l e a r n e r which t h e t e a c h e r h a n d l e s th rough a n a l o g y , example and r e f e r e n c e t o p a r t s o f an i n v i s i b l e computer . V i r t u a l l y a l l 206 of what the teacher says i s keyed to the learner's responses: an echo which could reasonably be taken as a request for c l a r i f i c a t i o n ("Cursor."), a second echo ("cursor") which appears to influence the teacher's reference to the cursor keys, a d i r e c t request f o r c l a r i f i c a t i o n which prompts further reference to objects on the imaginary screen, and a r e f e r e n t i a l question which obtains confirmation through r e p e t i t i o n and expansion. L e x i c a l uncertainty and other indications of trouble i n t h i s example, then, are a product of d e f i n i t i o n and generate negotiation over meaning and repair p r e c i s e l y where they are needed. D e f i n i t i o n and l e x i c a l uncertainty also co-occur within a given speaker's turn i n close proximity to each other. Given the nature of C0M1, however, the d i f f i c u l t y the speaker e x p e r i e n c e s — s p e c i f i c a l l y , the i n d i c a t i o n of l e x i c a l u n c e r t a i n t y — i s treated quite d i f f e r e n t l y depending on whether the speaker happens to be the teacher or the learner. The preference for s e l f - r e p a i r (see, f o r example, Sacks et a l . , 1977; Schwartz, 1980) i s v i t i a t e d i n the teacher's behavior (29); other-repair (see Kasper, 1985), however, i s more l i k e l y when the learner signals uncertainty (30) . (29) OR. UH. - EY ++ UH. I DON'T KNOW HOW TO CALL THIS. BUT UH. - IT'S CALLED UH. LLEX GROUP OF WORDS. OR. THE WORD - IS CALLED "STRING". + AND SO UH, YOU CAN, IF Y- DDEF WITHOUT HELP OF A COMPUTER, YOU CAN 207 LOCATE ++ UH, THE EXISTENCE - OF EACH, OF SUCH EXPRESSION OR WORDS. [8COM1] The turn continues f o r an additional 55 words and altogether includes seven r e l a t i v e l y long pauses. In perhaps more ordinary conversational environments, l i s t e n e r s can use these pauses to s t e a l a turn. In t h i s case, the learner did not attempt to help i n either the period of l e x i c a l uncertainty or the d e f i n i t i o n . When the learner, however, has been instructed to produce a d e f i n i t i o n , and then signals l e x i c a l uncertainty, i t may not be sur p r i s i n g to f i n d that i t i s the teacher who takes r e s p o n s i b i l i t y for making the repair. (30) WHAT'S TH- WHAT'S THE MENU? The menu i s the choice - uh - what - uh what - uh what can I ch- - uh, the LLEXS choice of - my - the section. DDEFS OKAY. IN THE COMPUTER - THERE - ARE RECORDED IN THE MEMORY OF THE COMPUTER A NUMBER OF - FILES. THESE FILES CONSIST /Hmm. OF PAPERS WITH INFORMATION. [4COMl] Based on t h i s b r i e f examination of a r e l a t i v e l y small group of co-occurring repairs i n tasks which are conducted through teaching and exposition, i t may be useful to suggest the ambivalence of the Group 1 task. Although negotiation over d e f i n i t i o n s and l e x i c a l uncertainty can and does occur i n Group 1 tasks, a Group 1 task can also be 208 conducted by d i r e c t i n g or otherwise co-opting the learner's responses. This sort of ambivalence i s generally not found i n the non-teaching and e x p e r i e n t i a l task structure, the f i n a l f i e l d of analysis to which the discussion now turns. Group 2: Confirmation Checks and Referential Questions Although the q u a l i t i e s of r e f e r e n t i a l questions have been outlined i n the analysis of discussion and problem-solving tasks, the confluence of r e f e r e n t i a l questions and confirmation checks requires a closer look. In general, t r a n s c r i p t s f o r both Lego tasks show that these forms of r e p a i r taken together are p i v o t a l sources of i n t e r a c t i o n . What are the basic patterns of i n t e r a c t i o n when r e f e r e n t i a l questions and confirmation checks co-occur? Perhaps the most t y p i c a l way that p a r t i c i p a n t s work through conversational problems, that i s , problems over the meaning and i n t e r p r e t a t i o n of an utterance, i s a speaker's reformulation of an utterance i n response to a partner's i n d i c a t i o n of non-comprehension, and then commitment by that partner to a course of action which i s i n turn evaluated by the o r i g i n a l speaker. Variations of t h i s pattern allow i n s e r t i o n of additional repair cycles based on the evaluation: Was the action proposed by the partner (here, the learner) l i k e l y to s a t i s f y the speaker's (teacher's) understanding of a "correct next step"? Re c a l l i n g that the task now under consideration (LEG2) enta i l e d continuous feedback on the e f f i c a c y of the partner's actions i n 209 assembling a toy model, i t comes as no surprise that the p a r t i c i p a n t s enforced a rather severe standard of c l a r i t y i n order to accomplish the task successfully. The basic pattern i s i l l u s t r a t e d i n (31), below: (31) THAT'S GOING TO GO ON - THEE - SSECOND SSET + OF POINTS. /second ? CCONS Second? CCONS THE SECOND SET. + Like t h i s ? RRQS EEXOS /YEAH. - I THINK SO, YES. Eval.+ [7LEG2] The i n d i c a t i o n of trouble (from the learner's perspective) i s underlined and l a b e l l e d (CCONS—a confirmation check), as i s the learner's attempted solution (RRQS EEXOS). The teacher's p o s i t i v e evaluation of the attempt (Eval.+) i s also noted i n the margin of the excerpt. In t h i s case the teacher's i n i t i a l response to the learner's d i f f i c u l t y i s an other-expansion, which turns out to be j u s t enough information to help the learner i d e n t i f y the correct placement of the object. The learner's commitment to a course of action i s a much more d i r e c t method of obtaining an unambiguous evaluation than the confirmation check alone, a purely verbal t a c t i c . The commitment i s indicated by use of exophoric reference, a verbal t a c t i c which i s frequently accompanied by gestures such as pointing or touching. This i n t e n s i f i e d approach to generating evaluation gives the learner a powerful l e v e l , o f control over the q u a l i t y and 210 p r e c i s i o n o f f e e d b a c k , and u l t i m a t e l y p r o d u c e s a more e f f i c i e n t s o l u t i o n t o the p r o b l e m . T h i s p a t t e r n o f i n t e r a c t i o n , t h e n , i s marked by a v e r b a l t r o u b l e s i g n a l and a c o m b i n a t i o n o f v e r b a l and p h y s i c a l s i g n a l s d e s i g n e d t o d i r e c t t h e q u a l i t y o f the t e a c h e r ' s nex t u t t e r a n c e . The p a t t e r n becomes more complex , o f c o u r s e , when t h e p a r t i c i p a n t s f i n d t h a t the s i m p l e s t l e v e l o f exchange does no t c a r r y t h e t a s k f o r w a r d . E s s e n t i a l l y , the l e v e l o f c o m p l e x i t y — t h e dep th o f n e g o t i a t i o n — i s keyed t o t h e c a p a c i t y o f c o n f i r m a t i o n checks and r e f e r e n t i a l q u e s t i o n s t o c r e a t e l o o p s i n what would o t h e r w i s e be s t r a i g h t f o r w a r d movement f rom one s t e p o f the t a s k t o the n e x t . T h i s l e v e l o f c o m p l e x i t y i s e x e m p l i f i e d i n the f o l l o w i n g e x c e r p t . (32) NEXT, YOU WANT TO FIND - TWO - SMALL RECTANGULAR - YELLOW PIECES. R e c t a n g u l a r ? ! CCONS REC - RECTANGULAR, WHICH MEANS THAT THEY ARE - S - NOT SQUARE BUT LONG AND THIN. NOW + NO, NOT THOSE. THEY, E v a l . -/ I s t h i s ? RRQS EEXOS THEY'RE FLAT P IECES, THEY'RE •- FLAT -/ F l a t ? CCONS THEY'RE OF THE SAME TYPE OF SHAPE AS THE - BASE - OF THE + SO, / B a s e ? CCONS YES, BUT - SMALLER THAN THAT. YOU NEED E v a l . -/ I s t h i s ? RRQS EEXOS SMALL + SMALLER, S E E ? . YES. OKAY. NOW - E v a l . + / Y e s . [5LEG2] W h i l e t h e l e a r n e r ' s language i s not e l a b o r a t e , i t appears s u f f i c i e n t t o g e n e r a t e a h i g h l y r e s p o n s i v e s t ream o f 211 d i r e c t i o n s and evaluation. The end of t h i s c o r r e c t i v e sequence i s signaled by the learner's i d e n t i f i c a t i o n of the correct piece, very much l i k e the simpler pattern examined above (31). Accomplishing t h i s i d e n t i f i c a t i o n , however, e n t a i l s much greater e f f o r t by both p a r t i c i p a n t s : a candidate i d e n t i f i c a t i o n which, i n fact, f a i l s , a d d i t i o n a l confirmation checks to recycle the search for the correct piece, informative responses to each confirmation check, and a f i n a l , p o s i t i v e evaluation which allows a new step to be i n i t i a t e d . Overall, the impression i s one of quick recovery from the l o c a l ambiguities of the task, e f f e c t i v e verbal cooperation but, at the same time, a rather terse and unexpressive q u a l i t y i n the learner's language—perhaps a function of the high l e v e l of information a v a i l a b l e to the learner about objects i n the task s i t u a t i o n . I t may be that learners i n such information-rich s i t u a t i o n s are simply unchallenged to use language beyond the minimum l i m i t s of necessity. Although the basic forms of genuine information exchange are i n place, as evinced by the central p o s i t i o n of confirmation checks and r e f e r e n t i a l questions, the par t i c i p a n t s are not e s p e c i a l l y prepared to make conversation beyond the p r a c t i c a l requirements of the task. The kind of language which the learner may reasonably apply during construction of the model thus c l e a r l y contrasts with the r e l a t i v e l y more expressive language used by learners during ordinary discussion. Educational implications of 212 t h i s d i s t i n c t i o n w i l l be explored i n Chapter 7. For the moment, however, the focus i s on the qu a l i t y of information exchange which i s supported by confirmation checks and r e f e r e n t i a l questions during the face-to-face Lego task. Beyond the general issue of pa r t i c i p a n t s using confirmation checks to encourage a partner to provide a d d i t i o n a l or expanded information, there i s a s p e c i f i c function f o r r e f e r e n t i a l questions used with confirmation checks i n o u t l i n i n g the l i m i t s of the l o c a l problem on the which the pa r t i c i p a n t s are at work. This function resembles the " s e t t i n g up" of responses previously noted i n other face-to-face tasks (COM2 and DIS) i n which the question i s apparently intended to set l i m i t s on the forthcoming response. Learners who are able to d i r e c t t h e i r partners i n t h i s way can be said to be successful i n helping to manage the task, as excerpt (33) indicates. (33) UH BETWEEN THE TWO SQUARE ONES. ++ YOU CAN SNAP IT ON TOP OF THE TWO SQUARE ONES. Uh t h i s way? RRQS EEXOS NO, THE OTHER WAY. /NO. Eval.-The other way? CCONS EEXOS YEAH. Eval.+ OKAY. PUT IT ON TOP OF THEM. Top of them? CCONS YEAH. 0- ON TOP OF THE TWO SQUARE PIECES SO THE TW- SO YOU PUSH THE TWO SQUARE PIECES TOGETHER. Eval.+ 213 Th l i k e t h i s ? RRQS EEXOS [3LEG2] The learner's r e f e r e n t i a l questions bracket attempts (in the form of confirmation checks) to remove layers from the mystery of how one piece i s to be placed i n r e l a t i o n to the others. This process i s related to zeroing i n on the pos i t i o n i n g of pieces, not to the meaning of the teacher's i n i t i a l d i r e c t i v e nor to the learner's d i f f i c u l t y i n sharing the teacher's view of the pieces. In eithe r case, however, the p a r t i c i p a n t s are able to move more-or-less s u c c e s s f u l l y through the task by combining language and s i t u a t i o n a l r e f e r e n c e — t h e hallmark of problem-solving done under conditions of shared perception. The f i n a l l i n k between confirmation check and r e f e r e n t i a l question to be described here i s the learner's s e l f - r e p a i r which obviates intervention by the teacher to add or change information i n order to make the di r e c t i o n s more comprehensible. Repair occurs immediately following the learner's public demonstration of a problem. (34) PLEASE FIND A BLACK ONE WITH A TWO TUBES /Hm ON THE BOTH SIDES, AND WITH TWO BUMPS ON IT. Tubes? This one? CCONS RRQS /OKAY, YEAH, THAT'S RIGHT, THAT'S RIGHT. OKAY. Eval.+ [6LEG2] In p r i n c i p l e , e i t h e r of the par t i c i p a n t s has the ri g h t to r e p a i r under these circumstances, although, as has been 214 pointed out with reference to d e f i n i t i o n s following l e x i c a l uncertainty, i t i s the teacher who i s more l i k e l y to intervene during i n s t r u c t i o n a l tasks and make the r e p a i r (Kasper, 1985). LEG2, however, renders the problem and i t s s o l u t i o n as the central issue and tends to suppress the importance of status differences between teacher and learner. This may be a c h a r a c t e r i s t i c of problem-centered tasks i n general (although i t should be pointed out that no evidence has been presented on behalf of the back-to-back Lego task to support the wider p o s s i b i l i t y ) . The t r a n s c r i p t s reviewed thus far, however, suggest the c e n t r a l r o l e of confirmation checks i n i n d i c a t i n g trouble and of r e f e r e n t i a l questions i n pointing out a candidate s o l u t i o n during the performance of problem-centered tasks. That these functions are often exercised by the learner i n the same turn during the face-to-face Lego task further supports the view raised here that the learner i s l a r g e l y capable of asserting normal conversational r i g h t s given the appropriate task structure. Summary This chapter has extended the findings of the analysis of variance into two f i e l d s of q u a l i t a t i v e analysis: 1) a comparison of display and r e f e r e n t i a l questions, and 2) an examination of two sets of complementary task structures, including d e f i n i t i o n s and expressions of l e x i c a l uncertainty during the computer demonstration task, and confirmation checks with r e f e r e n t i a l questions during the face-to-face 215 Lego task. A number of functions were found to characterize teacher-learner t a l k during the computer demonstration task: display of the learner's knowledge presumably acquired as a r e s u l t of i n s t r u c t i o n , pointing-out of objects or operations the teacher found useful for moving smoothly through the i n s t r u c t i o n a l procedure, se t t i n g up behavior and leading the learner to "inescapable conclusions" the teacher considered e s s e n t i a l to furthering the goals of i n s t r u c t i o n , and pre-empting the learner's opportunities to ask and answer questions which could lead to negotiation over meaning. The primary verbal medium for accomplishing these functions was found to be various forms of the display question. Re f e r e n t i a l questions were found most frequently i n the discussion and face-to-face Lego tasks, both of which reduced the importance of the teacher-learner status gap and encouraged active exchange of information. Referential questions were found to serve a number of l o c a l functions, including encouraging a partner to expand material j u s t introduced into the t a l k , gradually developing a topic, nominating topics for consideration and developing them across a number of conversational turns. Participants used r e f e r e n t i a l questions during cooperative problem-solving, i n p a r t i c u l a r , to request i d e n t i f i c a t i o n of objects or r e l a t i o n s h i p s between objects, assessment of performance and perception, explanations and d i r e c t i o n s . The central function of r e f e r e n t i a l questions, however, was to a s s i s t 216 the exchange of information between partners of r e l a t i v e l y equal status i n the task s e t t i n g . Examination of the two sets of complementary task structures emphasized the co-occurrence of c e r t a i n REs. Although the learner's l e x i c a l uncertainty during the computer explanation task (C0M1) was t y p i c a l l y the tr i p w i r e fo r a teacher-made d e f i n i t i o n , the d e f i n i t i o n s themselves could become the subject of negotiation. When repa i r d i d occur, however, i t was often a case of the teacher taking r e s p o n s i b i l i t y for repairing the t a l k of both p a r t i c i p a n t s . Confirmation checks and r e f e r e n t i a l questions occurring together during the face-to-face problem-solving task (LEG2) c h a r a c t e r i s t i c a l l y opened up the t a l k to negotiation over both content and language. Moreover, by combining the use of confirmation checks and r e f e r e n t i a l questions, both p a r t i c i p a n t s attempted to repair t h e i r own utterances, a t y p i c a l feature of normal conversational behavior, within t h e i r own turns. F i n a l l y , while confirmation checks took on the key function of i n d i c a t i n g trouble, co-occurring r e f e r e n t i a l questions were used by next-speakers eit h e r as a pivot f o r further work on the problem or as an opportunity to s i g n a l that p a r t i c i p a n t s had achieved a common understanding and could move on to something else. Perhaps the most general conclusion which can be reached from t h i s q u a l i t a t i v e phase of the study i s that some tasks are better for teachers and others are better for learners; from an i n t e r a c t i o n a l viewpoint, some encourage 217 display of the teacher's competence while others promote expression of the learner's competence. I t should be also be noted that while a l l tasks produced negotiation over meaning, the q u a l i t y and extent of t h i s negotiation c l e a r l y v a r i e d with the task. In general, the teaching tasks required the learner to become a l a r g e l y passive r e c i p i e n t of sometimes abstract explanation and curbed timely opportunities f o r making sense of the shower of explanation to which the learner was sometimes exposed (Chaudron, 1983 notes a s i m i l a r pattern i n classroom behavior). I r o n i c a l l y , the non-teaching tasks achieved greatest e f f i c i e n c y when they were conducted i n apparently roundabout fashion—when the p a r t i c i p a n t s had to make several attempts to reach a working l e v e l of mutual comprehensibility. The two task structures c l e a r l y diverged i n terms of what s i g n a l l e d trouble and how p a r t i c i p a n t s resolved i t . In COMl, movement from l e x i c a l uncertainty to d e f i n i t i o n constituted the l e a s t responsive, l e a s t complex l e v e l of negotiated exchange found i n the t r a n s c r i p t s . In contrast, the a l t e r n a t i o n of confirmation checks and r e f e r e n t i a l questions were key features of the r e l a t i v e l y complex, open-ended and cooperative exchanges i n LEG2. This depiction of the two sets of repairs, i t should be stressed, i s based on examination of a r e l a t i v e l y small number of t r a n s c r i p t s (24 of the t o t a l of 60). The t r a n s c r i p t s display the . d i s t i n c t i o n with c l a r i t y , however, and, hopefully, make findings from the analysis of variance more tangible to the 218 educational p r a c t i t i o n e r . Accordingly, the f i n a l chapter which follows w i l l extend these findings into implications for educational p r a c t i c e . 219 CHAPTER 7: SUMMARY, LIMITATIONS, IMPLICATIONS AND CONCLUSION C h a p t e r 7 c o n c l u d e s t h i s s t u d y o f c o n v e r s a t i o n a l r e p a i r and r e f e r e n c e . S e v e r a l major p o i n t s have been made d u r i n g t h e c o u r s e o f t h e s t u d y ; t h e s e w i l l now be summarized. L i m i t a t i o n s o f the s tudy o r a r e a s i n which i t might be m i s i n t e r p r e t e d w i l l be d i s c u s s e d and i m p l i c a t i o n s f o r e d u c a t i o n a l p r a c t i c e which a r e based on the e m p i r i c a l a n a l y s e s w i l l be h i g h l i g h t e d . Summary The s t u d y was c o n c e i v e d i n o r d e r t o examine s y s t e m a t i c a l l y the ways i n which members o f n a t i v e and n o n -n a t i v e t e a c h e r - l e d dyads modi fy t h e i r t a l k i n E n g l i s h i n o r d e r t o a c h i e v e u n d e r s t a n d i n g . U n d e r l y i n g t h i s i n t e r e s t were t h e o b s e r v a t i o n s , s u p p o r t e d i n the l i t e r a t u r e o f second language a c q u i s i t i o n , t h a t the e x t e n t and q u a l i t y o f i n t e r a c t i o n a l m o d i f i c a t i o n v a r i e s w i t h t h e c h a r a c t e r i s t i c s o f t h e s e t t i n g i n which the t a l k o c c u r s and t h a t t h e g r e a t e r t h e o p p o r t u n i t i e s f o r m o d i f i c a t i o n , the more l i k e l y l e a r n e r s a r e t o a c q u i r e the t a r g e t l a n g u a g e . The t r a d i t i o n a l c l a s s r o o m - f r o n t e d b e h a v i o r o f language t e a c h e r s i n t h i s v iew o f second language a c q u i s i t i o n has been found a r e l a t i v e l y i n e f f i c i e n t means o f g u i d i n g language i n s t r u c t i o n i n c o m p a r i s o n w i t h l e a r n e r - l e a r n e r i n t e r a c t i o n . What had not been examined, however , was the s y s t e m a t i c t r e a t m e n t o f n a t i v e and n o n - n a t i v e t e a c h e r s as c o - p a r t i c i p a n t s i n d y a d i c 220 i n t e r a c t i o n with learners, and the e f f e c t s such treatment has upon the t a l k produced within the dyad. The terms used to support the study were drawn from research i n second language a c q u i s i t i o n , discourse analysis and language education, three f i e l d s with s p e c i f i c and somewhat d i f f e r e n t i n t e r e s t s i n the uses of language i n context. The research focused on the e f f e c t s of d i f f e r e n t tasks on the ways the p a r t i c i p a n t s i n teacher-led dyads repa i r conversational trouble and r e f e r to things i n t h e i r spoken texts. Underlying t h i s formulation was a small group of d i s t i n c t i o n s of p o t e n t i a l value i n educational planning and of immediate use i n construction of the research design: that t a s k s — t h e settings for use of repair and r e f e r e n c e — vary q u a l i t a t i v e l y from those which emphasize educational goals to those which emphasize s o c i a l goals, and from those which employ ex p e r i e n t i a l , "hands-on" processes i n the achievement of the objectives to those which emphasize the r o l e of exposition and explanation. These ways of looking at repair, task and reference were operationalized through the research design, s p e c i f i c a l l y through a series of analyses of variance which examined the difference between native and non-native groups i n the use of repair and reference during a series of f i v e tasks. A second type of analysis focused on a small group of r e p a i r exponents which were found to d i s t i n g u i s h most c l e a r l y among the tasks. The second analysis emphasized the forms and functions of repair i n context, 221 pointing d i r e c t l y to the utterances of the p a r t i c i p a n t s i n the task t r a n s c r i p t s . Before summarizing the major conclusions of the study, i t w i l l be useful to put the conceptual bases and the findings of the study into a common perspective. The Knowledge Framework o r i g i n a l l y developed by Mohan (1986) suggested a r e l a t i o n s h i p between t h e o r e t i c a l knowledge and expository a c t i v i t y on the one hand, and between p r a c t i c a l knowledge and e x p e r i e n t i a l a c t i v i t y on the other. An extension of the framework (Figure 2) was offered as a ten t a t i v e and p a r t i a l approach to specifying the c h a r a c t e r i s t i c s of tasks, reference and repair which might be expected when the framework i s applied to problems i n language education. Among the l i n k s between the extended framework and the r e s u l t s of the study are the unambiguous r e l a t i o n s h i p s found between tasks emphasizing p r a c t i c a l discourse, and a group of r e p a i r behaviors centering on questions intended by one p a r t i c i p a n t to obtain information or guidance from the other. As noted i n Chapter 6, for example, the r e f e r e n t i a l question serves numerous discourse functions, alone or i n combination with other behaviors. I t i s primarily, however, a means of opening t a l k to unforseen t o p i c a l development and to a more equal d i s t r i b u t i o n of power i n the discourse environment. The other side of the extended Knowledge Framework suggested a l i n k between expository approaches to teaching and the teacher's use of display questions to t e s t 222 knowledge and ensure movement towards predetermined i n s t r u c t i o n a l goals. Following t h i s suggestion, the study found s i g n i f i c a n t relationships between the teaching tasks and display question and, indeed, a nearly symmetrical r e l a t i o n s h i p between display questions and r e f e r e n t i a l questions i n the teaching and non-teaching tasks. Other l i n k s , and f a i l u r e s to l i n k , can be pointed out, p a r t i c u l a r l y the finding that exophoric reference provides unambiguous evidence for the operation of e x p e r i e n t i a l a c t i v i t y , but that anaphoric reference i s f r e e l y used i n nearly a l l task settings. Although the framework should be treated with caution by other researchers, i t has served the useful function of generating f o c a l points f o r the quantitative and q u a l i t a t i v e study of r e p a i r i n i n s t r u c t i o n a l settings. The following l i s t summarizes the most general conclusions yielded by t h i s dual approach to the research. Conclusions Based on the Analysis of Variance 1. Teacher-led, dyadic tasks d i f f e r s i g n i f i c a n t l y depending on whether they are organized to support primarily educational or s o c i a l objectives. Repair and reference are c l e a r l y a l l o c a t e d to t h i s i n i t i a l d i v i s i o n of the task factor: d e f i n i t i o n s , display questions and echoes to the teaching tasks, c l a r i f i c a t i o n requests, comprehension checks, confirmation checks, expressions of l e x i c a l uncertainty, r e f e r e n t i a l questions, self-expansions, s e l f -r e p e t i t i o n s , and anaphoric reference and exophoric reference 223 to the non-teaching tasks. Tasks which are e x p l i c i t l y organized to e f f e c t i n s t r u c t i o n a l goals are thus r e l a t i v e l y l e s s l i k e l y to r e f l e c t the breadth and q u a l i t y of negotiation which characterizes tasks oriented towards achievement of s o c i a l goals. 2. The preference for s e l f - r e p a i r , as d i s t i n c t from other-repair, i s compelling and active when groups are organized as teacher-learner dyads. Among the 12 r e p a i r exponents, only those which highlighted the behavior of the other member of the dyad (other-repetition and other-expansion) were without s i g n i f i c a n t e f f e c t s i n any of the tasks. While the l i t e r a t u r e suggests that ordinary teacher-fronted language i n s t r u c t i o n encourages other-repair (largely r e p a i r of the learner's t a l k ) , dyadic i n t e r a c t i o n which i s not focused on the target language, even when oriented towards communicating i n s t r u c t i o n a l goals, does not support other-repair. Tasks which have no s p e c i a l focus on language i n s t r u c t i o n thus more c l o s e l y resemble normal conversational behavior i n terms of the preference f o r s e l f -r e p a i r . 3. In general, the l e a s t important d i s t i n c t i o n s i n the study were rel a t e d to the group factor. Given at l e a s t a professional l e v e l of competence i n English, Japanese teachers are at no special disadvantage over t h e i r native English-speaking counterparts i n working with learners i n dyadic task settings. This conclusion i s based on the v i r t u a l l y indistinguishable types and l e v e l s of repair and 224 reference produced by dyads i n which English language p r o f i c i e n c y was controlled. 4 . Like repair, reference i s allocated to tasks, although the es s e n t i a l d i s t i n c t i o n between tasks s h i f t s from a teaching-non-teaching dimension to an e x p e r i e n t i a l -expository dimension. Concern with the here-and-now and the sharing of perception as a conversational resource produce an intense use of exophoric reference. Exophoric reference serves as a signature of exp e r i e n t i a l conversational a c t i v i t y , and i s thus distinguished from anaphora which becomes a s i g n i f i c a n t resource only when normal access to the speech s i t u a t i o n has been cut o f f . In t h i s sense anaphoric reference serves to manage reduced contextualization and b u i l d l i n k s across the discourse as they are needed. 5. The cle a r e s t , l e a s t q u a l i f i e d a l l o c a t i o n of rep a i r to task occurs when tasks contain both goal and process dimensions, that i s , when a p a r t i c u l a r combination of social-educational and experiential-expository values has been applied to planning the task. The i n t e r s e c t i o n of s o c i a l goals and experi e n t i a l a c t i v i t y produces the most negotiation and repair i n teacher-led dyads (as measured by the mean frequency of confirmation checks and r e f e r e n t i a l questions) and thus resembles conversational behavior outside of t r a d i t i o n a l teacher-fronted classrooms. When educational objectives and expository a c t i v i t y are emphasized, dyadic t a l k i s oriented to t r a n s f e r r i n g 225 cognitive knowledge which the teacher possesses p r i o r to beginning the task to a learner who i s assumed to be a naive p a r t i c i p a n t . The c h a r a c t e r i s t i c repair behaviors which occur during t h i s kind of task include frequent d e f i n i t i o n s and expressions of l e x i c a l uncertainty—forms of verbal behavior of p a r t i c u l a r use to teachers and learners i n t r a d i t i o n a l classrooms. Conclusions Based on the Analysis of Transcripts 6. Display questions, e s s e n t i a l t o o l s of teaching s i t u a t i o n s , serve a v a r i e t y of functions r e l a t e d to t r a n s f e r of knowledge and control over the learner's opportunities to d i r e c t the t a l k . Beyond simply t e s t i n g the learner's knowledge, display questions also function to cut through ambiguity which might otherwise have to be negotiated, lead the learner d i r e c t l y to "inescapable conclusions" and pre-empt challenges (intended and unintended) to the d i r e c t l i n e the teacher has apparently already developed towards the goals of i n s t r u c t i o n . Within a teacher-fronted frame of reference with e x p l i c i t i n s t r u c t i o n a l goals (as was found here i n the computer demonstration task), display questions arguably improve the e f f i c i e n c y with which the i n s t r u c t i o n i s delivered. Unfortunately, however, t h i s kind of e f f i c i e n c y appears to have l i t t l e e f f e c t on enriching the i n t e r a c t i o n a l q u a l i t y of t a l k conducted between teachers and learners. 7. In tasks which are oriented toward expository communication of educational objectives, d e f i n i t i o n i s a 226 c e n t r a l concern of teachers. D e f i n i t i o n may anti c i p a t e or be triggered by the learner's l e x i c a l uncertainty. In ei t h e r case, d e f i n i t i o n can become a negotiable resource unless the teacher d i r e c t s or co-opts the learner's responses i n aid of movement towards an i n s t r u c t i o n a l goal. Comparatively viewed ( i . e . , viewed from the perspective of tasks allowing more-or-less equivalent r i g h t s to p a r t i c i p a t i o n ) , d e f i n i t i o n co-occurring with l e x i c a l uncertainty i s a useful although weak source f o r negotiation of meaning since i t depends l a r g e l y on the teacher's program for operation of the task. 8. Referential questions are a central feature of non-i n s t r u c t i o n a l , face-to-face t a l k . Beyond the general function of opening up the t a l k to negotiation over both language and content, r e f e r e n t i a l questions bring explanation and d i r e c t i o n into the discourse on an ad hoc basis. In conjunction with confirmation checks, r e f e r e n t i a l questions can also be used to e l i c i t evaluation, focus on objects or operations i n the speech s i t u a t i o n and undertake rep a i r . Exercise of these functions frequently means that information exchanged by the pa r t i c i p a n t s follows an i n d i r e c t pattern e n t a i l i n g evaluation, backtracking, r e v i s i o n and expansion before the conversation moves on to new top i c s . I t i s p r e c i s e l y t h i s p o t e n t i a l for roundabout pursuit and exchange of information, for recursive and open-ended i n t e r a c t i o n , which makes non-teaching tasks which employ some exp e r i e n t i a l processes prospectively useful 227 settings f o r a c q u i s i t i o n of another language. These conclusions outline the substance of the empirical study but they also point back to research which illuminates the p o s i t i o n of foreigner t a l k and interlanguage t a l k i n second language classrooms. Chaudron (1983), Long & Sato (1983) and Wesche & Ready (1985), for example, pointed out the r e l a t i v e i n e f f i c i e n c y of FT i n t r a d i t i o n a l , teacher-fronted classrooms. In a s i m i l a r vein, t h i s study found that the teaching tasks encouraged both the foreign and Japanese teachers to use such REs as display questions and d e f i n i t i o n s when attempting to accomplish t h e i r objectives and that the learners tended to support the teachers i n t h i s behavior by concerning themselves with demonstrations of attention, by use of echoes, for example. Studies on the functional properties of IT, on the other hand, have shown that NNS-NNS in t e r a c t i o n i n shared-goal environments produces re p a i r behavior comparable to and about as varied as NS-NSS i n t e r a c t i o n i n conversational settings (Duff, 1986; Porter, 1983; Porter and Long, 1985). The cooperative and information-exchange q u a l i t i e s of the non-teaching tasks i n t h i s study seem to have produced s i m i l a r r e s u l t s , p a r t i c u l a r l y with respect to such repairs as c l a r i f i c a t i o n requests, confirmation checks and r e f e r e n t i a l questions, although i t should be recognized that the NNSs i n t h i s study were Japanese teachers who were able to serve as sources of high-quality input. Although cooperation was evident i n a l l of the task 228 c o n d i t i o n s , v e r y d i f f e r e n t forms o f c o o p e r a t i o n p r e v a i l e d i n t h e two complementary t a s k s t r u c t u r e s o u t l i n e d i n Chapte r 5. Beyond t h e r e s e a r c h f i n d i n g s d i s c u s s e d e a r l i e r , the e s s e n t i a l c o n c l u s i o n o f the s tudy i s t h a t language t e a c h e r s — n a t i v e o r n o n - n a t i v e — a n d l e a r n e r s i n d y a d i c s e t t i n g s can n e g o t i a t e t h e i r d i s c o u r s e when t h e y have been o f f e r e d t h e r i g h t c o n d i t i o n s : an o r i e n t a t i o n t o c o n t e n t more t h a n l a n g u a g e , o b j e c t i v e s which a r e t r e a t e d as more s o c i a l t h a n e d u c a t i o n a l , and d i r e c t ( a l t h o u g h not u n l i m i t e d ) e x p e r i e n c e w i t h t h e o b j e c t s o f t h e i r t a l k . L i m i t a t i o n s o f the Study B e f o r e g o i n g on t o p o s s i b l e a p p l i c a t i o n s f o r t h e s t u d y , i t w i l l be u s e f u l t o p o i n t out some o f i t s l i m i t a t i o n s o r a r e a s i n which i t ought t o be c l a r i f i e d . As w i l l soon become e v i d e n t , t h e r e i s a c l o s e r e l a t i o n s h i p between p o i n t i n g out the l i m i t a t i o n s o f a s tudy and c o n s i d e r i n g ways i n wh ich o t h e r s may want t o ex tend i t . The f i r s t q u e s t i o n t o be r a i s e d h e r e i s how f a r one can g e n e r a l i z e from f i n d i n g s a c h i e v e d i n q u a s i - e x p e r i m e n t a l s t u d i e s t o t h e problems o f p r a c t i c e which o c c a s i o n such s t u d i e s i n t h e f i r s t p l a c e . A l t h o u g h the conduc t o f t h i s i n v e s t i g a t i o n l i e s w i t h i n a growing SLA r e s e a r c h t r a d i t i o n — s t a t i s t i c a l t r e a t m e n t o f c o n v e r s a t i o n a l d a t a g e n e r a t e d from groups o r g a n i z e d s p e c i f i c a l l y t o demonst ra te b e h a v i o r s o f i n t e r e s t t o t h e r e s e a r c h e r — i t s f i n d i n g s do no t t h e r e f o r e g a i n a u t o m a t i c a p p l i c a b i l i t y t o the w o r l d o f e d u c a t i o n a l p r a c t i c e . 229 As i n a l l cases i n which control over a large number of va r i a b l e s i s of importance to the researcher, l i n k s between the research design and educational practice are worth noting. For example, most of the tasks comprising the task factor are s i m i l a r to those which language teachers could employ i n a task-based syllabus (see, for example, Brown & Yule, 1983; Brumfit, 1986; Littlewood, 1981; Long, 1985b). A wide range of tasks, i n terms of content and focus, i s made possible when the teacher releases control over language and instead expends e f f o r t on organizing learners to solve problems or exchange information. To the extent that the reporting of the study may have obscured some of the l i n k s with educational practice, i t i s also worth pointing out that choice of several tasks was based on preliminary observations of NNS-NNS (Japanese-non-Japanese) teacher-learner i n t e r a c t i o n devoted to technology transfer. This point i s expanded somewhat i n the next section which outlines implications for educational p r a c t i c e . Another issue r e l a t e d to g e n e r a l i z a b i l i t y of the research findings i s the r e l a t i o n s h i p of method, of subject s e l e c t i o n or sampling procedures, for instance, to problems of p r a c t i c e . One area which has not received much discussion i n t h i s study i s the g e n e r a l i z a b i l i t y of r e s u l t s obtained through the q u a l i t a t i v e analysis, i n p a r t i c u l a r the bases f o r s e l e c t i o n and presentation of excerpts used to depict s p e c i f i c patterns of repair. A purely ideographic approach to conversational data would not have been 230 e s p e c i a l l y concerned with systematic sampling of data at any l e v e l i n the analysis nor with attempting to generalize beyond the defined context; a purely nomothetic (or "law-seeking") approach would have avoided opportunistic sampling altogether and would have emphasized the systematicity by which the data were sampled and examined ( B u r r e l l & Morgan, 1979). Rec a l l i n g that the text sampling procedure was something of a compromise between these methodological extremes, the study may have turned out a serie s of interp r e t a t i o n s which are neither a r t nor science and which may therefore f a i l to convince at ei t h e r l e v e l . Beyond simply accepting the q u a l i t a t i v e analysis on i t s own merits, however, there i s always the option f o r other researchers to re-examine and make sense of the o r i g i n a l data using the same assumptions which guided the f i r s t a n a l y s i s . This could mean i n practice, f o r example, using the o r i g i n a l repair exponents and re-examining the o r i g i n a l sets of t r a n s c r i p t s i n order to produce a comparable i n t e r p r e t a t i o n of the data. Rather than attempting to r e p l i c a t e the r e s u l t s i n t h i s way, however, i t might be a great deal more i n t e r e s t i n g to t r e a t the q u a l i t a t i v e analysis as a general format f o r examination of data generated by s i m i l a r groups i n s i m i l a r conversational se t t i n g s . What i s emphasized i n t h i s view of the analysis i s not so much the findings i n the o r i g i n a l study but the uses to which i t s methods can be put i n further studies. The l a s t l i m i t a t i o n taken up here i s the p r a c t i c a l 231 problem of teachers serving as partners i n classes with large numbers of learners, a problem which has not been c l a r i f i e d adequately thus f a r . Although t h i s study has argued that teacher-learner dyadic i n t e r a c t i o n may be a useful a l t e r n a t i v e to teacher-fronted whole-classroom i n s t r u c t i o n , r e l iance on teacher-learner groups to the exclusion of other organizational forms i s u n l i k e l y to be an e f f i c i e n t use of the teacher's time. One of the e x p l i c i t assumptions behind the study was that a l t e r n a t i v e r o l e s f o r teachers i n foreign language education had not been adequately explored i n the l i t e r a t u r e . Many teachers and the educational systems which support them would be reluctant to give up d i r e c t , teacher-learner contact during the course of an i n s t r u c t i o n a l program i n favor of a syllabus organized around learner-learner i n t e r a c t i o n . Nothing i n t h i s study should be taken to suggest that teacher-led dyads are the only or best approach to foreign language i n s t r u c t i o n , however. On the contrary, while consideration of alter n a t i v e s among p o s s i b i l i t i e s f o r teacher-learner i n t e r a c t i o n served to motivate the study, i t should be stressed that opportunities f o r NNS-NNS exchange i n classroom settings can take many useful forms beyond the cooperative p a r t i c i p a t i o n of NNS teachers and learners i n dyadic groups. Implications f o r Educational Practice What does the study o f f e r f o r those who plan and conduct foreign language education? 232 F i r s t , the r e s u l t s support the capacity of non-native teachers with a professional l e v e l of competence i n the target language to serve as co-participants with learners i n task-oriented groups. The non-native teachers i n t h i s study were l a r g e l y indistinguishable from the native English speaking teachers i n terms of t h e i r use of reference and r e p a i r . The r e s u l t s suggest, for example, that non-native teachers are able to employ what Tarone (1983) and others (Bialystock, 1983; Faerch & Kasper, 1983) have c a l l e d s t r a t e g i c competence i n a v a r i e t y of discourse settings, and that learners do not therefore s u f f e r from exposure to t h e i r non-native teachers. At the same time, i t appears that the threshold of s t r a t e g i c competence for learners to employ various repairs i n a foreign language u s e f u l l y may be r e l a t i v e l y low. The lowest common denominator i n t h i s study, the "intermediate l e v e l " learners, comprehensibly requested c l a r i f i c a t i o n or expanded t h e i r utterances, for example, equally well with t h e i r Japanese and native i n t e r l o c u t o r s . A further, and related, a p p l i c a t i o n for the r e s u l t s i s a measure of support for t r e a t i n g English as an i n t e r n a t i o n a l language (Smith, 1981) which does not require (but c l e a r l y does not exclude) the input of native speakers. Even though a large number of i n s t r u c t i o n a l s i t u a t i o n s around the world are quite obviously successful examples of content taught and learned by non-natives through the medium of English, there i s s t i l l considerable controversy about the q u a l i t y of 233 l anguage l e a r n e d from n o n - n a t i v e s . T h i s s t u d y , l i m i t e d t o t h e compar ison o f s e v e r a l forms o f r e p a i r and r e f e r e n c e between n a t i v e - l e d and n o n - n a t i v e l e d g r o u p s , found no b a s i s f o r e x c l u d i n g n o n - n a t i v e t e a c h e r s from d y a d i c exchange on t h e grounds t h a t t h e i r b rand o f E n g l i s h i s somehow a s e c o n d -r a t e s o u r c e o f i n p u t . A t h i r d i m p l i c a t i o n r e l a t e s t o t h e c h o i c e o f t a s k s f o r o r g a n i z i n g d y a d i c i n t e r a c t i o n . The f i n d i n g s show t h a t a v a r i e t y o f t a s k s can s u p p o r t n e g o t i a t i o n and r e p a i r , a l t h o u g h t h e q u a l i t y and e x t e n t o f n e g o t i a t i o n i s c l e a r l y i n f l u e n c e d by the t a s k s e l e c t e d . Because the v a l u e s o f the e d u c a t i o n a l system i n which t e a c h e r s and o t h e r p l a n n e r s o p e r a t e e v e n t u a l l y f i n d t h e i r way i n t o i n s t r u c t i o n a l p r a c t i c e , i t may be u n r e a s o n a b l e t o expec t c o n s e r v a t i v e f o r e i g n language e d u c a t i o n a l systems t o t r e a t t e a c h e r s and l e a r n e r s as e q u a l p a r t n e r s i n t a s k - b a s e d i n t e r a c t i o n d e s i g n e d f o r c l a s s r o o m s e t t i n g s . The image and p r a c t i c e o f t e a c h e r - f r o n t e d , t e s t - o r i e n t e d f o r e i g n language i n s t r u c t i o n a r e e s p e c i a l l y c o m p e l l i n g i n J a p a n , f o r example , and i t i s o n l y i n t h e a d u l t (and t y p i c a l l y p r i v a t e ) f o r e i g n language system t h a t t e a c h e r s and l e a r n e r s b e g i n t o t h i n k o f a f o r e i g n language as a medium o f i n t e r p e r s o n a l communica t ion . Indeed , a l l o f the Japanese t e a c h e r s employed i n t h i s s t u d y were v e t e r a n s o f b o t h the f o r m a l f o r e i g n language i n s t r u c t i o n a l s y s t e m , w i t h i t s emphasis on r o t e l e a r n i n g and e x a m i n a t i o n s , and o f n o n - t r a d i t i o n a l r o u t e s t o f l u e n c y i n E n g l i s h , i n c l u d i n g a d u l t c l a s s e s , E n g l i s h s p e a k i n g c l u b s and 234 educational experiences overseas. The most e f f e c t i v e model for classroom language learning, then, i s probably not to be found i n t r a d i t i o n a l language classrooms at a l l . I t i s more l i k e l y to be found i n content-area i n s t r u c t i o n and s o c i a l exchanges conducted i n non-school environments—at worksites, conferences, meetings and other places i n which NNSs gather to obtain expertise, exchange information or solve problems. This model for the exchange of content can be applied i n foreign language classrooms by t r e a t i n g tasks as environments which allow d i f f e r e n t facets of content-area problems to be explored at d i f f e r e n t points i n the syllabus. Some tasks are demonstrably more appropriate f o r t r e a t i n g information as a p r a c t i c a l resource, others as a t h e o r e t i c a l resource. I t i s e n t i r e l y possible, for example, to design a syllabus i n which teachers i n i t i a t e a cycle of tasks with r e l a t i v e l y undemanding experi e n t i a l a c t i v i t i e s allowing for a high l e v e l of shared information, move on to a c t i v i t i e s which make more performance demands on the learner and then conclude with t h e o r e t i c a l generalization and explanation. Within l i m i t s , the teacher's r o l e can be adjusted to s u i t c u l t u r a l expectations—the teacher can, a f t e r a l l , e x p l i c i t l y d i r e c t i n s t r u c t i o n i n the dyad (or, for that matter, commission learner-learner groups)—and thus design opportunities to t r e a t content i n d i f f e r e n t ways. The point i s that choice of task c a r r i e s consequences, some of which have been elaborated here, for the kind and quantity of 235 i n t e r a c t i o n i n which the learner i s going to p a r t i c i p a t e . Implications for Task-based Research Even though the present discussion i s about the ap p l i c a t i o n of research to educational practice, there i s one f i n a l implication of the study which necessarily returns the discussion to the conduct of task-based research. A major problem for anyone undertaking applied research i s the development of categories which capture some of the r e a l i t y of an applied world: teachers, learners, classrooms, tasks, c l a r i f i c a t i o n requests, exophoric reference, and so on. In general, the categories i n t h i s study r e f l e c t d i s t i n c t i o n s which are v i r t u a l l y i d e n t i c a l to those employed by other researchers. I t makes some sense to b u i l d on a body of published research by appropriate a p p l i c a t i o n of i t s categories and frames of reference rather than pioneer an e n t i r e l y novel set of categorical d i s t i n c t i o n s . A l t e r n a t i v e , inductive approaches to generating a n a l y t i c categories (see, f o r example, Schwartz, 1980; Gaies, 1983) have at l e a s t as much r e s p e c t a b i l i t y i n the t r a d i t i o n of s o c i a l sciences, however, and should be considered as a useful means of establishing v a l i d markers through the r e l a t i v e l y unexplored t e r r i t o r y of task-based research. To some extent the q u a l i t a t i v e analysis of Chapter 6 was an attempt to move somewhat beyond the currently used d e s c r i p t i v e frames. This phase of the study should be considered an extension of hypothesis t e s t i n g accomplished through quantitative analysis, however, rather than what i t 236 might have been under an e n t i r e l y d i f f e r e n t research d e s i g n — a n attempt to b u i l d new frames of reference and categories for the study of dyadic i n t e r a c t i o n between non-native speakers of English. Further studies of NNS-NNS discourse can stake out a middle ground between the extremes of deductive and inductive approaches to treatment of data. They can, for example, s t a r t with some rather broad categories which have grown out of empirical research or others' attempts at model-building, construct a set of narrower categories on the basis of a q u a l i t a t i v e study and then go on to suggest or t e s t a manageable set of hypotheses. The simple, two-dimensional framework outlined i n t h i s study, i t s e l f based on a two-factor a c t i v i t y framework for educational settings, can accommodate a v a r i e t y of novel categories which have yet to be developed i n future observational research. The notion of complementary task structures can be used to s e l e c t rough observational boundaries and a l l o c a t e behavioral categories which emerge during analysis of conversational data. The framework i t s e l f can be validated; i t should c e r t a i n l y be modified to allow for more accurate d e s c r i p t i o n of discourse i n educational settings. Conclusion The l i m i t a t i o n s and implications of the study present an o u t l i n e of areas into which others may extend t h e i r own research. The r o l e of classroom teachers i n t h i s prospective research process has only been implied; i t ought 237 to be made e x p l i c i t . Considering the r e s p o n s i b i l i t y and c e n t r a l p o s i t i o n of teachers i n foreign language education, there are very few others who are i n a better p o s i t i o n to pose good questions about t h e i r work, to seek an understanding of the verbal processes which occur within t h e i r classrooms or i n related experimental settings and to o f f e r explanations with a basis i n the r e a l i t y of professional pr a c t i c e . I f t h i s study has helped foreign language teachers to view t h e i r teaching from a fresh perspective i t w i l l have f u l f i l l e d one of i t s goals. I f , i n addition, i t encourages teachers to examine t h e i r work from a research perspective, i t w i l l have achieved an unanticipated bonus. The focus of the study has been sharpened by the a v a i l a b i l i t y of conceptual thinking accomplished by others. The debt i s recognized and a hope expressed that the study w i l l prove to be of both technical and conceptual value to others i n t h e i r pursuit of solutions to problems i n language education. 238 REFERENCES Abunahleh, L., Al l e n , S., Arthur, B., Beals, S., Butler, M., Drezner, B., Frydenberg, G., Galal, M., Gass, S., Hildebrandt, K., Marios, L., & Ostrander, T. (1982). The scope and function of language re p a i r i n foreigner discourse. Interlanguage Studies B u l l e t i n . 112-120. Bachman, L. F., & Palmer, A. S. (1981). The construct v a l i d a t i o n of the FSI or a l interview. Language Learning. 31. 67-86. Bi y a l i s t o c k , E. 1983. Some factors i n the s e l e c t i o n and implementation of communication strategies. In C. Faerch & G. Kasper (Eds.), Strategies i n interlanguage communication (pp. 100-118). London: Longman. Breen, M. (1987a). Contemporary designs i n syllabus design: Part I. Language Teaching. A p r i l , 81-92. Breen, M. (1987b). Contemporary designs i n syllabus design: Part I I . Language Teaching. July, 157-174. Brown, G., & Yule, G. (1983a). Discourse analysis. Cambridge: Cambridge University Press. Brown, G., & Yule, G. (1983b). Teaching the spoken language: An approach based on the analysis of conversational English. Cambridge: Cambridge University Press. Brulhart, M. M. (1985). Foreigner t a l k i n the ESL classroom: Interactional adjustments to adult students at two language p r o f i c i e n c y l e v e l s . Unpublished master's t h e s i s , University of B r i t i s h Columbia, Vancouver. B u r r e l l , G., & Morgan, G. 1979. So c i o l o g i c a l paradigms and organizational analysis: Elements of the sociology of corporate l i f e . London: Heinemann Educational Books. Campbell, C., G a s k i l l , W., & Vanderbrook, S. (1977). Some aspects of foreigner t a l k . In C. R. Henning (Ed.), Proceedings of the F i r s t Los Angeles Second Language Research Forum. Los Angeles: University of C a l i f o r n i a . Cazden, C. (1972). The s i t u a t i o n : A neglected source of s o c i a l c l a s s differences i n language use. In J . B. Pride & J . Holmes (Eds.), S o c i o l i n g u i s t i c s (pp. 294-313). Harmondsworth, England: Penguin Books. Chaudron, C. (1983). Foreigner t a l k i n the classroom—An aid to learning? In H. Seliger & M. H. Long (Eds.), Classroom oriented research i n second language a c q u i s i t i o n (pp. 127-143). Rowley, MA: Newbury House. 239 Clearlake Research. (1985). CLR ANOVA [Computer program]. Houston, TX: Author. Crookes, G. (1986). Task c l a s s i f i c a t i o n : A cross- d i s c i p l i n a r y review (Tech. Rep. No. 4). Manoa: Univer s i t y of Hawaii Social Science Research I n s t i t u t e , Center f o r Second Language Classroom Research. Cummins, J . (1983). Language prof i c i e n c y and academic achievement. In J . O i l e r (Ed.), Issues i n language t e s t i n g research (pp. 108-129). Rowley, MA: Newbury House. Doughty, C., & Pica, T. (1986). "Information gap" tasks: Do they f a c i l i t a t e second language acquisition? TESOL Quarterly. 20. 305-325. Duff, P. A. (1986). Another look at interlanguage t a l k : Taking task to task. In R. R. Day (Ed.), Talking to learn: Conversations i n second language a c q u i s i t i o n (pp. 147-181). Rowley, MA: Newbury House. E l l i s , R. (1984). Classroom second language development. Oxford: Pergamon. E l l i s , R. (1985). Understanding second language a c q u i s i t i o n . Oxford: Oxford University Press. Elmes, D. G., Kantowitz, B. H., & Roediger, H. L. (1985). Research methods i n psychology. St. Paul, MN: West Publishing Company. Educational Testing Service. (1982). ETS Oral Proficiency Testing Manual. Princeton, NJ: Author. Faerch, C , & Kasper, G. (1983). Plans and strategies i n foreign language communication. In C. Faerch & G. Kasper (Eds.), Strategies i n interlanguage communication (pp. 20-60). London: Longman. Feldman, D. S., J r . , & Gagnon, J . (1986). StatView 512+ [Computer program]. Calabasas, CA: Abacus Concepts. Ferguson, C. A. (1971). Absence of copula and the notion of s i m p l i c i t y . In D. Hymes (Ed.), P i d g i n i z a t i o n and c r e o l i z a t i o n of language (pp. 141-150). London: Cambridge University Press. Ferguson, C. A. (1975). Towards a characterization of English foreigner t a l k . Anthropoligical L i n g u i s t i c s , 17, 1-14. Ferguson, G. A. (1981). S t a t i s t i c a l analysis i n psychology and education (3rd ed.). Tokyo: McGraw-Hill. 240 Freed, B. F. (1978). Foreigner t a l k : A study of speech adjustments made by native speakers of English i n conversation with non-native speakers. Unpublished doctoral d i s s e r t a t i o n , University of Pennsylvania, Philadelphia. Freed, B. F. (1980). Talking to foreigners versus t a l k i n g to children: S i m i l a r i t i e s and differences. In R. C. S c a r c e l l a & S. D. Krashen (Eds.), Research i n second language a c q u i s i t i o n : Selected papers of the Los Angeles Second Language Ac q u i s i t i o n Research Forum (pp. 19-27). Rowley, MA: Newbury House. Gaies, S. J . (1982). Native speaker - nonnative speaker i n t e r a c t i o n among academic peers. Studies i n Second Language Acquisition , 5, 74-81. Gaies, S. J . (1983). Learner feedback: An exploratory study of i t s r o l e i n the second language classroom. In W. H. Se l i g e r & M. H. Long (Eds.), Classroom oriented research i n language a c q u i s i t i o n (pp. 190-212). Rowley, MA: Newbury House. Gass, S. M., & Varonis, E. M. (1984). The e f f e c t of f a m i l i a r i t y on the comprehensibility of non-native speech. Language Learning, 34., 65-89. Gass, S. M., & Varonis E. M. (1985a). V a r i a t i o n i n native speaker speech modification to non-native speakers. Studies i n Second Language Acquisition , 7, 37-57. Gass, S. M., fie Varonis, E. M. (1985b). Task v a r i a t i o n and nonnative/nonnative negotiation of meaning. In S. M. Gass & C. G. Madden (Eds.), Input i n second language a c q u i s i t i o n (pp. 149-161). Rowley, MA: Newbury House. Gass. S. M., & Varonis, E. M. (1986). Sex differences i n NSS/NSS intera c t i o n s . In R. R. Day (Ed.), Talking to learn: Conversation i n second language a c q u i s i t i o n (327-351). Rowley, MA: Newbury House. Goetz, J . P., & LeCompte, M. D. (1984). Ethnography and q u a l i t a t i v e design i n educational research. Orlando, FL: Academic Press. Haastrup, K., & P h i l l i p s o n , R. (1983). Achievement strategies i n learner/native speaker i n t e r a c t i o n . In C. Faerch & G. Kasper (Eds.), Strategies i n interlanguage communication (pp. 14 0-158). London: Longman. Halliday, M. A. K., & Hasan, R. (1976). Cohesion i n English. London: Longman. 241 H a r r i s , D. P . , & Pa lmer , L . A . (1986a) . CELT e x a m i n e r 1 s i n s t r u c t i o n s and t e c h n i c a l manual . New Y o r k : McGraw-H i l l . H a r r i s , D. P . , & Pa lmer , L . A . (1986b). CELT: Form A t e s t book . New York : M c G r a w - H i l l . H a t c h , E . (1983) . P s y c h o l i n g u i s t i c s : A second language p e r s p e c t i v e . Rowley, MA: Newbury House. H e n z l , V . M. (1974) . L i n g u i s t i c r e g i s t e r o f f o r e i g n language i n s t r u c t i o n . Language L e a r n i n g . 23., 207-222. H e n z l , V . M. (1979) . F o r e i g n t a l k i n the c l a s s r o o m . I n t e r n a t i o n a l Review o f A p p l i e d L i n g u i s t i c s . 17, 159-167. H o l d e r , W. (1982) . The WORD p l u s [Computer p r o g r a m ] . San D i e g o , CA: O a s i s Systems. Huck, S . W. , C o r m i e r , W. H . , & Bounds, W. G . , J r . (1974) . Read ing s t a t i s t i c s and r e s e a r c h . New Y o r k : Harper and Row. Hymes, D. (1967) . Models o f the i n t e r a c t i o n o f language and s o c i a l s e t t i n g . J o u r n a l o f S o c i a l I s s u e s . 2 7 ( 2 ) , 8 -28 . Hymes, D. (1972) . On communicat ive competence. In J . B. P r i d e & J . Holmes ( E d s . ) , S o c i o l i n g u i s t i c s (269-293) . Harmondsworth, E n g l a n d : Penguin Books . K a s p e r , G . (1985) . R e p a i r i n f o r e i g n language t e a c h i n g . S t u d i e s i n Second Language A c q u i s i t i o n , 7 , 200-215. Kramsch, C . J . 1985. C l a s s r o o m i n t e r a c t i o n and d i s c o u r s e o p t i o n s . S t u d i e s i n Second Language A c q u i s i t i o n , 7, 169-183. K r a s h e n , S . (1978) . The m o n i t o r model f o r second language a c q u i s i t i o n . In R. G i n g r a s ( E d . ) , Second language a c q u i s i t i o n and f o r e i g n language t e a c h i n g (pp. 1 -27 ) . A r l i n g t o n , VA: C e n t e r f o r A p p l i e d L i n g u i s t i c s . K r a s h e n , S . (1980) . The i n p u t h y p o t h e s i s . In J . E . A l a t i s ( E d . ) , C u r r e n t i s s u e s i n b i l i n g u a l e d u c a t i o n (pp. 144-158) . Wash ing ton , DC: Georgetown U n i v e r s i t y P r e s s . K r a s h e n , S . (1981) . Second language a c q u i s i t i o n and second language l e a r n i n g . O x f o r d : Pergamon P r e s s . K r a s h e n , S . (1982) . P r i n c i p l e s and p r a c t i c e i n second language a c q u i s i t i o n . O x f o r d : Pergamon P r e s s . 242 L i t t l e w o o d , W. (1981) . Communicat ive language t e a c h i n g . Cambr idge: Cambridge U n i v e r s i t y P r e s s . L o n g , M. H. (1980) . I n p u t , i n t e r a c t i o n , and second language a c q u i s i t i o n . U n p u b l i s h e d d o c t o r a l d i s s e r t a t i o n , U n i v e r s i t y o f C a l i f o r n i a a t Los A n g e l e s . L o n g , M. H. (1981) . I npu t , i n t e r a c t i o n , and s e c o n d - l a n g u a g e a c q u i s i t i o n . In H. W i n i t z ( E d . ) , N a t i v e language and f o r e i g n language a c q u i s i t i o n (Annals o f t h e New York Academy o f S c i e n c e s , Volume 379, p p . 259 -278) . New Y o r k : The New York Academy o f S c i e n c e s . L o n g , M. H. (1983a) . L i n g u i s t i c and c o n v e r s a t i o n a l ad jus tments t o n o n - n a t i v e s p e a k e r s . S t u d i e s i n Second Language A c q u i s i t i o n , 5, 177-193. L o n g , M. H. (1983b). N a t i v e s p e a k e r / n o n - n a t i v e s p e a k e r c o n v e r s a t i o n and the n e g o t i a t i o n o f c o m p r e h e n s i b l e i n p u t . A p p l i e d L i n g u i s t i c s , 4, 126-141. L o n g , M. H. (1985a) . Input and second language a c q u i s i t i o n t h e o r y . In S . M. G a s s , & C . G. Madden ( E d s . ) , Input i n second language a c q u i s i t i o n (pp. 377-393) . Rowley , MA: Newbury House . L o n g , M. H. (1985b). A r o l e f o r i n s t r u c t i o n i n second language a c q u i s i t i o n : T a s k - b a s e d language t e a c h i n g . In K. H y l t e n s t a m & M. Pienemann ( E d s . ) , M o d e l l i n g and a s s e s s i n g second language development (pp. 7 7 - 9 9 ) . San D i e g o : C o l l e g e H i l l P r e s s . L o n g , M. H . , Adams, L . , McLean, M . , & C a s t a n o s . (1976) . Do ing t h i n g s w i t h w o r d s — V e r b a l i n t e r a c t i o n i n l o c k s t e p and s m a l l group c l a s s r o o m s i t u a t i o n s . In J . Fanselow & R. Crymes ( E d s . ) , On TESOL '76 (pp. 137 -153) . Wash ing ton , DC: TESOL. L o n g , M. H . , & P o r t e r , P. A . (1985) . Group work, i n t e r l a n g u a g e t a l k , and second language a c q u i s i t i o n . TESOL Q u a r t e r l y , 19, 207-228. L o n g , M. H . , & S a t o , C . (1983) . C l a s s r o o m f o r e i g n e r t a l k d i s c o u r s e : Forms and f u n c t i o n s o f t e a c h e r s ' q u e s t i o n s . In W. H. S e l i g e r & M. H. Long ( E d s . ) , C l a s s r o o m o r i e n t e d r e s e a r c h i n language a c q u i s i t i o n (pp. 268-285) . Rowley, MA: Newbury House. Lowe, P . , J r . (1987) . I n t e r a g e n c y Language Roundtab le O r a l P r o f i c i e n c y I n t e r v i e w [Review o f the L P I ] . In J . C . A l d e r s o n , K. J . Krahnke & C. W. S t a n s f i e l d ( E d s . ) , Reviews o f E n g l i s h language p r o f i c i e n c y t e s t s (pp. 43 -4 7 ) . Wash ing ton , DC: T e a c h e r s o f E n g l i s h t o Speakers o f O t h e r Languages . 243 Malamah-Thomas, A . (1987). C l a s s r o o m i n t e r a c t i o n . O x f o r d : O x f o r d U n i v e r s i t y P r e s s . M a r t i n , J . R. (1983) . The development o f r e g i s t e r . In J . F i n e & R. 0 . F r e e d l e ( E d s . ) , Deve lopmenta l i s s u e s i n d i s c o u r s e (pp. 1 -39) . Norwood, N J : A b l e x . M e r r i a m , S . B . , & S impson, E . L . (1984) A g u i d e t o r e s e a r c h f o r e d u c a t o r s and t r a i n e r s o f a d u l t s . M a l a b a r , F L : R o b e r t E . K r e i g e r . Mohan, B. (1986) . Language and c o n t e n t . R e a d i n g , MA: A d d i s o n - W e s l e y . Newport , E . (1976) . M o t h e r e s e : The speech o f mothers t o young c h i l d r e n . U n p u b l i s h e d d o c t o r a l d i s s e r t a t o n , U n i v e r s i t y o f P e n n s y l v a n i a , P h i l a d e l p h i a . N i e , N. H . , H u l l , C . H . , J e n k i n s , J . G . , S t e i n b r e n n e r , K . , & B r e n t , D. H. (1975) . S t a t i s t i c a l package f o r t h e s o c i a l s c i e n c e s (2nd e d . ) . New York : M c G r a w - H i l l . O x f o r d , R. (1987) . Comprehensive E n g l i s h Language T e s t [Review o f the C E L T ] . In J . C . A l d e r s o n , K. J . Krahnke & C . W. S t a n s f i e l d ( E d s . ) , Reviews o f E n g l i s h language p r o f i c i e n c y t e s t s (pp. 2 2 - 2 4 ) . Wash ing ton , DC: T e a c h e r s o f E n g l i s h t o Speakers o f Other Languages . P i c a , T . (1987) . S e c o n d - l a n g u a g e a c q u i s i t i o n , s o c i a l i n t e r a c t i o n , and the c l a s s r o o m . A p p l i e d L i n g u i s t i c s . 8, 3 -21 . P i c a , T . , & Doughty , C . (1985) . Input and i n t e r a c t i o n i n the communicat ive language c l a s s r o o m : A compar ison o f t e a c h e r - f r o n t e d and group a c t i v i t i e s . In S . M. Gass & C . G . Madden ( E d s . ) , Input i n second language a c q u i s i t i o n (pp. 115-13 2 ) . Rowley, MA: Newbury House. P i c a , T . , & L o n g , M. H. (1986) . The l i n g u i s t i c and c o n v e r s a t i o n a l per formance o f e x p e r i e n c e d and i n e x p e r i e n c e d t e a c h e r s . In R. R. Day ( E d . ) , T a l k i n g t o l e a r n : C o n v e r s a t i o n i n second language a c q u i s i t i o n (pp. 8 5 - 9 8 ) . Rowley, MA: Newbury House. P i c a , T . , Young, R . , & Doughty, C . (1987) . The impact o f i n t e r a c t i o n on comprehens ion . TESOL Q u a r t e r l y , 21 , 737-758. P o r t e r , P. A . (1983) . V a r i a t i o n s i n the c o n v e r s a t i o n s o f a d u l t l e a r n e r s o f E n g l i s h as a f u n c t i o n o f t h e p r o f i c i e n c y l e v e l o f the p a r t i c i p a n t s . U n p u b l i s h e d d o c t o r a l d i s s e r t a t i o n , S t a n f o r d U n i v e r s i t y , P a l o A l t o . 244 P o r t e r , P. A . (1986) . How l e a r n e r s t a l k t o each o t h e r : Input and i n t e r a c t i o n t a s k - c e n t e r e d d i s c u s s i o n s . In R. R. Day ( E d . ) , T a l k i n g t o l e a r n : C o n v e r s a t i o n s i n second language a c q u i s i t i o n (pp. 200-222) . Rowley, MA: Newbury House . R u l o n , K. A . , & M c C r e a r y , J . (1986) . N e g o t i a t i o n o f c o n t e n t : T e a c h e r - f r o n t e d and s m a l l - g r o u p i n t e r a c t i o n . In R. R. Day, T a l k i n g t o l e a r n : C o n v e r s a t i o n s i n second language a c g u i s i t i o n (pp. 183-199) . Rowley, MA: Newbury House . S c a r c e l l a , R. & H i g a C . (1981) . Input n e g o t i a t i o n and age d i f f e r e n c e s i n second language a c q u i s i t i o n . Language L e a r n i n g , 31, 409-437. S c h a c h t e r , J . (1985) . Three approaches t o t h e s t u d y o f i n p u t . Language L e a r n i n g . 36, 211-225. S c h e g l o f f , E . A . , J e f f e r s o n , G . , & S a c k s , H. (1977) . The p r e f e r e n c e f o r s e l f - c o r r e c t i o n i n t h e o r g a n i z a t i o n o f r e p a i r i n c o n v e r s a t i o n . Language. 53, 361-382. S c h i n k e - L l a n o , L . (1986) . F o r e i g n e r t a l k i n j o i n t c o g n i t i v e a c t i v i t i e s . In R. R. Day ( E d . ) , T a l k i n g t o l e a r n : C o n v e r s a t i o n s i n second language a c q u i s i t i o n (pp. 99 -117) . S c h w a r t z , J . (1980) . The n e g o t i a t i o n f o r meaning: R e p a i r i n c o n v e r s a t i o n s between second language l e a r n e r s o f E n g l i s h . In D. La rsen -F reeman ( E d . ) , D i s c o u r s e a n a l y s i s i n second language r e s e a r c h (pp. 138-153) . Rowley, MA: Newbury House . S h o r t r e e d , I. (1986) . The e f f e c t s o f t a s k c o m p l e x i t y and p r o f i c i e n c y on f o r e i g n e r t a l k d i s c o u r s e and communicat ion s t r a t e g i e s i n NS-NSS i n t e r a c t i o n . U n p u b l i s h e d m a s t e r ' s t h e s i s . U n i v e r s i t y o f B r i t i s h C o l u m b i a , V a n c o u v e r . S i e g e l , S . (1956) . Nonparamet r ic s t a t i s t i c s f o r t h e b e h a v i o r a l s c i e n c e s . New York : McGraw H i l l . S m i t h , L . E . ( E d . ) . 1981. E n g l i s h f o r c r o s s - c u l t u r a l c o m m u n i c a t i o n . London: M a c m i l l a n . Snow, C . (1987) . Beyond c o n v e r s a t i o n : Second language l e a r n e r s ' a c q u i s i t i o n o f d e s c r i p t i o n and e x p l a n a t i o n . In L a n t o l f , J . P. & L a b a r c a , A . ( E d s . ) , R e s e a r c h i n second language l e a r n i n g : Focus on the c l a s s r o o m (pp. 3 - 1 6 ) . Norwood, N J : A b l e x . T a b a c h n i c k , B. G . , & F i d e l l , L . S . (1983) . U s i n g m u l t i v a r i a t e s t a t i s t i c s . New Y o r k : Harper and Row. 245 T a r o n e , E . (1983) . Some t h o u g h t s on t h e n o t i o n o f ' communica t ion s t r a t e g y ' . In C . F a e r c h & G. K a s p e r ( E d s . ) , S t r a t e g i e s i n i n t e r l a n g u a g e communicat ion (pp. 6 1 - 7 4 ) . London: Longman. T u r n e r , R. ( E d . ) . (1974) . Ethnomethodolocry. Harmondsworth, E n g l a n d : Penguin Books . V a r o n i s , E . M . , & G a s s , S . M. (1982) . The c o m p r e h e n s i b i l i t y o f n o n - n a t i v e s p e e c h . S t u d i e s i n Second Language A c q u i s i t i o n , 4, 114-136. V a r o n i s , E . M . , & G a s s , S . M. (1985) . N o n - n a t i v e / n o n - n a t i v e c o n v e r s a t i o n s : A model f o r n e g o t i a t i o n o f meaning . A p p l i e d L i n g u i s t i c s , 6, 71 -90 . Wagner, J . (1983) . Dann du tagen e i n e e e e e — w e i s s e P l a t t e — A n a n a l y s i s o f i n t e r l a n g u a g e communicat ion i n s t r u c t i o n s . In C . F a e r c h & G. Kasper ( E d s . ) , S t r a t e g i e s i n i n t e r l a n g u a g e communicat ion (pp. 159-174) . London: Longman. Wesche, M. B . , & Ready, D. (1985) . F o r e i g n e r t a l k i n t h e u n i v e r s i t y c l a s s r o o m . In G a s s , S . M. & Madden C . G . ( E d s . ) , Input i n second language a c g u i s i t i o n (89 -113) . 246 APPENDICES (A-K) 247 ask f o r feedback and c l e a r up any p o i n t d u r i n g t h e c o n v e r s a t i o n s . I t i s most impor tan t t h a t you t r y t o ge t the most out o f t h e s e t a l k s : The t e a c h e r w i l l be g i v i n g you i n f o r m a t i o n and i s ready and w i l l i n g t o h e l p y o u , bu t you w i l l have t o do y o u r b e s t t o communicate t o o . I do a p p r e c i a t e y o u r h e l p i f you can s p a r e t h e t i m e . We s h o u l d be a b l e t o f i n i s h a l l t e s t i n g and c o n v e r s a t i o n s i n a t o t a l t ime o f one hour and 45 m i n u t e s . I t h i n k you w i l l f i n d t h e e x p e r i e n c e t o be an e x c e l l e n t form o f p r a c t i c e and a way o f l e a r n i n g what your l e v e l i s . I f you would l i k e t o p a r t i c i p a t e , p l e a s e c a l l me w i t h i n t h e nex t two weeks a t the t e l e p h o n e number l i s t e d above . In a d d i t i o n , f e e l f r e e a t any t ime t o c a l l me w i t h any q u e s t i o n s you may have . F i n a l l y , I want everyone t o know t h a t you may f r e e l y d e c i d e no t t o p a r t i c i p a t e and t h a t you can f r e e l y wi thdraw from t h e p r o j e c t a t any t ime . Your h e l p on t h e p r o j e c t w i l l be e n t i r e l y v o l u n t a r y and c o n f i d e n t i a l . A n d , i f you do d e c i d e no t t o p a r t i c i p a t e , o r t o wi thdraw, i t w i l l not be h e l d a g a i n s t you i n any way. YOROSHIKU ONEGAI SHIMASU! 250 Appendix C Statement o f Informed Consent I c o n s e n t t o p a r t i c i p a t i o n i n the r e s e a r c h p r o j e c t , "The E f f e c t o f Task V a r i a t i o n i n T e a c h e r - l e d Groups on R e p a i r o f E n g l i s h as a F o r e i g n Language" c o n d u c t e d by R i c h a r d B e r w i c k , Kobe U n i v e r s i t y o f Commerce. I u n d e r s t a n d t h a t the main p u r p o s e s o f t h e p r o j e c t a r e t o examine 1) how v a r i o u s c o n v e r s a t i o n a l s i t u a t i o n s i n f l u e n c e t h e k i n d o f v e r b a l a s s i s t a n c e p a r t i c i p a n t s o f f e r t o each o t h e r and 2) i f the a s s i s t a n c e depends on whether a group i s mixed ( n a t i v e E n g l i s h and Japanese) o r homogeneous (Japanese o n l y ) . I a l s o u n d e r s t a n d t h a t I w i l l be work ing w i t h one o t h e r p a r t i c i p a n t i n a two-member g r o u p , t h a t our group w i l l engage i n f i v e b r i e f (5-7 minute) c o n v e r s a t i o n s t o t a l i n g about one hour w i t h b r e a k s , and t h a t the c o n v e r s a t i o n s w i l l be r e c o r d e d f o r l a t e r a n a l y s i s . My t o t a l p a r t i c i p a t i o n t i m e , i n c l u d i n g n e c e s s a r y b r i e f i n g s o r E n g l i s h p r o f i c i e n c y t e s t s and the c o n v e r s a t i o n s , w i l l be l e s s than one hour and 45 m i n u t e s . The f i v e s i t u a t i o n s i n c l u d e a f o r m a l d i s c u s s i o n , a d e m o n s t r a t i o n , two model r e c o n s t r u c t i o n t a s k s and a p e r i o d o f i n f o r m a l d i s c u s s i o n , i n a d d i t i o n t o an o p p o r t u n i t y t o e v a l u a t e the t a s k s immed ia te ly a f t e r t h e y have been c o m p l e t e d . Mr. Berwick has a s s u r e d me t h a t my i d e n t i t y w i l l remain c o n f i d e n t i a l ( tha t my name w i l l not be used d u r i n g a n a l y s i s o f d a t a and r e p o r t i n g o f r e s u l t s ) . He has a l s o o f f e r e d t o answer any f u r t h e r q u e s t i o n s I may have about the s t u d y and i t s p r o c e d u r e s i n o r d e r t o ensure my f u l l u n d e r s t a n d i n g . Mr. Berw ick has a l s o in fo rmed me t h a t I may r e f u s e t o p a r t i c i p a t e i n the s t u d y , t h a t my s e r v i c e s may be wi thdrawn a t any t ime f o r any r e a s o n I choose and t h a t such w i thdrawa l w i l l i n no way be h e l d a g a i n s t me. F i n a l l y , I acknowledge r e c e i p t o f a copy o f t h i s s t a t e m e n t , i n c l u d i n g a l l a t t a c h m e n t s . DATE SIGNATURE Appendix D E n g l i s h Language P r o f i c i e n c y T e s t s Used t o S e l e c t S u b j e c t s 1. The CELT - Comprehensive E n g l i s h Language T e s t (see H a r r i s & Pa lmer , 1986b; O x f o r d , 1987) . The CELT ( S t r u c t u r e S e c t i o n ) i s a 7 5 - i t e m t e s t o f E n g l i s h language p r o f i c i e n c y d e s i g n e d f o r h i g h s c h o o l , c o l l e g e and a d u l t l e a r n e r s o f E S L / E F L a t i n t e r m e d i a t e and advanced l e v e l s . K u d e r - R i c h a r d s o n 20 r e l i a b i l i t i e s (Form A) range from .88 t o .96 f o r the s i x r e f e r e n c e groups u s e d t o e s t a b l i s h t e s t norms. The S t r u c t u r e S e c t i o n emphasizes the s o l u t i o n o f grammar problems i n a c o n v e r s a t i o n a l c o n t e x t . The examinees a r e asked t o choose the word o r phase a n a t i v e s p e a k e r o f E n g l i s h would use i n t h e c o n v e r s a t i o n and t h e n t o mark t h e answer shee t w i t h the c o r r e c t answer . Two examples o f t h e t e s t q u e s t i o n s f o l l o w (see H a r r i s & Pa lmer , 1986b, p . 11) . Example I: "How o l d i s George?" " H e ' s two y e a r s younger h i s b r o t h e r P a u l . " (A) t h a t (B) o f (C) as (D) than (x) Example I I : "Have you f i n i s h e d the r e p o r t f o r Mr . J o n e s ? " " Y e s , I t h i s m o r n i n g . " (A) i t t o him gave (B) gave i t t o him (x) (C) t o him gave i t (D) gave t o him i t 253 No o v e r a l l r e l i a b i l i t y e s t i m a t e i s p r o v i d e d nor a r e norms r e p o r t e d s p e c i f i c a l l y f o r Japanese l e a r n e r s o f E n g l i s h as f o r e i g n l a n g a u g e . O x f o r d (1987) c o n c l u d e s : In sum, the a u t h o r s have c r e a t e d a t e s t t h a t i s r e l i a b l e , v a l i d , and u s e f u l f o r n o n n a t i v e s p e a k e r s o f E n g l i s h . The t e s t appears t o measure E n g l i s h language p r o f i c i e n c y i n a way t h a t i s easy t o a d m i n i s t e r and s c o r e . More work on norming would make the t e s t more u s e f u l , (p. 24) 2 . The LPI - Language P r o f i c i e n c y I n t e r v i e w (see E d u c a t i o n a l T e s t i n g S e r v i c e , 1982; Lowe, 1987) . The L P I , a l s o known as the FSI ( F o r e i g n S e r v i c e In te rv iew) and the ILR ( In te ragency Language R o u n d t a b l e O r a l P r o f i c i e n c y I n t e r v i e w ) , i s a t e s t o f an i n d i v i d u a l ' s a b i l i t y t o c o n v e r s e i n a f o r e i g n language i n an i n t e r v i e w s i t u a t i o n . The manual (see E d u c a t i o n a l T e s t i n g S e r v i c e , 1982) p r o v i d e s t h e f o l l o w i n g d e s c r i p t i o n : The i n t e r v i e w c o n s i s t s o f a f a c e - t o - f a c e c o n v e r s a t i o n w i t h one o r two t r a i n e d t e s t e r s f o r a p e r i o d o f 10 t o 3 0 m i n u t e s . The r e s u l t i n g speech sample i s t h e n r a t e d on a s c a l e o f 0 ( f o r no p r a c t i c a l a b i l i t y t o f u n c t i o n i n t h e language) t o 5 ( f o r a b i l i t y e q u i v a l e n t t o t h a t o f an educa ted n a t i v e speaker) w i t h p l u s r a t i n g s f o r s t r o n g per formance w i t h i n a l e v e l . Examinees a r e g u i d e d th rough s e v e r a l s t a g e s d u r i n g the i n t e r v i e w . F o l l o w i n g a warm-up, q u e s t i o n s a r e d e s i g n e d t o e s t a b l i s h an i n i t i a l l e v e l (based on a d e s c r i p t i o n o f 254 f u n c t i o n , c o n t e n t and a c c u r a c y f o r each o f t h e f i v e l e v e l s ) , p robe a t t h e next h i g h e s t l e v e l o r u n t i l t he examinee i s u n a b l e t o o f f e r s u s t a i n e d r e s p o n s e s , and t h e n r e t u r n t o the l e v e l a t wh ich the examinee i s a b l e t o r e s p o n d . The manual n o t e s t h a t the o r a l i n t e r v i e w was d e v e l o p e d t o overcome the d i f f i c u l t y o f p r o d u c i n g v a l i d assessments o f o r a l p r o d u c t i o n w i t h p a p e r - a n d - p e n c i l t e s t s (p. 9 ) , and argues t h a t such t e s t s , g i v e n the o r a l assessment g o a l , l a c k b o t h c o n t e n t v a l i d i t y and f a c e v a l i d i t y . Lowe (1987) p o i n t s out t h a t " the t e s t i s not an i n s t r u m e n t because the p r o c e d u r e i s n e i t h e r f i x e d i n p r i n t n o r i n v a r i a b l e " (p. 46 ) . Lowe a l s o r e p o r t s a h i g h degree o f i n t e r - r a t e r r e l i a b i l i t y f o r the t e s t (Pearson product-moment c o r r e l a t i o n s e x c e e d i n g . 8 7 ) , and n o t e s t h a t the c o n t e n t v a l i d i t y o f the i n t e r v i e w depends on t h e t r a i n e d i n t e r v i e w e r ' s a b i l i t y t o employ q u e s t i o n t y p e s o r o r a l t a s k s wh ich a r e a p p r o p r i a t e f o r examinees a t d i f f e r e n t l e v e l s o f p r o f i c i e n c y . Lowe c i t e s Bachman and Palmer (1981) who found c o n v e r g e n t and d i s c r i m i n a n t v a l i d i t y f o r " t h e i r v e r s i o n o f t h e o r a l i n t e r v i e w p r o c e d u r e " (p. 4 6 ) , and n o t e s t h a t the p r o c e d u r e p o s s e s s e s a h i g h degree o f f a c e v a l i d i t y . 255 Appendix E I n s t r u c t i o n s t o R a t e r s and Index o f Dependent V a r i a b l e s RATING CASES OF CONVERSATIONAL REPAIR AND REFERENCE I have l i s t e d below t y p i c a l examples o f how r e p a i r and r e f e r e n c e l o o k i n c o n t e x t . You w i l l be r a t i n g o n l y seven o f t h e 14 t y p e s l i s t e d ; you need r e a d o n l y the seven which a r e headed w i t h an a s t e r i s k (* ) . F I R S T — P l e a s e r e a d the examples below i n o r d e r t o g e t a f e e l f o r each t y p e . You w i l l f i n d the f o l l o w i n g i n f o r m a t i o n f o r each example: (a) an index number i n p a r e n t h e s e s f o r t h e i t e m , (b) a name, (c) a b r i e f d e s c r i p t i o n and (d) the c o n v e r s a t i o n a l e x c e r p t w i t h i n which the r e p a i r o r r e f e r e n c e o c c u r s . ( I n f o r m a t i o n i n the square b r a c k e t s f o l l o w i n g each e x c e r p t r e f e r s t o the t r a n s c r i p t i n which i t o c c u r s and has no b e a r i n g on y o u r r a t i n g . ) SECOND—Once you have read the examples , you can go on t o t h e nex t f i v e p a g e s . I have l i s t e d 21 e x c e r p t s from t r a n s c r i p t s o f c o n v e r s a t i o n s between a t e a c h e r and a l e a r n e r . Note t h a t each e x c e r p t e x e m p l i f i e s o n l y one form o f r e p a i r o r r e f e r e n c e , which i s u n d e r l i n e d . A f t e r r e a d i n g each e x c e r p t , p l e a s e d e c i d e which k i n d o f r e p a i r o r r e f e r e n c e i t i l l u s t r a t e s . T h e n , w r i t e the index number f o r the r e p a i r o r r e f e r e n c e i n the space p r o v i d e d . The seven r e p a i r and r e f e r e n c e c a t e g o r i e s , and t h e i r index numbers, have been r e p r o d u c e d f o r you a t t h e t o p o f each r a t i n g s h e e t . NOTE THE FOLLOWING EXAMPLES (1) C l a r i f i c a t i o n R e q u e s t : The l i s t e n e r i n d i c a t e s l a c k o f u n d e r s t a n d i n g t h r o u g h an i m p l i e d o r e x p l i c i t r e q u e s t f o r the s p e a k e r t o expand o r r e f o r m u l a t e an u t t e r a n c e . CAN YOU FIND THAT PIECE? +++ I beg y o u r pardon? [2LEGl] *(2) Comprehension Check: A speaker checks whether t h e l i s t e n e r has u n d e r s t o o d the u t t e r a n c e . BUMPS IS A KIND OF SMALL, UH, LIKE A CIRCLE. - C IRCLE, •S A L ITTLE BIT ELEVATED - C IRCLE, OKAY? [8LEG1] *(3) C o n f i r m a t i o n Check: A speaker r e q u e s t s c o n f i r m a t i o n t h a t t h e p r e v i o u s u t t e r a n c e has been h e a r d c o r r e c t l y by r e p e a t i n g a word o r p h r a s e from the u t t e r a n c e and a d d i n g r i s i n g i n t o n a t i o n . Where? UHH, ++ STARTING ON THE SECOND LINE . Second l i n e ? [10LEG2] (4) D e f i n i t i o n : A speaker s t a t e s what a word o r p h r a s e means, e i t h e r i n r e s p o n s e t o o r i n a n t i c i p a t i o n o f the l i s t e n e r ' s l a c k o f c o m p r e h e n s i o n ; the d e f i n i t i o n t y p i c a l l y t a k e s t h e form "A i s a ( type of ) B " . THAT'S IT . A STRING IS JUST A THICK PIECE OF THREAD BUT IN, IN COMPU UH, / r i b - r i b b o n . r i b b o n . No. COMPUTER DOESN'T HAVE A RIBBON. [2C0M1] *(5) D i s p l a y Q u e s t i o n : Requests the l i s t e n e r t o demonst ra te knowledge o r i n f o r m a t i o n a l r e a d y p o s s e s s e d by t h e s p e a k e r and known by t h e l i s t e n e r t o be p o s s e s s e d by the s p e a k e r . The " d i s p l a y " may a l s o t ake the form o f a r h e t o r i c a l q u e s t i o n which i s answerable by the s p e a k e r who p o s e s i t . APPEARING - ONE LETTER BEFORE [emph] T - H - E . SEE? - SO WHY DON'T WE KEEP- ++ IN ORDER TO FIND ANOTHER T - H - E WHAT SSSHALL I DO? [8COM2] (6) E c h o : E x a c t comple te o r ( t y p i c a l l y ) p a r t i a l r e p e t i t i o n , w i t h f l a t o r f a l l i n g i n t o n a t i o n , o f the p r e c e d i n g s p e a k e r ' s u t t e r a n c e . MHMM. NOW PRESS THE ONE THAT GOES DOWN. Down [2COM2] 257 *(7) L e x i c a l U n c e r t a i n t y : H e s i t a n t o r t e n t a t i v e a t tempt t o r e c a l l o r p r o p e r l y employ a p a r t i c u l a r word; o f t e n c h a r a c t e r i z e d by r e p e t i t i v e p r o d u c t i o n o f i n c o m p l e t e o r i n c o r r e c t forms o f the l e x i c a l i t e m . u h , I - I- I h e l d - I h - u h , I h e l d - the s p - u h - - o r a t o r i c a l c o n t e s t , - and I t o o k - the management - o f t h a t c o n t e s t . [8DIS] *(8) R e f e r e n t i a l Q u e s t i o n : A means o f e l i c i t i n g i n f o r m a t i o n wh ich i s unknown and o f i n t e r e s t t o the s p e a k e r , and which may be p o s s e s s e d by the h e a r e r . R e f e r e n t i a l q u e s t i o n s a r e o r i e n t e d t o the t o p i c r a t h e r than t o the q u a l i t y o f language by wh ich t h e t o p i c i s e x p r e s s e d . UM - I THINK KOCHI IS FAMOUS FOR - UH ++ FIGHTING DOGS. Ahhh! UH - WHAT - WHAT DO YOU CALL THEM? [2DIS] (9) S e l f - e x p a n s i o n : P a r t i a l o r comple te r e p h r a s i n g o f o n e ' s own u t t e r a n c e , o f t e n o c c u r r i n g w i t h i n t h e s p e a k e r ' s t u r n bu t p o s s i b l y o c c u r r i n g w i t h i n the s p e a k e r ' s nex t t u r n . (The ">" i n d i c a t e s the r e f e r e n t i a l p o i n t f o r t h e e x p a n s i o n . ) YES , THAT ONE- AND THEN - >PUT IT - ON THAT - - THE LONGER ONE. + PUT THE SQUARE ONE ON THE LONG- LONG ONE . [10LEG2] (10) S e l f - r e p e t i t i o n : E x a c t , p a r t i a l o r semant ic ( e q u i v a l e n t ) r e p e t i t i o n o f o n e ' s p r e v i o u s u t t e r a n c e w i t h i n f i v e t u r n s o f t h a t u t t e r a n c e . The s e l f - r e p e t i t i o n f r e q u e n t l y o c c u r s w i t h i n the s p e a k e r ' s own t u r n . (The ">" i n d i c a t e s the r e f e r e n t i a l p o i n t f o r the r e p e t i t i o n . ) YEAH + UMMM ++ L E T ' S JUST TRY THAT. YEAH >JUST TRY IT THERE. L E T ' S JUST TRY IT THERE. WE'LL BE CREATIVE WITH THIS THING. [3LEG2] (11) O t h e r - e x p a n s i o n : P a r t i a l r e p h r a s i n g o f t h e p r e v i o u s s p e a k e r ' s u t t e r a n c e . R e p h r a s i n g t y p i c a l l y i n c l u d e s new m a t e r i a l i n a d d i t i o n t o the r e p e t i t i o n . (The ">" i n d i c a t e s the r e f e r e n t i a l p o i n t f o r the e x p a n s i o n . ) 258 I SEE IT ON PEOPLE'S FRONT DOORS OR ON THEIR CAR. / Y e s , we p u t on c a r s OR ON THE FRONT OF THE CAR. IS IT FOR >GOOD LUCK FOR / A h THE NEW YEAR FOR THE ++ DRIVING, OR / y e s Yeah . I t means the c e l e b r a t i o n o r good l u c k . Urn, I t h i n k . [12DIS] (12) O t h e r - r e p e t i t i o n : E x a c t , p a r t i a l o r s e m a n t i c r e p e t i t i o n o f t h e p r e v i o u s s p e a k e r ' s u t t e r a n c e w i t h i n f i v e t u r n s o f the u t t e r a n c e . (The ">" i n d i c a t e s the r e f e r e n t i a l p o i n t f o r t h e r e p e t i t i o n . ) U h , not i n s i d e . S o , i n f r o n t o f the g a t e . UH HUH. I S E E , GATE. YOU SAID GATE. UM HM. [12DIS] *(13) A n a p h o r i c R e f e r e n c e : A n a p h o r i c r e f e r e n c e p o i n t s back t o someth ing c o n c r e t e l y i d e n t i f i e d a t a p r e v i o u s p o i n t i n t h e t e x t . Anaphora t y p i c a l l y t a k e s the form o f a pronoun ( t h u s , >BOOK . . . I T ) . IT cannot be i n t e r p r e t e d w i t h o u t i d e n t i f i c a t i o n o f the r e f e r e n t i a l s o u r c e (">"). SO. ARE ALL OF THE >PIECES TURNED RIGHT SIDE UP? (/Mhmm./) Y e s , y e s they a r e . [2LEG1] *(14) E x o p h o r i c R e f e r e n c e : E x o p h o r i c r e f e r e n c e p o i n t s out o b j e c t s o r r e l a t i o n s h i p s i n the c o n v e r s a t i o n a l c o n t e x t . I t i s e n t i r e l y c o n t e x t - b o u n d and o r d i n a r i l y cannot be i n t e r p r e t e d w i t h o u t s h a r e d p e r c e p t i o n o r e x p e r i e n c e . The t e x t does not show a p r i o r c o n c r e t e r e f e r e n t f o r an e x o p h o r i c p r o n o u n . TO THE RIGHT. ++ YEAH! AND DOWN! ONE- ONE - YEAH, DOWN! BEAUTIFUL! A h ! THAT 1 S [emph] THE ONE I WANT. [8LEG2] Comprehension Check (2) R e f e r e n t i a l Q u e s t i o n (8) C o n f i r m a t i o n Check (3) A n a p h o r i c R e f e r e n c e (13) D i s p l a y Q u e s t i o n (5) E x o p h o r i c R e f e r e n c e (14) L e x i c a l U n c e r t a i n t y (7) NOW PLEASE RATE THE FOLLOWING EXCERPTS BY FILLING IN THE APPROPRIATE NUMBER W r i t e t h e a p p r o p r i a t e index number on the l i n e t o t h e r i g h t o f each e x c e r p t . F o r example: THAT' S [emph] THE ONE I WANT. 14 [8LEG2] << 1 >> YOU UNDERSTAND? Y e s , y e s , y e s . Hm. OKAY. ++ DID YOU GET IT? [6LEG1] << 2 » WHEN WE WERE GOING THROUGH, I SHOULD HAVE STOPPED YOU AT TWO POINTS. WE HAD THE WORD, "THEORY", T - H - E - O - R - - I - E - S . WHY DID IT STOP AT THE WORD, "THEORY"? BECAUSE WE TYPED IN T - H - E , BUT WE DIDN'T LEAVE A SPACE IN FRONT OR A SPACE IN ( / H n n . / ) BACK, SO WHAT WE DID JUST NOW WAS TO TYPE IN THE LETTERS, -[4COM2] << 3 » NEXT TO IT . RIGHT NEXT TO IT . THAT'S IT . YEAH. - AND THEN - HMMM. THIS IS A - THE NEXT ONE IS VERY COMPLICATED. TAKE TWO PIECES - OF UHHH - YELLOW BUMPS, PLEASE. /Two p i e c e s ? What k i n d o f ? [9LEG2] Comprehension Check (2) R e f e r e n t i a l Q u e s t i o n (8) C o n f i r m a t i o n Check (3) A n a p h o r i c R e f e r e n c e (13) D i s p l a y Q u e s t i o n (5) E x o p h o r i c R e f e r e n c e (14) L e x i c a l U n c e r t a i n t y (7) « 4 » THAT'S GOING TO GO ONN- THEEE - SSECOND SSET + OF POINTS. / s e c o n d ? Second? THE SECOND SET . [7LEG2] << 5 >> THEN F I L E "A" BUTTON, F ILE " B " BUTTON, F I L E " C " BUTTON AND THEN - THE ENTIRE DIFF , DIFFERENTLY. ENTIRELY DIFFERENT MANUSCRIPT APPEARS JUST LIKE MAGIC. YOU WILL LOVE IT . [10COM1] << 6 » YOU DO THAT? Put i t on - where? YEAH, IN THE NEXT [emph] STAGE. [8LEG2] « 7 >> Ohh, o h h , I s e e , I s e e . . OKAY? - PY- LIKE A PYRAMID. Okay. [10LEG] << 8 » SO WHAT ABOUT UH " T - H - E " THIS TIME? BUT, UH, BEFORE THAT, WE HAVE TO PUT, UH, BUTTON, OK, THAT ONE. DO YOU REMEMBER THE /mmm BUTTON YOU PRESS? OK, THAT ONE. YES THIS ONE. [6COM2] 261 Comprehension Check (2) R e f e r e n t i a l Q u e s t i o n (8) C o n f i r m a t i o n Check (3) A n a p h o r i c R e f e r e n c e (13) D i s p l a y Q u e s t i o n (5) E x o p h o r i c R e f e r e n c e (14) L e x i c a l U n c e r t a i n t y (7) « 9 >> THERE'S A BLACK PIECE WITH A KIND OF ANTENNA. Antenna? D'YOU SEE IT? [4LEG1] « 10 » THAT ONE. T h i s ? OKAY, YEAH. - RIGHT THERE. THAT'S RIGHT. YOU'RE VERY SMART. [6LEG2] « 11 » OKAY. SO, THE NEXT CAN YOU FIND - A SQUARE BOARD? Square b o a r d Mm. OKAY, UH, WITH- ONE TWO THREE FOUR FIVE SIX SEVEN EIGHT NINE TEN TWELVE, OKAY, TWELVE BUMPS ON IT . IT IS - BLACK COLOR. [6LEG1] << 12 >> YEAH. I LIKED KOCHI AND I WAS A L ITTLE BIT - SURPRISED. A h ! On what p o i n t ? WELL, BEFORE I WENT TO SHIKOKU, I TOLD SOME OF MY FRIENDS THAT I WAS GOING TO SHIKOKU, AND THEY SAID, "SHIKOKU! A H ! " [2DIS] « 13 » We ' re u n d e r s t a n d . OKAY, YOU UNDERSTAND, OKAY? + OKAY, SO NOW YOU HAVE THREE -[6LEG1] 262 Comprehension Check (2) R e f e r e n t i a l Q u e s t i o n (8) C o n f i r m a t i o n Check (3) A n a p h o r i c R e f e r e n c e (13) D i s p l a y Q u e s t i o n (5) E x o p h o r i c R e f e r e n c e (14) L e x i c a l U n c e r t a i n t y (7) « 14 » T H E - COME THIS WAY. COME TOWARDS ME. BRING YOUR HAND -THA- - NEXT TO THAT. + THAT'S RIGHT. + THAT'S IT . [4LEG2] << 15 >> YE - UHHH, I THINK- GRAY [emph] A h , g r a y ? YEAH, - GRAY. (3) MAYBE WOULD YOU TURN IT DOWN? TURN [8LEG2] << 16 >> JUST EXPERIMENT. ++ WHY - WHY DON'T YOU MAKE IT GO - OVER -HERE? PUT IT IN THE - UH - NOW WHAT DOES THAT SAY THERE? Mmm NEED DO. MAKE IT GO TO NEED DO. THAT'S IT . [2COM2] « 17 » OKAY, THEN, UH, PLEASE FIND, UH, THE LONNNG + UH, WHITE BOARD WITH - 1, 2, 3, 4, 5, 6, 7, 8! A h ! Y e s , I s e e . / 8 BUMPS. + DID YOU FIND IT? [8LEG1] « 18 » PLEASE, YEAH. - UH - , NO, - THE BLACK P I E C E , PLEASE. B l a c k ? + YEAH, THAT'S IT . - AND PUT IT + UHHH, YEAH. [9LEG2] 263 Comprehension Check (2) R e f e r e n t i a l Q u e s t i o n (8) C o n f i r m a t i o n Check (3) A n a p h o r i c R e f e r e n c e (13) D i s p l a y Q u e s t i o n (5) E x o p h o r i c R e f e r e n c e (14) L e x i c a l U n c e r t a i n t y (7) « 19 » THINK IT MIGHT BE CLOSE + IS THAT IT? OR IS THAT THE ONE YOU T h i s i s no t + t h i s one? [2LEG2] « 20 » Hahhh! A t t h a t t i m e , i n you - how - what p l a c e d i d you - - v i s i t - - v i s i t - f o r t h e t r i p ? [6DIS] « 21 » IN - ON THAT PAGE. NOW. AT THE TOP OF THE + OF THE + BARS. THERE ARE - FIVE BUTTONS. [4COM1] 264 Appendix F T r a n s c r i p t i o n C o n v e n t i o n s Most o f t h e f o l l o w i n g c o n v e n t i o n s have been adopted from Brown & Y u l e (1983b). 1. P a u s e s : - a v e r y b r i e f pause (about 1/2 s e c . ) + a s h o r t pause (about 1 s e c . ) ++ a l o n g pause (about 1 1/2 s e c . ) (4) f o r pauses o f two seconds o r l o n g e r , e s t i m a t e d number o f seconds i s i n p a r e n t h e s e s 2. The p o i n t o f o v e r l a p p i n g speech i s r e p r e s e n t e d by a s l a s h e d l i n e : Four bumps w i t h square bumps? Ah h a h . Mm /FOUR BUMPS. OH, THAT'S IT . B IG, OK. 3. Omi t ted and u n c l e a r segments: F o r words no t h e a r d c l e a r l y - ?WORD F o r words and which cannot be guessed - ??? WORD F o r words and p h r a s e s c u t s h o r t by s p e a k e r - WORD 3. O t h e r : R i s i n g i n t o n a t i o n - WORD? I n b r e a t h - ASHHH O u t b r e a t h - 'HAAA Lengthened sounds - SSSSO. UHHHH. T r a n s c r i b e r ' s comments - [emph] 4. S p a c i n g o f t r a n s c r i b e d t e x t : S i n g l e spaced f o r c o n t i n u o u s o r o v e r l a p p i n g s p e e c h . Double spaced between t u r n s . 265 Appendix G ANOVA F r a t i o s f o r S e l e c t e d T rans fo rmed and Unt rans formed V a r i a b l e s L i s t e d i n T a b l e 3 Treatment o f dependent v a r i a b l e by t r a n s f o r m a t i o n (Group/Task) Dependent v a r i a b l e square r o o t l o g a r i t h m i c u n t r a n s f o r m e d Comprehension . 07 . 09 .04 check 12.53 12.74 11.19 C o n f i r m a t i o n .32 .68 . 18 check 13.99 12.68 11. 68 D e f i n i t i o n 2 .10 2.22 1.90 3.77 4.00 3.43 E x o p h o r i c . 12 . 07 .00 r e f e r e n c e 49.18 30.73 48.11 O t h e r - .33 . 11 .53 r e p e t i t i o n 2 .46 2.71 2.21 S e l f - .32 .50 .17 r e p e t i t i o n 9.30 10.33 7.83 Echo .26 .26 .30 4.67 4.20 4.46 N o t e . F r a t i o s a r e l i s t e d f i r s t by group and t h e n by t a s k f o r each dependent v a r i a b l e 266 Appendix H S i g n i f i c a n t R e p a i r C a t e g o r i e s and S o u r c e s o f V a r i a n c e f o r E x p e r i e n t i a l and E x p o s i t o r y T a s k s U s i n g LEG1 as the E x p o s i t o r y Stem ANOVA Main s o u r c e s o f EXPER-EXP0S1 EXPER-EXPOS2 v a r i a n c e (p_ <) R e p a i r exponent F r a t i o p_ F r a t i o p_ .05 .01 Comp. 14. ,533 . 0034 EY > EL EY > EL check 18. , 343 .0016 EY > EL EY > EL D i s p l a y 65. ,681 .0001 E L > EY EL > EY q u e s t i o n — — — — L e x i c a l 5. ,912 . 0354 H > M — u n c e r t . 12. .707 .0051 H > M -17. .962 .0017 EY > E L EY > EL 11, .942 . 0062 EY > EL EY > EL 5, .293 .0442 a t EXPOS1 H > M a t EXPER1 H > M S e l f - 4, . 179 . 0682 r e p a i r 4, . 166 .0685 N o t e . G r o u p s : M = M i x e d , H = Homogeneous; T a s k s : EL = E x p e r i e n t i a l , EY = E x p o s i t o r y ; d f = 1 i n a l l c a s e s 267 Appendix I ANOVA T a b l e s Comparing E x p o s i t o r y w i t h E x p e r i e n t i a l T a s k s (COMl v s . LEG2 + COM2) T a b l e 1-1 A n a l y s i s o f V a r i a n c e f o r C l a r i f i c a t i o n Requests Source o f Sum o f Mean E p s i l . v a r i a t i o n d f squares square F p c o r r . Group 1 37.500 37.500 11.125 .0075 E r r o r 10 33.708 3.371 Task 1 7.042 7.042 1.366 .2696 G X T 1 .667 .667 .129 .7266 E r r o r 10 51.542 5.154 1.00 T a b l e 1-2 A n a l y s i s o f V a r i a n c e f o r C o n f i r m a t i o n Checks Source o f Sum o f Mean E p s i l . v a r i a t i o n d f s q u a r e s square F p c o r r . Group 1 .260 .014 .014 .9093 E r r o r 10 190.771 19.077 Task 1 106.260 106.260 8.360 .0161 G X T 1 3.010 3.101 .237 .6370 E r r o r 10 127.104 12.710 1.00 268 T a b l e 1-3 A n a l y s i s o f V a r i a n c e f o r D e f i n i t i o n s Source o f v a r i a t i o n d f Sum o f squares Mean square E p s i l . E c o r r . Group E r r o r Task G x T E r r o r 1 10 1 1 10 1.500 13.708 7.042 . 000 12.208 1.500 1.371 7. 042 . 000 1.221 1.094 . 3202 5.768 .0372 .000 1.0000 1.00 T a b l e 1-4 A n a l y s i s o f V a r i a n c e f o r E x p r e s s i o n s o f L e x i c a l U n c e r t a i n t y Source o f v a r i a t i o n d f Sum o f s q u a r e s Mean square E p s i l , c o r r . Group E r r o r Task G x T E r r o r 1 10 1 1 10 1.260 28.854 29.260 .844 19.521 1.260 2.885 29.260 .844 1.952 .437 .5236 14.989 .432 . 0031 .5257 1.00 T a b l e 1-5 Source o f Sum o f Mean E p s i l . v a r i a t i o n d f squares square F P c o r r . Group 1 16.667 16.667 1.555 .2408 E r r o r 10 107.167 10.717 Task 1 253.500 253.500 28.324 .0003 G x T 1 6. 000 6.000 .670 .4320 E r r o r 10 89.500 8.950 1.00 269 T a b l e 1-6 A n a l y s i s o f V a r i a n c e f o r S e l f - R e p e t i t i o n Source o f Sum o f Mean E p s i l . v a r i a t i o n d f squares square F p c o r r . Group 1 60.167 60.167 .597 .4574 E r r o r 10 1007.042 100.704 Task 1 1. 042 1.042 . 039 .8473 G x T 1 181.500 181.500 6.805 . 0261 E r r o r 10 266.708 26.671 270 Appendix J ANOVA T a b l e s Comparing E x p o s i t o r y w i t h E x p e r i e n t i a l Tasks (COMl v s . LEG2) T a b l e J - l A n a l y s i s o f V a r i a n c e f o r C l a r i f i c a t i o n Requests Source o f Sum o f Mean E p s i l . v a r i a t i o n d f squares square F p c o r r . Group 1 37.500 37.500 7.895 .0185 E r r o r 10 47.500 4.750 Task 1 13.500 13.500 2.462 .1477 G X T 1 .667 .667 .122 .7346 E r r o r 10 54.833 5.483 1.00 T a b l e J - 2 A n a l y s i s o f V a r i a n c e f o r C o n f i r m a t i o n Checks S o u r c e o f Sum o f Mean E p s i l . v a r i a t i o n d f squares square F p c o r r . Group 1 1.042 1.042 .025 .8782 E r r o r 10 421.083 42.108 Task 1 442.042 442.042 14.426 .0035 G X T 1 .042 .042 .001 .9713 E r r o r 10 127.104 12.710 1.00 271 T a b l e J - 3 A n a l y s i s o f V a r i a n c e f o r D e f i n i t i o n s S o u r c e o f Sum o f Mean E p s i l . v a r i a t i o n d f squares square F P c o r r . Group 1 . 375 .375 .220 .6495 E r r o r 10 17.083 1.708 Task 1 12.042 12.042 10.865 .0081 G x T 1 . 375 .375 .338 .5737 E r r o r 10 12.208 1.221 1.00 T a b l e J - 4 A n a l y s i s o f V a r i a n c e f o r E x p r e s s i o n s o f L e x i c a l U n c e r t a i n t y Source o f Sum o f Mean E p s i l . v a r i a t i o n d f s q u a r e s square F P c o r r . Group 1 1.500 1.500 .545 .4772 E r r o r 10 27.500 2.750 Task 1 28.167 28.167 12.707 .0051 G x T 1 . 667 .8667 .301 .5954 E r r o r 10 22.167 2.217 1.00 T a b l e J - 5 A n a l y s i s o f V a r i a n c e f o r R e f e r e n t i a l Q u e s t i o n s S o u r c e o f Sum o f Mean E p s i l . v a r i a t i o n d f s q u a r e s square F P c o r r . Group 1 5. 042 5. 042 .287 . 6036 E r r o r 10 175.417 17.542 Task 1 630.375 630.375 44.471 . 0001 G x T 1 18.375 18.375 1.296 .2814 E r r o r 10 141.750 14.175 1.00 272 T a b l e J - 6 A n a l y s i s o f V a r i a n c e f o r S e l f - R e p e t i t i o n S o u r c e o f Sum o f Mean E p s i l . v a r i a t i o n d f s q u a r e s square F p_ c o r r . Group 1 16.667 16.667 . 117 .7395 E r r o r 10 1425.667 142.567 Task 1 6. 000 6.000 .169 .6897 G x T 1 294.000 294.000 8.282 .0164 E r r o r 10 266.708 26.671 273 Appendix K Means and S tandard D e v i a t i o n s f o r M i x e d - and Homogeneous-group Tasks T a b l e K - l Means and S t a n d a r d D e v i a t i o n s f o r M i x e d - g r o u p T a s k s Task COMl COM2 DIS L E G l LEG2 Mean Mean Mean Mean Mean V a r i a b l e SD SD SD SD SD A n a p h o r i c r e f e r e n c e dyad 33.67 3.78 t e a c h e r 3 6.17 4.45 l e a r n e r 2.50 2.67 C l a r i f i c a t i o n r e q u e s t dyad 3.67 2.16 t e a c h e r 1.00 2.45 l e a r n e r 2.67 2.25 Comprehens i o n check dyad 2.67 2.73 t e a c h e r 2.67 2.73 l e a r n e r .00 . 00 t a b l e c o n t i n u e s 31.33 11.84 30.33 11.43 1. 00 1.27 40.83 20.02 29.167 16.38 11. 67 6.98 41.83 17.79 38.17 15.74 3 .67 3.27 42 .50 7.79 41.50 7.84 1.00 .89 4.00 3.41 1.33 3.27 2 . 67 3.08 1.17 1.33 . 17 .41 1. 00 1.27 5.50 1.87 .50 .84 5.00 1.90 4, 3 4 3 83 43 50 84 33 01 .83 .75 .83 .75 . 00 . 00 .17 .41 .00 . 00 . 17 .41 5, 4, 5, 4, 00 34 00 34 00 00 .83 .75 .83 .75 .00 .00 274 T a b l e K - l ( c o n t ' d . ) Means and S t a n d a r d D e v i a t i o n s f o r M i x e d - g r o u p Tasks Task C0M1 COM2 DIS LEG1 LEG2 Mean Mean Mean Mean Mean V a r i a b l e SD SD SD SD SD C o n f i r m a t i o n check dyad 3.5 4.33 4. 00 7.33 12.00 2 . 35 2.50 4.43 1.97 10.26 t e a c h e r .33 .33 2.83 .50 .17 .82 .52 4.92 .84 .41 l e a r n e r 3 .17 4.00 1.17 6.83 11.83 2.40 2.83 1.17 2.40 10.48 ' i n i t i o n dyad 2.00 1.50 1.00 . 67 .33 1.27 1.38 1.55 .82 .82 t e a c h e r 1.67 1.33 1.00 . 67 .33 1.37 1.37 1.55 .82 .82 l e a r n e r .33 . 17 .00 .00 .00 .52 .41 . 00 .00 .00 D i s p l a y q u e s t i o n dyad 5. 67 9 .33 2 . 00 . 33 1.00 2.66 3.56 3.16 .82 1.67 t e a c h e r 5.67 9.33 1.83 .33 1.00 2.66 3.56 2.86 .82 1.67 l e a r n e r .00 . 00 . 17 . 00 .00 .00 . 00 .41 .00 .00 Echo dyad 7.67 8.83 4.17 5.50 7.33 1.75 4 . 02 2.48 4.76 4.18 t e a c h e r .33 .17 . 50 .33 . 17 .52 .41 .55 .52 .41 l e a r n e r 7.33 8.67 3.67 5.17 7.17 1.37 4 . 03 2.42 4.88 4.07 t a b l e c o n t i n u e s 275 T a b l e K - l ( c o n t ' d . ) Means and S t a n d a r d D e v i a t i o n s f o r M i x e d - g r o u p T a s k s Task COMl COM2 DIS L E G l LEG2 Mean Mean Mean Mean Mean V a r i a b l e SD SD SD SD SD E x o p h o r i c r e f e r e n c e dyad 1. 67 21.83 4.83 1.50 57.83 1.63 7.14 7.47 1.38 21.78 t e a c h e r 1.17 19.50 2.50 .33 42 .17 1.47 6.89 3.99 .52 18.85 l e a r n e r .50 2 . 33 2 . 33 1.17 15.67 .84 1. 63 3.93 1.17 8.12 L e x i c a l u n c e r t a i n t y dyad 2.17 .67 3 . 67 1.00 .33 1.17 . 52 2.25 1.27 .52 t e a c h e r 1.17 . 67 .83 . 17 .17 1.33 .52 .75 .41 .41 l e a r n e r 1.00 . 00 2.83 .83 .17 1.27 . 00 2 . 32 1.17 .41 O t h e r -e x p a n s i o n dyad 4.00 2 . 67 5. 00 6.17 2.67 3.58 3.27 4 . 00 3 . 37 1.86 t e a c h e r 2.00 2 . 00 2.50 3.33 1.67 1.79 3 .10 2.43 2.66 1.51 l e a r n e r 2.00 . 67 2.50 2.83 1.00 2 .10 .52 2.07 1. 60 1.27 t a b l e c o n t i n u e s 276 T a b l e K - l ( c o n t ' d . ) Means and S t a n d a r d D e v i a t i o n s f o r M i x e d - g r o u p Tasks Task COMl COM2 DIS L E G l LEG2 Mean Mean Mean Mean Mean V a r i a b l e SD SD SD SD SD O t h e r -r e p e t i t i o n dyad 4.67 7.00 7.50 6.83 4.50 1.51 4 . 65 3.78 3.49 2.35 t e a c h e r 2.00 2.83 3.83 3 . 00 2.67 1.41 1.94 1.47 1.10 1.63 l e a r n e r 2 . 67 4.17 3 . 67 3.83 1.83 1.21 3.66 2.81 3.43 2.40 R e f e r e n t i a l q u e s t i o n dyad .83 4.67 17.17 10.50 12.83 .75 5.13 4 .75 5.58 4.62 t e a c h e r .17 2.83 11.00 5.83 5.00 .41 4.07 5.76 3.31 5.59 l e a r n e r . 67 1.83 6.17 4.67 7.83 .82 2 .14 4.67 3 .88 7.20 S e l f -e x p a n s i o n dyad 29.50 24.83 21.17 26.83 36.67 9.52 11. 02 6.31 9.81 20.77 t e a c h e r 26.83 23.33 12.83 24.33 28.67 8.64 10.46 5.71 9.40 11.31 l e a r n e r 2 . 67 1.50 8.33 2.50 8.00 2 . 66 1.23 3.33 1.87 16.77 t a b l e c o n t i n u e s T a b l e K - l ( c o n t ' d . ) Means and S t a n d a r d D e v i a t i o n s f o r M i x e d - g r o u p Tasks Task V a r i a b l e COMl Mean SD COM2 DIS L E G l Mean SD Mean SD Mean SD LEG2 Mean SD S e l f -r e p e t i t i o n dyad t e a c h e r l e a r n e r 25.50 5. 61 22.83 7.78 2 3 67 67 29 . 83 8.26 27.00 6.72 2.83 2.56 16.33 9.93 10.83 9.20 5.50 3.15 33.50 10.75 21.50 12.99 3.00 2.37 24.50 12.50 32.67 11.15 .83 .75 N o t e , n = 6 f o r a l l mixed d y a d s . 278 T a b l e K-2 Means and S t a n d a r d D e v i a t i o n s f o r Homogeneous-group Tasks Task COMl COM2 DIS L E G l LEG2 Mean Mean Mean Mean Mean V a r i a b l e SD SD SD SD SD A n a p h o r i c r e f e r e n c e dyad 27.50 18.67 18.26 7.82 t e a c h e r 22.17 16.00 16.38 5.25 l e a r n e r 5.3 3 2.67 5.01 2.66 C l a r i f i c a t i o n r e q u e s t dyad .83 1.83 1.17 4.50 2.67 .75 1. 47 1.94 2.88 1.86 t e a c h e r . 00 . 17 .33 .50 .50 .00 .41 .52 .84 1.23 l e a r n e r .83 1. 67 .83 4.00 2.17 .75 1. 63 1.60 2.61 .98 Comprehension check dyad 1.00 2.67 .00 5.50 1.17 1.10 2.58 .00 3.21 1.17 t e a c h e r 1.00 2.67 . 00 5.50 1.17 1.10 2 . 58 . 00 3.21 1.17 l e a r n e r . 00 . 00 .00 . 00 . 00 . 00 .00 .00 . 00 . 00 t a b l e c o n t i n u e s 30.50 13.55 20.17 12.07 10. 33 6.77 45.00 12.99 38.00 10.24 7.00 39.17 16.25 36.17 13.38 3 . 00 279 T a b l e K-2 ( c o n t ' d . ) Means and S t a n d a r d D e v i a t i o n s f o r Homogeneous-group Tasks Task C0M1 COM2 DIS LEG1 LEG2 Mean Mean Mean Mean Mean V a r i a b l e SD SD SD SD SD C o n f i r m a t i o n check dyad 3.83 2.33 4.26 2.07 t e a c h e r .17 .50 .41 .84 l e a r n e r 3.67 1.83 3.93 2.14 D e f i n i t i o n dyad 1.33 .50 1.75 .55 t e a c h e r 1.3 3 .50 1.75 .55 l e a r n e r .00 .00 .00 .00 D i s p l a y q u e s t i o n dyad 4.00 12.67 3.80 3.45 t e a c h e r 4.00 12.33 3.80 3.45 l e a r n e r .00 .33 .00 .52 Echo dyad 9.50 8.67 5.36 3.77 t e a c h e r .33 3.3 3 .52 5.35 l e a r n e r 9.17 5.3 3 5.53 3.39 1.33 7.33 12.33 2 . 07 3.39 4.08 .50 . 17 .33 1.23 .41 .52 .83 7.17 12.00 1.60 3 . 06 3.90 .00 .83 .33 .00 .98 .82 .00 .83 .33 . 00 .98 .82 .00 . 00 .00 . 00 .00 .00 2 .33 .50 3.17 3.33 .55 4.58 2.17 . 50 3 .00 3.37 .55 4.69 . 17 .00 . 17 .41 .00 .41 2. 67 8.00 7. 67 2.58 2.37 4.08 1. 17 .67 .00 1.47 .82 .00 1.50 7.33 7.67 1.38 2 .25 4.08 t a b l e c o n t i n u e s 280 T a b l e K-2 ( c o n t ' d . ) Means and S t a n d a r d D e v i a t i o n s f o r Homogeneous-group Tasks Task C0M1 COM2 DIS ' LEG1 LEG2 Mean Mean Mean Mean Mean V a r i a b l e SD SD SD SD SD E x o p h o r i c r e f e r e n c e dyad 10.17 21. 67 1.33 3 . 00 51.17 16.65 3.67 1.97 3.80 15.04 t e a c h e r 8.33 16. 00 .83 1.83 38.83 14.67 4.52 .93 3.60 12.21 l e a r n e r 1.83 5. 67 .50 1.17 12.33 2.99 5.09 1.23 2.40 8.17 L e x i c a l u n c e r t a i n t y dyad 3.00 .83 3.00 2.67 .50 2.83 .98 2.28 1. 03 .55 t e a c h e r 1.17 .50 . 67 1.33 .50 1.94 .84 .52 1.37 .55 l e a r n e r 1.83 . 33 2.33 1.33 .00 2.23 . 52 2 . 07 1. 63 .00 O t h e r -e x p a n s i o n dyad 4.50 4.50 6.50 5. 00 5.50 3.78 3.45 3.39 3.16 4.32 t e a c h e r 2.17 2.83 4.17 2.83 3.00 1.72 2.79 2.04 2.79 2 .19 l e a r n e r 2 . 33 1.67 2.33 2 .17 2.50 2.94 1. 03 2 .16 1.33 2.51 O t h e r -r e p e t i t i o n dyad 7.17 9.17 6.83 7.83 4.17 6.43 5. 00 5. 64 4.17 3.43 t e a c h e r 4.00 5.67 3.33 4.67 3.83 3.29 2 . 58 2.25 2.25 3.19 l e a r n e r 3.17 3 . 50 3.50 3.17 . 33 3.31 2 . 59 3 . 51 2 .99 .52 t a b l e c o n t i n u e s 281 T a b l e K-2 ( c o n t ' d . ) Means and S t a n d a r d D e v i a t i o n s f o r Homogeneous-group Tasks Task COMl COM2 DIS L E G l LEG2 Mean Mean Mean Mean Mean V a r i a b l e SD SD SD SD SD R e f e r e n t i a l q u e s t i o n dyad 3.50 6. 00 13 .33 13 . 17 12. 00 2.88 4 . 19 8 . 38 6. 40 5. 72 t e a c h e r .50 1. 83 10 .83 9. 33 2. 00 1.23 1. 47 8 .98 7. 71 2. 68 l e a r n e r 3 . 00 4. 17 2 .50 3. 83 10. 00 2.83 4. 02 3 .73 3. 76 4. 15 S e l f -e x p a n s i o n dyad 25.50 19. 50 19 . 67 22. 17 20. 33 3.15 5. 72 5 .24 6. 91 5. 13 t e a c h e r 20.83 17. 33 10 .83 21 . 00 20. 33 6.43 5. 54 6 .56 6. 78 5. 13 l e a r n e r 4 . 67 2 . 17 8 .83 1. 17 • 00 4.13 2 . 32 6 .49 1. 33 • 00 S e l f -r e p e t i t i o n dyad 34 .17 28. 33 12 .00 28. 17 21. 50 8.28 11. 74 3 .41 11. 86 7. 82 t e a c h e r 27.83 26. 33 7 .67 18. 17 28. 00 11. 57 11. 41 4 .59 7. 25 11. 97 l e a r n e r 6.33 2 . 00 4 . 33 3. 33 • 17 4 . 68 1. 79 3 .83 2 . 94 m 41 N o t e , n = 6 f o r a l l homogeneous d y a d s . 282