TESOL Quarterly (Spring 1995)

TABLE OF CONTENTS
A Journal for Teachers of English to Speakers of Other Languages and of Standard English as a Second Dialect
Editor
SANDRA McKAY, San Francisco State University
BONNY NORTON PEIRCE, Ontario Institute for Studies in Education
Teaching Issues Editor
Brief Reports and Summaries Editors

GRAHAM CROOKES and KATHRYN A. DAVIS, University of Hawaii
at Manoa Review Editor
H. DOUGLAS BROWN, San Francisco State University MARILYN KUPETZ, TESOL Central Office
Assistant Editor
Editorial Assistant
CATHERINE HARTMAN, San Francisco State University Lyle Bachman, Kathleen Bardovi-Harlig, Anna Uhl Chamot Keith Chick, Deborah Curtis, Sharon Hilles,
University of Natal San Francisco State University Indiana University Georgetown University University of California, Los Angeles
Editorial Advisory Board

Anne Lazaraton, Patsy M. Lightbown, Brian Lynch,
Concordia University University of Melbourne The Pennsylvania State University
Peter Master,
Mary McGroarty,
California State University, Fresno
Rebeca L. Oxford,
Northern Arizona University
California State Polytechnic University, Pomona
Bonny Norton Peirce, Patricia Porter, Kamal Sridhar,

Ontario Institute for Studies in Education
University of Alabama
Nancy Hornberger, Sarah Hudelson, Thom Hudson,

University of Pennsylvania Arizona State University University of Hawaii at Manoa University of Arizona
San Francisco State University State University of New York
Donna M. Johnson,
Additional Readers
Bradford Arthur, Elsa Auerbach, Ellen Block, H. Douglas Brown, Joan G. Carson, Graham Crcokes, Patricia A. Dunkel, Rod Ellis, Sandra Fotos, Jan Frodesen, Linda Harklau, William Harshbarger, Barbara Hoekje, Thomas Huckin, Ann M.Johhs, Michael H. Long, John Murphy, Patricia Nichols, David Nunan,Cindy Pease-Alvarez, Teresa Pica, Thomas Scovel, May Shih, Ruth Spack, Jean Turner, Gail Weinstein-Shr, Vivian Zamel
Credits
Advertising arranged b Jennifer Delett, TESOL Central Office, Alexandria, Virginia Typesetting by World Composition Services, Inc. Printing and binding by Pantagraph Printing, Bloomington, Illinois Copies of articles that appear in the TESOL Quarterly are available through The Genuine Article, 3501 Market Street, Philadelphia, Pennsylvania 19104 U.S.A.
Copyright 1995 Teachers of English to Speakers of Other Languages, Inc. US ISSN 0039-8 322
VOLUMES MENU
QUARTERLY
CONTENTS
Click any article to jump directly to that article.
ARTICLES
Social Identity, Investment, and Language Learning 9 Bonny Norton Peirce Student Reactions to Teacher Response in Multiple-draft Composition Classrooms 33 Dana Ferris On the Teachability of Communication Strategies 55 Zoltn Drnyei Interpretation Tasks for Grammar Teaching 87 Rod Ellis The Role of Lexical Aspect in the Acquisition of Tense and Aspect 107 Kathleen Bardovi-Harlig and Dudley W. Reynolds 133 Closing the Gap Between Learning and Instruction David Nunan
To print, select PDF page nos. in parentheses. (10-32) (34-54) (56-86) (88-106) (108-132) (134-159)
THE FORUM
Comments on Joy Reids Responding to ESL Students Texts: 159 The Myths of Appropriation There are Myths and Then There are Myths . . . Chris Hall The Author Responds . . . Joy Reid Comments on Virginia LoCastros Learning Strategies and Learning Environments 166 Two Readers React . . . Rebecca L. Oxford and John M. Green The Author Responds . . . Virginia LoCastro Comments on B. Kumaravadivelus The Postmethod Condition: (E)merging Strategies for Second/Foreign Language Teaching Alternative to or Addition to Method? Dilin Liu The Author Responds . . . B. Kumaravadivelu
174
TEACHING ISSUES
180 Language Tests and ESL Teaching Examining Standardized Test Content: Some Advice for Teachers Felicia DeVincenzi Assessing Student Performance in the ESL Classroom J. Charles Alderson and Caroline Clapham
Volume 29, Number 1 u Spring 1995
BRIEF REPORTS AND SUMMARIES

Mainstream Classroom Teachers and ESL Students Nancy Clair 189
REVIEWS
How Languages are Learned 197 Patsy M. Lightbown and Nina Spada Reviewed by Kenneth Rose The Discovery of Competence: Teaching and Learning with Diverse Student Writers Eleanor Kutz, Suzy Q. Groden, and Vivian Zamel Reviewed by Trudy Smoke The Other Tongue (2nd ed.) Braj Kachru Reviewed by Kimberley Brown
BOOK NOTICES
Classroom Observation Tasks, 207 Ruth Wajnryb (James Riedel) Literature and Language Teaching: A Guide for Teachers and Teacher Trainers, Gillian Lazar (Gina Keefer) Longman Language Activator: The Worlds First Production Dictionary, (Cheryl Eason and Robert Yates) Process Your ThoughtsWriting with Computers, Marianne Phinney (Gladys Vega Scott) Information for Contributors 213 Editorial Policy General Information for Authors Publications Received 219 Publications Available from the TESOL Central Office TESOL Membership Application
221
iS an international professional organization for those concerned with the teaching of English as a second or foreign language and of standard English as a second dialect. TESOLS mission is to strengthen the effective teaching and learning of English around the world while respecting individuals language rights. Information about membership and other TESOL services is available from TESOL Central Office at the address below. TESOL Quarterly is published in Spring, Summer, Autumn, and Winter. Contributions should be sent to the Editor or the appropriate Section Editors at the addresses listed in the Information for Contributors section. Publishers representative is Helen Kornblum, Director of Communications & Marketing. All material in TESOL Quarterly is copyrighted. Copying without the permission of TESOL, beyond the exemptions specified by law, is an infringement involving liability for damages. TESOL Journal is published in Spring, Summer, Autumn, and Winter. Send contributions to Elliot Judd, Editor, TESOL Journal, Department of English (M/C 162), University of Illinois at Chicago, 601 S. Morgan, Chicago, Illinois 60607 U.S.A. Announcements should be sent directly to the Editor, TESOL Matters, 2 months prior to the month of publication desired and must be received by the first of that month (e.g., February 1 for the April issue). Use Central Office address below. TESOL Matters is published in February, April, June, August, October, and December. Neither TESOL Quarterly nor TESOL Journal publishes announcements. Advertising in all TESOL publications is arranged by the marketing assistant, TESOL Central Office, Suite 300, 1600 Cameron Street, Alexandria, Virginia 22314-2751 U. S. A., Tel. 703836-0774. Fax 703-836-7864. E-mail [email protected]
OFFICERS AND EXECUTIVE BOARD 1994-95

President FRED GENESEE McGill University Montreal, Canada First Vice President JOY REID University of Wyoming Laramie, WY Second Vice President MARY ANN CHRISTISON Snow College Ephraim, UT Executive Director SUSAN C. BAYLEY Alexandria, VA Treasurer MARTHA EDMONDSON Washington, DC Jo Ann Aebersold Eastern Michigan University Ypsilanti, MI Kathleen Bailey Monterey Institute of International Studies Monterey, CA Nick Collins Capilano College Vancouver, Canada Ellen Ducy Perez Universidad National Pedro Henriquez Urena Santo Domingo, Dominican Republic Donald Freeman School for International Training Brattleboro, VT Mary Hines Hines Video Design New York, NY Lars-ke Kll Fredricka Bremerskolan Stockholm, Sweden Denise E. Murray San Jos State University San Jos, CA David Nunan University of Hong Kong Hong Kong Sharon Seymour City College of San Francisco San Francisco, CA Pat Spring Chatswood High School Sydney, Australia Susan Stempleski City University of New York New York, New York Beth Witt Chinle Public Schools Chinle, AZ
QUARTERLY
Editors Note
s This is the last issue to include book notices submitted by the readers.
Starting in the summer issue, book notices (i.e., one-paragraph descriptions of recent publications) will be compiled by the book review editor.
In this Issue
s The lead article in this issue examines the relationship between the
language learner, the social context, and individual identity. The next two articles, both classroom-based research studies, report on teacher responses to ESL compositions and the teaching of communication strategies. Grammar teaching is the focus of the next two articles. Whereas the first deals with theoretical issues in teaching grammar, the second reports on a study of adult learners acquisition of past tense. Both articles, however, present examples of pedagogical strategies for teaching grammar. The final article discusses how learners can become more involved in their own learning process.
q Maintaining that second language acquisition theorists have not devel-
oped a comprehensive theory of social identity, Bonny Norton Peirce argues for a concept of the individual that depicts individual identity as multiple, a site of struggle, and subject to change. In the article, she reports on her study of immigrant women in Canada to illustrate the complex relationship between language learners and the target language, arguing for a concept of investment rather than motivation to explain learners often ambivalent desire to speak. In closing, Peirce maintains that second language teachers need to help language learners claim the right to speak outside of the classroom, and she suggests what she terms classroom-based social research as a pedagogical strategy to help learners improve their language learning outside of the classroom and claim the right to speak. 5
q Dana Ferris reports on ESL students reactions to teacher response
in a multiple-draft composition setting. Surveying the students in two levels of a university ESL composition course, Ferris found that students pay more attention to teachers comments on their preliminary drafts than to final drafts, that they pay the most attention to comments on grammar, and that overall they find their teachers comments helpful. The survey also showed that students had a variety of problems in understanding their teachers comments. In conclusion, Ferris discusses the implications of the study in terms of the role of grammar feedback, the need to clarify teachers responding behavior, and the role of positive comments versus constructive criticism.
q In another classroom-based research study, Zoltn Drnyei examines
whether or not communication strategies can be taught. He suggests three possible reasons for the controversy surrounding the teachability of communication strategies: the use of indirect evidence to support claims of teachability, the range of strategies included in discussing communication strategies, and varying definitions of the concept of teaching. He then reports on his study of communication strategy training with Hungarian secondary students. The results of the study suggest that both the quality and quantity of learners use of some communication strategies does improve with focused instruction.
q In the first of two articles on grammar teaching, Rod Ellis points out
that traditional grammar teaching has focused on having learners produce specific grammatical structures. Ellis argues that such an approach does not take into account the fact that learners pass through a number of stages in learning the language so that requiring them to produce particular structures may be difficult. Hence, he proposes an alternative approach to grammar teaching that focuses learners attention on a target structure using what he terms interpretation tasks. In the article, Ellis provides a rationale for this approach to grammar teaching and delineates principles for designing interpretation tasks.
q Kathleen Bardovi-Harlig and Dudley W. Reynolds report on the
results of a study of adult learners use of the simple past tense. The study shows that the acquisition of the past tense occurs in stages that are determined by the meaning of verbs termed the lexical aspect of verbs. The study suggests that even advanced language learners have a low rate of appropriate use of the simple past tense with activity and state verbs and with the use of the past with adverbs of frequency. Supporting Elliss emphasis on teaching grammar through focused practice, they close by describing a pedagogical approach for increasing the appropriate use of the past tense by providing students with contextualized examples of the past tense in authentic texts and by using focused noticing exercises. 6 TESOL QUARTERLY
q In his article, David Nunan explores ways in which the relationship
between teachers and learners and teaching and learning can be strengthened. Specifically he examines how this relationship can be strengthened in terms of experiential content, the learning process, and language content. He argues that in all domains the key is learner centredness, which he maintains is not an all-or-nothing concept but a relative matter ranging from awareness to transcendence. Throughout the article, he provides examples of activities that can be used to increase the depth of learner involvement in the learning process. Also in this issue:
q The Forum: Chris Halls commentary on Joy Reids Responding to
ESL Students Texts: The Myths of Appropriation is followed by a response from the author. Rebecca L. Oxford and John M. Green respond to a Brief Reports and Summaries article, Learning Strategies and Learning Environments, written by Virginia LoCastro. Finally, Dilin Liu and B. Kumaravadivelu exchange comments regarding B. Kumaravadivelus article, The Postmethod Condition: (E)merging Strategies for Second/Foreign Language Teaching.
q Brief Reports and Summaries: Nancy Clair reports on a qualitative
study investigating the beliefs, practices, and professional development needs of three mainstream classroom teachers who have ESL students in their classes.
q Teaching Issues: Felicia DeVincenzi exchanges views with J. Charles
Alderson and Caroline Clapham on language tests and ESL teaching.

q Reviews: Kenneth Rose reviews Patsy M. Lightbown and Nina Spadas
book How Languages are Learned. Eleanor Kutz, Suzy Q. Groden, and Vivian Zamels The Discovery of Competence: Teaching and Learning with Diverse Student Writers is reviewed by Trudy Smoke, and Kimberley Brown reviews Braj Kachrus second edition of The Other Tongue.
q Book Notices: Four ESL texts are highlighted in this section.
Sandra McKay
IN THIS ISSUE
NEW WAYS SERIES II

TESOL invites contributions to nine new volumes of innovative classroom techniques. Moving beyond the traditional skills-based approach to language instruction, New Ways Series II offers volumes on how to incorporate technology, culture, innovative assessment, content-based lessons, authentic materials, communicative games, drama, and literature into many kinds of classroom activities and settings. Series editor is Jack C. Richards. The following specialists will edit each volume: Tim Boswood, Using Technology in Language Teaching Donna Brinton, Content-Based Instruction J.D. Brown, Using Alternative Assessment in Language Teaching Alvino Fantini, Teaching Culture Ruth Larimer, Using Authentic Materials in Language Teaching Marilyn Lewis, Teaching Adults Deborah Short, Teaching English in Secondary Programs Makhan Tickoo, Using Communicative Games in Language Teaching Valerie Whiteson, Using Drama and Literature in Language Teaching
q q q q q q q q q
Guidelines for contributors

TESOL professionals are invited to contribute to one or more of the series. More than one contribution may also be submitted for any particular book in the series. Contributors will not be paid but their names will appear with their submissions.
How to contribute
Contributors should send a request for guidelines, indicating the volume to which the contributor would like to submit manuscripts, to TESOL Central Office. Include a selfaddressed envelope with US$.64 postage for every two sets of guidelins requested. Requests for more than two sets of guidelines will require additional envelopes with postage provided. Members outside the US do not have to provide postage on their envelopes. Detailed guidelines will then be provided on how to develop, reference, and format submissions and to which address to send them. Contributors must follow the submission guidelines. Send a selfaddressed envelope with your request for guidelines. To: TESOL Central Office q New Ways Series Alexandria, Virginia 22314-2751 USA
q
1600 Cameron St., Suite 300l
TESOL QUARTERLY Vol. 29, No. 1, Spring 1995
Social Identity, Investment, and Language Learning*

BONNY NORTON PEIRCE
Ontario Institute for Studies in Education
The author argues that second language acquisition (SLA) theorists have struggled to conceptualize the relationship between the language learner and the social world because they have not developed a comprehensive theory of social identity which integrates the language learner and the language learning context. She also maintains that SLA theorists have not adequately addressed how relations of power affect interaction between language learners and target language speakers. Using data collected in Canada from January to December 1991 from diaries, questionnaires, individual and group interviews, and home visits, the author illustrates how and under what conditions the immigrant women in her study created, responded to, and sometimes resisted opportunities to speak English. Drawing on her data analysis as well as her reading in social theory, the author argues that current conceptions of the individual in SLA theory need to be reconceptualized, and she draws on the poststructuralist conception of social identity as multiple, a site of struggle, and subject to change to explain the findings from her study. Further, she argues for a conception of investment rather than motivation to capture the complex relationship of language learners to the target language and their sometimes ambivalent desire to speak it. The notion of investment conceives of the language learner, not as ahistorical and unidimensional, but as having a complex social history and multiple desires. The article includes a discussion of the implications of the study for classroom teaching and current theories of communicative competence. Everybody working with me is Canadian. When I started to work there, they couldnt understand that it might be difficult for me to understand everything and know about everything what its normal for them. To explain it more clearly I can write an
*Earlier drafts of this paper were presented at the Social Issues/Social Change Conference in Toronto, Canada, in July 1993, and the 28th Annual TESOL convention in Baltimore, United States, in March 1994.
example, which happened few days ago. The girl [Gail] which is working with me pointed at the man and said: Do you see him?I said Yes, Why? Dont you know him? No. I dont know him. How come you dont know him. Dont you watch TV. Thats Bart Simpson. It made me feel so bad and I didnt answer her nothing. Until now I dont know why this person was important. Eva, February 8, 1991 1 No researcher today would dispute that language learning results from participation in communicative events. Despite any claims to the contrary, however, the nature of this learning remains undefined. Savignon, 1991, p. 271
ow would second language acquisition (SLA) theorists conceptualize the relationship between Eva, an immigrant language learner, and Gail, an anglophone Canadian, both of whom are located in the same North American workplace in the 1990s? Because they have struggled to conceptualize the relationship between the individual language learner and larger social processes, a question such as this poses a problem for SLA theorists. In general, many SLA theorists have drawn artificial distinctions between the language learner and the language learning context. On the one hand, the individual is described with respect to a host of affective variables such as his/her motivation to learn a second language. Krashen (1981, 1982), for example, has hypothesized that comprehensible input in the presence of a low affective filter is the major causal variable in SLA. In Krashens view, this affective filter comprises the learners motivation, self-confidence, and anxiety stateall of which are variables that pertain to the individual rather than the social context. Furthermore, the personality of the individual has been described unidimensionally as introverted or extroverted, inhibited or uninhibited, field dependent or field independent. 2 With reference to these theories, Eva might be described as someone who is unmotivated with a high affective filter; perhaps an introverted personality who is unable to interact appropriately with her interlocutors. Or she might be portrayed as a poor language learner who has not developed sociolinguistic competence.
Quoted in Peirce, 1993, p. 197. Eva explained that the man her co-worker pointed to had a Bart Simpson t-shirt on. Spelling mistakes in the original have been corrected. 2 See Brown (1987) for an overview of the literature on personality variables and language learning.
1
10
TESOL QUARTERLY
Other theories of SLA focus on social rather than individual variables in language learning. The social frequently refers to group differences between the language learner group and the target language group (Schumann, 1976). In this view, where there is congruence between the second language group and the target language group, what Schumann (1976) terms social distance between them is considered to be minimal, in turn facilitating the acculturation of the second language group into the target language group and enhanced language learning. Where there is great social distance between two groups, little acculturation is considered to take place, and the theory predicts that members of the second language group will not become proficient speakers of the target language. Supporters of the Acculturation Model of SLA (Schumann, 1978) might argue that despite the fact that Eva and Gail are in contact, there is great social distance between them because there is little congruence between Evas culture and that of Gail. For this reason, Eva might struggle to interact successfully with members of the target language community. Because of the dichotomous distinctions between the language learner and the social world, there are disagreements in the literature on the way affective variables interact with the larger social context. For example, although Krashen regards motivation as a variable independent of social context, Spolsky (1989) regards the two as inextricably intertwined. Although Krashen draws distinctions between selfconfidence, motivation, and anxiety, Clement, Gardner, and Smythe (quoted in Spolsky, 1989) consider motivation and anxiety as a subset of self-confidence. Although Krashen considers self-confidence as an intrinsic characteristic of the language learner, Gardner (1985) argues that self-confidence arises from positive experiences in the context of the second language: Self-confidence . . . develops as a result of positive experiences in the context of the second language and serves to motivate individuals to learn the second language (p. 54). Such disagreements in the SLA literature should not be dismissed, as Gardner (1989) dismisses them, as more superficial than real (p. 137). I suggest that this confusion arises because artificial distinctions are drawn between the individual and the social, which lead to arbitrary mapping of particular factors on either the individual or the social, with little rigorous justification. In the field of SLA, theorists have not adequately addressed why it is that a learner may sometimes be motivated, extroverted, and confident and sometimes unmotivated, introverted, and anxious; why in one place there may be social distance between a specific group of language learners and the target language community, whereas in another place the social distance may be minimal; why a learner can sometimes speak and other times remains silent. Although muted, there is an uneasy recognition by some theorists that
SOCIAL IDENTITY, INVESTMENT, AND LEARNING 11
current theory about the relationship between the language learner and the social world is problematic. Scovel (1978) for example, has found that research on foreign language anxiety suffers from several ambiguities, and Gardner and MacIntyre (1993) remain unconvinced of the relationship between personality variables (p. 9) and language achievement. The central argument of this paper is that SLA theorists have not developed a comprehensive theory of social identity that integrates the language learner and the language learning context. Furthermore, they have not questioned how relations of power in the social world affect social interaction between second language learners and target language speakers. Although many SLA theorists (Ellis, 1985; Krashen, 1981; Schumann, 1978; Spolsky, 1989; Stern, 1983) recognize that language learners do not live in idealized, homogeneous communities but in complex, heterogeneous ones, such heterogeneity has been framed uncritically. Theories of the good language learner have been developed on the premise that language learners can choose under what conditions they will interact with members of the target language community and that the language learners access to the target language community is a function of the learners motivation. Thus Gardner and MacIntyre (1992), for example, argue that the major characteristic of the informal context is that it is voluntary. Individuals can either participate or not in informal acquisition contexts (p. 213). SLA theorists have not adequately explored how inequitable relations of power limit the opportunities L2 learners have to practice the target language outside the classroom. In addition, many have assumed that learners can be defined unproblematically as motivated or unmotivated, introverted or extroverted, inhibited or uninhibited, without considering that such affective factors are frequently socially constructed in inequitable relations of power, changing over time and space, and possibly coexisting in contradictory ways in a single individual. Drawing on a recent study (Peirce, 1993) as well as my reading in social theory, I will propose a theory of social identity that I hope will contribute to debates on second language learning. This theory of social identity, informed by my data, assumes that power relations play a crucial role in social interactions between language learners and target language speakers. In March 1991, for example, when I asked Eva why the communication breakdown between her and Gail had taken place, Eva indicated she had felt humiliated at the time. She said that she could not respond to Gail because she had been positioned as a strange woman. What had made Eva feel strange? When I analyzed Evas data more closely, I realized that Gails questions to Eva were in fact rhetorical. Gail did not expect, or possibly even desire a
12 TESOL QUARTERLY
response from Eva: How come you dont know him. Dont you watch TV. Thats Bart Simpson. It was Gail and not Eva who could determine the grounds on which interaction could proceed; it was Gail and not Eva who had the power to bring closure to the conversation. If, as Savignon (1991) argues, language learning results from participation in communicative events, it is important to investigate how power relations are implicated in the nature of this learning. I therefore take the position that notions of the individual and the language learners personality in SLA theory need to be reconceptualized in ways that will problematize dichotomous distinctions between the language learner and the language learning context. I argue that SLA theory needs to develop a conception of the language learner as having a complex social identity that must be understood with reference to larger, and frequently inequitable social structures which are reproduced in day-to-day social interaction. In taking this position, I foreground the role of language as constitutive of and constituted by a language learners social identity. It is through language that a person negotiates a sense of self within and across different sites at different points in time, and it is through language that a person gains access toor is denied access topowerful social networks that give learners the opportunity to speak (Heller, 1987). Thus language is not conceived of as a neutral medium of communication but is understood with reference to its social meaning. I support these arguments with findings from a longitudinal case study of the language learning experiences of a group of immigrant women in Canada (Peirce, 1993).
THE STUDY: IMMIGRANT WOMEN AS LANGUAGE LEARNERS

From January to June 1990 I helped teach a 6-month ESL course to a group of recent immigrants at Ontario College in Newtown, Canada.3 After the course was complete, I invited the learners to participate in a longitudinal case study of their language learning experiences in Canada. Five women agreed to participate in the study: Mai from Vietnam, Eva and Katarina from Poland, Martina from Czechoslovakia, and Felicia from Peru. My research questions were divided into two parts:
Part I How are the opportunities for immigrant women in Canada to practice ESL socially structured outside the classroom? How do immigrant women
3
The names of places and participants have been changed to protect tbe identities of participants.
SOCIAL IDENTITY, INVESTMENT, AND LEARNING
13
respond to and act upon these social structures to create, use, or resist opportunities to practice English? To what extent should their actions be understood with reference to their investment in English and their changing social identities across time and space? Part II How can an enhanced understanding of natural language learning and social identity inform SLA theory, in general, as well as ESL pedagogy for immigrant women in Canada? (Peirce, 1993, p. 18)
The study lasted 12 monthsfrom January to December 1991. A major source of data collection was a diary study: From January to June 1991, the participants kept records of their interactions with anglophone Canadians and used diaries to reflect on their language learning experiences in the home, workplace, and community. During the course of the study, we met on a regular basis to share some of the entries the women had made in their diaries and to discuss their insights and concerns. I also drew a substantial amount of data from two detailed questionnaires I administered before and after the study, as well as personal and group interviews, and home visits. One of the assumptions on which I based my research questions was that practice in the target language is a necessary condition of second language learning. As Spolsky (1989) argues, extensive exposure to the target language, in relevant kinds and amounts, and the opportunity to practice the target language are essential for second language learning: Learning cannot proceed without exposure and practice. These conditions, furthermore, are graded: The more exposure and practice, the more proficient the learner will become. Spolsky (1989) argues that the language learner can have exposure to and practice in the target language in two qualitatively different settings: the natural or informal environment of the target language community or the formal environment of the classroom. The focus of my research was on the natural language learning experiences of the women in their homes, workplaces, and communities.
THE THEORY: SOCIAL IDENTITY, INVESTMENT, AND THE RIGHT TO SPEAK Social Identity as Multiple, a Site of Struggle, and Changing Over Time
In examining the relationship between the language learners in my study and the social worlds in which they lived, I drew in particular on
14 TESOL QUARTERLY
Weedons (1987) conception of social identity or subjectivity. Feminist poststructuralism, like much postmodern educational theory (Cherryholmes, 1988; Giroux, 1988; Simon, 1992), explores how prevailing power relations between individuals, groups, and communities affect the life chances of individuals at a given time and place. Weedons work, however, is distinguished from that of other postmodern theorists in the rigorous and comprehensive way in which her work links individual experience and social power in a theory of subjectivity. Weedon (1987) defines subjectivity as the conscious and unconscious thoughts and emotions of the individual, her sense of herself and her ways of understanding her relation to the world (p. 32). Furthermore, like other poststructuralist theorists who inform her work (Derrida, Lacan, Kristeva, Althusser, and Foucault), Weedon does not neglect the central role of language in her analysis of the relationship between the individual and the social: Language is the place where actual and possible forms of social organization and their likely social and political consequences are defined and contested. Yet it is also the place where our sense of ourselves, our subjectivity, is constructed (p. 21). Three defining characteristics of subjectivity, as outlined by Weedon, are particularly important for understanding my data: the multiple nature of the subject; subjectivity as a site of struggle; and subjectivity as changing over time. First, Weedon (1987) argues, the terms subject and subjectivity signify a different conception of the individual from that associated with humanist conceptions of the individual dominant in Western philosophy. Whereas humanist conceptions of the individualand most definitions of the individual in SLA researchpresuppose that every person has an essential, unique, fixed, and coherent core (introvert/extrovert; motivated/unmotivated; field dependent/ field independent), poststructuralism depicts the individual as diverse, contradictory, and dynamic; multiple rather than unitary, recentered rather than centered. By way of example (and at the risk of oversimplification) a humanist might be attracted by a book with the title How to Discover Your True Self. A poststructuralist, on the other hand, might prefer a book titled Its OK to Live with Contradictions. Second, the conception of social identity as a site of struggle is an extension of the position that social identity is multiple and contradictory. Subjectivity is produced in a variety of social sites, all of which are structured by relations of power in which the person takes up different subject positionsteacher, mother, manager, criticsome positions of which may be in conflict with others. In addition, the subject is not conceived of as passive; he/she is conceived of as both subject of and subject to relations of power within a particular site, community, and society: The subject has human agency. Thus the subject positions that a person takes up within a particular discourse
are open to argument: Although a person may be positioned in a particular way within a given discourse, the person might resist the subject position or even set up a counterdiscourse which positions the person in a powerful rather than marginalized subject position. Third, in arguing that subjectivity is multiple, contradictory, and a site of struggle, feminist poststructuralism highlights the changing quality of a persons social identity. As Weedon (1987) argues, the political significance of recentering the subject and abandoning the belief in essential subjectivity is that it opens up subjectivity to change (p. 33). This is a crucial point for second language educators in that it opens up possibilities for educational intervention. I will demonstrate below that although it might be tempting to argue that Eva was essentially an introverted language learner, the data which follows provides convincing evidence that Evas social identity was not fixed; it was a site of struggle and changed dramatically over time as did her interactions with anglophone Canadians. At the time of the Bart Simpson exchange, however, Gail was in a powerful subject position and Eva did not actively resist being positioned as strange. Because of the construction of Evas social identity in Canada as immigrant, the social meaning of Gails words to her were understood by Eva in this context. Had Eva been, for example, an anglophone Canadian who endorsed public rather than commercial television, she could have set up a counterdiscourse to Gails utterance, challenging Gails interest in popular culture. However, because of the unequal relations of power between Gail and Eva at that point in time, it was Gail who was subject of the discourse on Bart Simpson; Eva remained subject to this discourse. Thus while Eva had been offered the opportunity to engage in social interaction, to practice her English, her subject position within the larger discourse of which she and Gail were a part undermined this opportunity: It made me feel so bad and I didnt answer her nothing. This discourse must be understood not only in relation to the words that were said, but in relationship to larger structures within the workplace, and Canadian society at large, in which immigrant language learners often struggle for acceptance in Canadian society.
From Motivation to Investment

A logical extension of reconceptualizing notions of the individual in SLA theory is the need to problematize the concept of motivation. In the field of second language learning, the concept of motivation is drawn primarily from the field of social psychology, where attempts have been made to quantify a learners commitment to learning the target language. The work of Gardner and Lambert (1972) and Gard16 TESOL QUARTERLY
ner (1985) has been particularly influential in introducing the notions of instrumental and integrative motivation into the field of SLA. In their work, instrumental motivation references the desire that language learners have to learn a second language for utilitarian purposes, such as employment, whereas integrative motivation references the desire to learn a language to integrate successfully with the target language community. Such conceptions of motivation, which are dominant in the field of SLA, do not capture the complex relationship between relations of power, identity, and language learning that I have been investigating in my study of immigrant women. In my view, the conception of investment rather than motivation more accurately signals the socially and historically constructed relationship of the women to the target language and their sometimes ambivalent desire to learn and practice it. My conception of investment has been informed by my reading in social theory, although I have not as yet found a comprehensive discussion of the term in these contexts. It is best understood with reference to the economic metaphors that Bourdieu (1977) uses in his workin particular the notion of cultural capital. Bourdieu and Passeron (1977) use the term cultural capital to reference the knowledge and modes of thought that characterize different classes and groups in relation to specific sets of social forms. They argue that some forms of cultural capital have a higher exchange value than others in a given social context. I take the position that if learners invest in a second language, they do so with the understanding that they will acquire a wider range of symbolic and material resources,4 which will in turn increase the value of their cultural capital. Learners will expect or hope to have a good return on that investmenta return that will give them access to hitherto unattainable resources. Furthermore, drawing on Ogbu (1978), I take the position that this return on investment must be seen as commensurate with the effort expended on learning the second language. It is important to note that the notion of investment I am advocating is not equivalent to instrumental motivation. The conception of instrumental motivation generally presupposes a unitary, fixed, and ahistorical language learner who desires access to material resources that are the privilege of target language speakers. In this view, motivation is a property of the language learnera fixed personality trait. The notion of investment, on the other hand, attempts to capture the relationship of the language learner to the changing social world. It conceives of the language learner as having a complex social identity
4
By symbolic resources I refer to such resources as language, education, and friendship, whereas I use the term material resources to include capital goods, real estate, and money.
17
and multiple desires. The notion presupposes that when language learners speak, they are not only exchanging information with target language speakers but they are constantly organizing and reorganizing a sense of who they are and how they relate to the social world. Thus an investment in the target language is also an investment in a learners own social identity, an identity which is constantly changing across time and space.
Communicative Competence and the Right to Speak

Given the position that communication and social interaction are implicated in the construction of a language learners social identity, my research on immigrant women in Canada develops questions I have raised in earlier research (Peirce, 1989) about the way Hymes (1971) views on communicative competence have been taken up by many theorists in the field of second language learning over the past 15 years. I have argued (Peirce, 1989) that although it is important for language learners to understand the rules of use of the target language, it is equally important for them to explore whose interests these rules serve. What is considered appropriate usage is not selfevident but must be understood with reference to relations of power between interlocutors. I take the position that theories of communicative competence in the field of second language learning should extend beyond an understanding of the appropriate rules of use in a particular society, to include an understanding of the way rules of use are socially and historically constructed to support the interests of a dominant group within a given society. Drawing on Bourdieu (1977), I argue in this paper that the definition of competence should include an awareness of the right to speakwhat Bourdieu calls the power to impose reception (p. 75). His position is that the linguist takes for granted the conditions for the establishment of communication: that those who speak regard those who listen as worthy to listen and that those who listen regard those who speak as worthy to speak. However, as Bourdieu argues, it is precisely such assumptions that must be called into question.
THE ANALYSIS: IDENTITY, INVESTMENT, AND LANGUAGE LEARNING

Although the findings from my study are extensive (Peirce, 1993), I wish to highlight data that address the question, How can an enhanced understanding of natural language learning and social identity inform SLA theory? First, I will address how the notion of investment helps
18 TESOL QUARTERLY
explain the contradictions between the womens motivation to learn English and their sometimes ambivalent desire to speak it. Second, I highlight data from two of the participantsMartina and Evato analyze the relationship between investment, social identity, and language learning.
Investment and Social Identity

All the participants in the study were highly motivated to learn English. They all took extra courses to learn English; they all participated in the diary study; they all wished to have more social contact with anglophone Canadians; and all of them, except Martina, indicated that they felt comfortable speaking English to friends or people they knew well. It is significant, however, that all the women felt uncomfortable talking to people in whom they had a particular symbolic or material investment. Eva, who came to Canada for economical advantage, 5 and was eager to work with anglophones, practice her English and get better jobs, was silenced when the customers in her workplace made comments about her accent. Mai, who came to Canada for her life in the future and depended on the wishes of management for her job security and financial independence, was most uncomfortable speaking to her boss. Katarina, who came to Canada to escape a communist and atheistic system, and had a great affective investment in her status as a professional, felt most uncomfortable talking to her teacher, the doctor, and other anglophone professionals. Martina, who had given up a surveyors job to come to Canada for the children, was frustrated and uncomfortable when she could not defend her familys rights in the public world. Felicia, who had come to Canada to escape terrorism, and had great affective investment in her Peruvian identity, felt most uncomfortable speaking English in front of Peruvians who speak English fluently. The concept of motivation as currently taken up in the SLA literature conceives of the language learner as having a unified, coherent identity which organizes the type and intensity of a language learners motivation. The data indicate that motivation is a much more complex matter than hitherto conceived. Despite being highly motivated, there were particular social conditions under which the women in my study were most uncomfortable and unlikely to speak (See also Auerbach & McGrail, 1991; Cumming & Gill, 1992; Goldstein, 1991; Peirce, Harper, & Burnaby, 1993; Rockhill, 1987). The data suggest that a language learners motivation to speak is mediated by investments
5
The only alterations that have been made to the written contributions of the participants are spelling corrections.
19
that may conflict with the desire to speak. Paradoxically, perhaps, the decision to remain silent or the decision to speak may both constititute forms of resistance to inequitable social forces. For example, although Felicia resisted speaking English in front of strangers because she did not want to be identified as an immigrant in Canada, other immigrant language learners are anxious to speak English for the express purpose of resisting unscrupulous social practices. For example, in his Torontobased study of Spanish-speaking immigrants, Klassen (1987) found that some language learners wanted to learn English as a means of defence in their daily lives. An understanding of motivation should therefore be mediated by an understanding of learners investments in the target languageinvestments that are closely connected to the ongoing production of a language learners social identity. This position will be defended more comprehensively in the following discussion of Martina and Evas experiences of learning English in Canada. In the following discussion, I demonstrate how the conception of social identity as multiple, a site of struggle, and subject to change helps to explain the conditions under which Martina and Eva spoke or remained silent.
Martina: Social identity as Multiple and a Site of Struggle

Martina was born in Czechoslovakia in 1952. She came to Canada in March 1989 when she was 37 years old, with her husband Petr and their three children (Jana 17, Elsbet 14, Milos 11 at the time). She came to Canada for a better life for children. Neither she nor her husband knew any English before they came to Canada, but her children had received some English language training in Austria where the family had spent 19 months waiting for Canadian visas. Although Martina had a professional degree as a surveyor, she worked as a cook help at a restaurant, Fast Foods, before she started the ESL course in January 1990. Initially, Martina was dependent on her children to perform the public and domestic tasks of settling into a new country. When Martina went looking for a job, she took her eldest daughter with her, even though her daughter would become distressed because nobody wanted to employ her mother. When Martina wanted to help serve customers at Fast Foods, she asked her daughters to tell her what words to use. As Martinas English improved, she took on more of the parental tasks in the home. Many of Martinas diary entries describe the way that she used English to perform a wide variety of tasks in the home and community. It was Martina rather than her husband Petr who did most of the organization in the family, like finding accommodations, organizing telephones, buying appliances, finding schools for the chil20 TESOL QUARTERLY
dren. Martina also helped her husband to perform public tasks in English. When Petr was laid off work, he relied on Martina to help him get unemployment insurance and he asked Martina to help him prepare for his plumbers certificate by translating the preparation book from English to Czech. I wish to argue that Martinas investment in English was largely structured by an identity as primary caregiver in the family. It was important that she learn English so that she could take over the parental tasks of the home from her children. The very reason why Martina and Petr came to Canada was to find a better life for children. Martina was anxious not to jeopardize the childrens future by having them take on more public and domestic tasks than were absolutely necessary. Furthermore, because Martina had the responsibility for dealing with the public world, she was also anxious to understand the Canadian way of lifehow things get done in Canadian society. The poststructuralist view that social identity is nonunitary and contradictory helps to explain how Martina responded to and created opportunities to practice English. To illustrate this point, I will address some of the multiple sites of Martinas identity formation: She was an immigrant, a mother, a language learner, a worker, a wife. As a socially constructed immigrant woman (Ng, 1987; Boyd, 1992), Martina never felt comfortable speaking. Despite the fact that Martina showed remarkable resourcefulness and progress in her language learning, she frequently referred to herself as stupid and inferior because she could not speak English fluently. As she wrote in December 1991:
1. I feel uncomfortable using English in the group of people whose English language is their mother tongue because they speak fluently without any problems and I feel inferior.
Significantly, however, despite feelings of inferiority and shame, despite what could be described as a high affective filter, Martina refused to be silenced. I suggest that the reasons why Martina refused to be silenced were because her social identity as a mother and primary caregiver in the home led her to challenge what she understood to be appropriate rules of use governing interactions between anglophone Canadians and immigrant language learners. The multiple sites of identity formation explain the surprises in Martinas dataoccasions when Martina would speak despite the fact that she was not a legitimate speaker (Bourdieu, 1977, p. 650) in the particular discourse. To mention only two occasions: First, Martina surprised her children (and no doubt her landlord and herself) by entering into a long conversation with her landlord on the phone in which she insisted that her family had not broken their lease agreement. In her diary of March 8, 1991, she wrote:
2. The first time I was very nervous and afraid to talk on the phone. When the phone rang, everybody in my family was busy, and my daughter had to answer it. After ESL course when we moved and our landlords tried to persuade me that we have to pay for whole year, I got upset and I talked with him on the phone over one hour and I didnt think about the tenses rules. I had known that I couldnt give up. My children were very surprised when they heard me.
Second, Martina surprised customers at Fast Foods (who looked at her strangely) and co-workers (who were surprised, but said nothing) by taking the initiative to serve the customers while the other workers were playing a video game in the managers office. Consider the following entry from her diary on March 7, 1991.
3. My experiences with young Canadians were very bad, maybe I didnt have fortune. Usually I worked only with my manager, but when was P.A. day or some holidays for students, the manager stayed in his office and I worked with some students. Very often I worked with two sisters Jennifer (12 years) and Vicky (15 years) and the assistant manager who was at a cash [register]. These two girls loved talking but not with me. Even though I was very busy, they talked with young customers and laughed and sometime looked at me. I didnt know, if they laughed at me or not. When we didnt have any customers, they went to the manager office and tried to help the manager with wheel of fortune on the computer. Later when some customers came in and I called these girls, they went but they made faces. I felt bad and I wanted to avoid this situation. In the evening I asked my daughter what I have to tell the customer. She answered me May I help you then pardon and something else. When I tried first time to talk to two customers alone, they looked at me strangely, but I didnt give up. I gave them everything they wanted and then I went looking for the girls and I told them as usually only cash. They were surprised but they didnt say anything.
I suggest that Martinas perseverance with speaking (I couldnt give up, I didnt give up) and her courage to resist marginalization intersect with her social identity as a mother in two ways. First, as a primary caregiver, she could not rely on her husband to deal with the public world and defend the familys rights against unscrupulous social practices. Martina had to do this herself, regardless of her command of the English tense system, the strange looks she received from her interlocutors, and her feelings of inferiority. Second, Martina drew on her symbolic resources as a mother to reframe the power relations between herself and her co-workers. Thus, instead of conceding to their power as legitimate speakers of Eng1ish, she reframed their relationship as a domestic one in which, as children they had no authority over her, as a parent. Consider the following extract taken from an interview with Martina on March 17, 1991:
22 TESOL QUARTERLY
4. In restaurant was working a lot of children, but the children always thought that I amI dont knowmaybe some broom or something. They always said Go and clean the living room. And I was washing the dishes and they didnt do nothing. They talked to each other and they thought that I had to do everything. And I said No. The girl is only 12 years old. She is younger than my son. I said No, you are doing nothing. You can go and clean the tables or something.
Martinas social identity was a site of struggle. By setting up a counterdiscourse in her workplace and resisting the subject position immigrant woman in favor of the subject position mother, Martina claimed the right to speak. It is precisely this ability to claim the right to speak that I suggest should be an integral part of an expanded notion of communicative competence.
Eva: Social Identity as Changing Over Time

Eva was born in Poland in 1967 and came to Canada as a refugee in 1989 when she was 22 years old. She immigrated because she wanted economical advantage. Eva had finished high school and worked as a bartender before she left Poland. She chose to come to Canada because it is one of the few industrialized countries that encourages immigration. She came alone, with no family or friends, but did know one person in Newtown before she arrived. Before Eva came to Canada, she spent 2 years in Italy where she became fluent in Italian. She knew no English before she arrived in Canada. When Eva arrived in Newtown, she found employment at what she calls The Italian store which is situated in the heart of an established Italian neighborhood in Newtown. Eva herself lived in this neighborhood, as do many recent immigrants to Newtown. Eva was given the job at the Italian store because she was a fluent speaker of Italian. Eva was happy at the Italian store but was concerned because she wanted to learn English and had little opportunity to practice English while working in this store. After she finished the ESL course in June 1990, she began looking for another job in earnest, at a place where she could become a more proficient speaker of English. She found employment at a restaurant in Newtown called Munchies, where she was the only employee who could not speak English fluently. Eva was a fulltime employee whose main job was to clean the store and prepare the food for cooking. The conception of social identity as subject to change helps explain the way Eva over time responded to and created opportunities to practice English in her workplace. The central point I wish to make here is that it was only over time that Evas conception of herself as an immigrantan illegitimate speaker of Englishchanged to a
conception of herself as a multicultural citizen with the power to impose reception. When Eva first started working at Munchies, she did not think it was appropriate for her to approach her co-workers and attempt to engage them in conversation. As she said in an interview on March 7, 1991,
5. When I see that I have to do everything and nobody cares about me becausethen how can I talk to them? I hear they doesnt care about me and I dont feel to go and smile and talk to them.
Note that Eva does not complete a crucial part of her sentence. Nobody cares about me because. The data suggest that nobody acknowledged Eva because she had the subject position immigrant in the workplace: As Eva put it, she was someone who was not fluent in English; she was not Canadian, she was stupid, she had the worst type of work in the store. To speak under such conditions would have constituted what Bourdieu (1977) calls heretical usage (p. 672). Eva accepted the subject position immigrant; she accepted that she was not a legitimate speaker of English and that she could not command reception of her interlocutors. As she herself said, when she first arrived in Canada, she assumed that if people treated her with disrespect, it was because of her own limitations. She conceded to these rules of use in her workplace, rules that Eva herself accepted described as normal. As she said in an interview on January 23, 1991,
6. I think because when I didnt talk to them, and they didnt ask me, maybe they think Im just likebecause I had to do the worst type of work there. Its normal.
As Evas sense of who she was, and how she related to the social world began to change, she started to challenge her subject position in the workplace as an illegitimate speaker of English. An extract from an interview on January 23, 1991, indicates how Eva claimed spaces in conversations with co-workers. Her purpose was to introduce her own history and experiences into the workplace in the hope that her symbolic resources would be validated. This surprised her co-workers.
7. (B refers to Bonny and E refers to Eva.) B: You were saying Eva that you are starting to speak to other people? The other people who work [at Munchies]? E: Ya. Because before B: Is everybody there Canadian? E: Ya. Because there everybody is Canadian and they would speak to each other, not to mebecauseI always was likethey sent me off to do something else. I felt bad. Now its still the same but I have to do something. I try to speak. B: How are you doing that? 24 TESOL QUARTERLY
E: For example, we have a half-hour break. SometimesI try to speak. For example, they talk about Canada, what they like here, the places which they like B: Like to visit? Vacations? E: Ya. Then I started to talk to them about how life is in Europe. Then they started to ask me some questions. But its still hard because I cannot explain to them how things, like B: How do you actually find an opportunity in the conversation to say something. Like, if theyre talking to each other, do you stop them? E: No. B: You wait for a quietThen what do you say? E: No. I dont wait for when they are completely quiet, but when its the moment I can say something about what they are talking about. B: When you started doing that, were they surprised? E: A little bit.
As Eva continued to develop what I have called an identity as a multicultural citizen, she developed with it an awareness of her right to speak. If people treated her with disrespect, it was their problem and not her problem. Thus when, after a years experience in the workplace, a male customer said to her in February 1992,6 Are you putting on this accent so that you can get more tips? Eva had been angry, rather than ashamed; she had spoken out, rather than been silenced. When she said to him, I wish I did not have this accent because then I would not have to listen to such comments, she was claiming the right to speak as a multicultural citizen of Canada. Over time, then, Evas communicative competence developed to include an awareness of how to challenge and transform social practices of marginalization.
THE IMPLICATIONS: CLASSROOM-BASED SOCIAL RESEARCH

Although it is beyond the scope of this article to offer a comprehensive analysis of ways in which my research might inform second language teaching, I take in good faith Savignons (1991) comment that communicative language teaching looks to further language acquisition research to inform its development. I have argued thus far that SLA theorists have struggled to define the nature of language learning because they have drawn artificial distinctions between the individual language learner and larger, frequently inequitable social structures. I have drawn on Martina and Evas data to argue that the individual language learner is not ahistorical and unidimensional but has a complex and sometimes
6
Although the diary study was officially over by February 1992, I continued to maintain contact with the participants.
25
contradictory social identity, changing across time and space. I have drawn on my data to argue that motivation is not a fixed personality trait but must be understood with reference to social relations of power that create the possibilities for language learners to speak. I have suggested that even when learners have a high affective filter, it is their investment in the target language that will lead them to speak. This investment, in turn, must be understood in relation to the multiple, changing, and contradictory identities of language learners. An important implication of my study is that the second language teacher needs to help language learners claim the right to speak outside the classroom. To this end, the lived experiences and social identities of language learners need to be incorporated into the formal second language curriculum. The data indicates, however, that students social identities are complex, multiple, and subject to change. What kind of pedagogy, then, might help learners claim the right to speak? Drawing on insights from my research project in general and the diary study in particular (see Peirce, 1994), as well as a wide range of classroom research (e.g., Auerbach, 1989; Cummins, 1994; Heath, 1983, 1993; Heller & Barker, 1988; Morgan, 1992; Stein & Janks, 1992; Stein & Pierce, in press), I suggest that what I call classroom-based social research might engage the social identities of students in ways that will improve their language learning outside the classroom and help them claim the right to speak. It may help students understand how opportunities to speak are socially structured and how they might create possibilities for social interaction with target language speakers. Furthermore, it may help language teachers gain insight into the way their students progress in language learning intersects with their investments in the target language. I define classroom-based social research (CBSR) as collaborative research that is carried out by language learners in their local communities with the active guidance and support of the language teacher. In many ways, language learners become ethnographers in their local communities. Like the students in Heaths (1983) study, learners will develop their oral and literacy skills by collapsing the boundaries between their classrooms and their communities. Adult immigrants, however, differ from native-born students in that they do not have easy access to the linguistic codes or cultural practices of their local communities. The emphasis on CBSR, therefore, is to focus precisely on these aspects of social life, with a view to enhancing language learning and social interaction. As will be discussed below, a crucial component of CBSR is the use of the written word for reflection and analysis. As Ngo (1994) has convincingly argued from her personal experience of immigration, writing can build bridges not only across geographic space but across historical time:
26 TESOL QUARTERLY
Through my writing I found myself again after a long time of being lost. I learned who I was in the past, who I was then, and who I wanted to be in the future. There I finally found freedom in writing. I flew in the sky with my pencil and notebook.
CBSR might include the following objectives and methodologies.
Objective 1: Investigative Opportunities to Interact with Target Language Speakers

Learners can be encouraged to investigate systematically what opportunities they have to interact with target language speakers, whether in the home, the workplace, or the community. To this end, they might make use of observation charts or logbooks.
Objective 2: Reflect Critically on Engagement with Target Language Speakers

Learners can be encouraged to reflect critically on their engagement with target language speakers. That is, learners might investigate the conditions under which they interact with target language speakers; how and why such interactions take place; and what results follow from such interaction. This might help learners develop insight into the way in which opportunities to speak are socially structured and how social relations of power are implicated in the process of social interaction. As a result, they may learn to transform social practices of marginalization.
Objective 3: Reflect on Observations in Diaries or Journals

Learners can be encouraged to reflect on their observations in diaries or journals. This will create opportunities for learners to write about issues in which they have a particular investment, and in so doing, develop their talents as writers. Specifically, learners could use their diaries to examine critically any communication breakdowns that may have occurred with target language speakers. These diaries could be written in the target language and collected regularly by the teacher. The diaries might give the language teacher access to information about the students opportunities to practice the target language outside the classroom, their investments in the target language, and their changing social identities. The teacher could help students critically reflect on findings from their research and make suggestions for further research, reflection, and action where necessary.
Objective: Pay Attention to and Record Unusual Events

Learners could be encouraged to pay particular attention to those moments when an occurrence, action, or event, surprises them or strikes them as unusual. By recording their surprises in the data collection process, the learners may become conscious of differences between social practices in their native countries and those in the target language community. Given the subject position student researcher rather than language learner or immigrant, learners may be able to critically engage their histories and their experiences from a position of strength rather than a position of weakness. With this enhanced awareness, learners may also be able to use the language teacher as an important resource for further learning.
Objective 5: Compare Data with Fellow Students and Researchers

Students could use the data they have collected as material for their language classrooms, to be compared with the findings of their fellow students and researchers. In comparing their data with other learners, the students will have an investment in the presentations that their fellow students make and a meaningful exchange of information may ensue. Students may begin to see one another as part of a social network in which their symbolic resources can be produced, validated, and exchanged. The teacher may also be able to use this information to structure classroom activities and develop classroom materials that will help learners claim the right to speak outside the classroom. Drawing on Heath (1993), the teacher could make use of drama to help students develop confidence in interacting with target language speakers. Furthermore, the teacher may be able to guide classroom discussion from a description of the findings of the research, to a consideration of what the research might indicate about broader social processes in the society. In this way, the teacher could help students interrogate their relationship to these larger social processes, understand how feelings of inadequacy are frequently socially constructed, and find spaces for the enhancement of human possibility. In sum, second language theorists, teachers, and students cannot take for granted that those who speak regard those who listen as worthy to listen, and that those who listen regard those who speak as worthy to speak.
ACKNOWLEDGMENTS
I would like to acknowledge and thank the participants in my study, who generously allowed me to bring their personal insights into public debate. I would also like 28 TESOL QUARTERLY
to thank Roger Simon, Monica Heller, Jim Cummins, Barbara Burnaby, Sandra Silberstein, Kathleen Troy, and two TESOL Quarterly reviewers for their diverse contributions to my research and analysis. The research on which this article is based was supported by the Social Sciences and Humanities Research Council of Canada. This support is gratefully acknowledged.
THE AUTHOR
Bonny Norton Peirce is a postdoctoral fellow in the Modern Language Centre, Ontario Institute for Studies in Education, Canada. She is interested in the relationship between social theory and language learning, teaching, and testing internationally. Recent research has been accepted for publication in Applied Linguistics and the Harvard Educational Review.
REFERENCES
Auerbach, E. R. (1989). Toward a social-contextual approach to family literacy. Harvard Educational Review, 59, 165181. Auerbach, E., & McGrail, L. (1991). Rosas challenge: Connecting classroom and community contexts. In S. Benesch (Ed.), ESL in America: Myths and possibilities (pp. 96111). Portsmouth, NH: Heinemann. Bourdieu, P. (1977). The economics of linguistic exchanges. Social Science Information, 16, 645668. Bourdieu, P., & Passeron J. (1977). Reproduction in education, society, and culture. Beverley Hills, CA: Sage Publications. Boyd, M. (1992). Immigrant women: Language, socio-economic inequalities, and policy issues. In B. Burnaby & A. Cumming (Eds.), Socio-political aspects of ESL in Canada. (pp. 14ll59). Toronto, Canada: OISE Press. Brown, H. D. (1987). Principles of language learning and teaching. Englewood Cliffs, NJ: Prentice Hall. Cherryholmes, C. (1988). Power and criticism: Poststructuralist investigations in education. New York: Teachers College Press. Cumming, A. & Gill, J. (1991). Motivation or accessibility? Factors permitting Indo-Canadian women to pursue ESL literacy instruction. In B. Burnaby & A. Cumming (Eds.), Socio-political aspects of ESL education in Canada (pp. 24l 252). Toronto, Canada: OISE Press. Cummins, J. (1994). Knowledge, power, and identity in teaching English as a second language. In F. Genesee (Ed.), Educating second language children: The whole child, the whole curriculum, the whole community (pp. 3358). Cambridge: Cambridge University Press. Gardner, R. C. (1985). Social psychology and second language learning. The role of attitudes and motivation. London: Edward Arnold. Gardner, R. C. (1989). Attitudes and motivation. Annual Review of Applied Linguistics, 1988, 9, 135148. Gardner, R. C., & Lambert, W. C. (1972). Attitudes and motivation in second language learning. Rowley, MA: Newbury House. Gardner, R. C., & MacIntyre, P. D. (1992). A students contributions to secondlanguage learning. Part I: Cognitive variables. Language Teaching, 25, 211220. SOCIAL IDENTITY, INVESTMENT, AND LEARNING 29
Gardner, R. C., & MacIntyre, P. D. (1993). A students contributions to secondlanguage learning. Part II: Affective variables. Language Teaching, 26, 111. Giroux, H. A. (1988). Schooling and the struggle for public life: Critical pedagogy in the modern age. Minneapolis: University of Minnesota Press. Goldstein, T. (1991). Immigrants in the multicultural/multilingual workplace: Ways of communicating and experience at work. Unpublished doctoral dissertation, Ontario Institute for Studies in Education/University of Toronto, Canada. Heath, S. B. (1983). Ways with words: Language, life, and work in communities and classrooms. Cambridge: Cambridge University Press. Heath, S. B. (1993). Inner city life through drama: Imagining the language classroom. TESOL Quarterly, 27, 177192. Heller, M. (1987). The role of language in the formation of ethnic identity. In J. Phinney & M. Rotheram (Eds.), Childrens ethnic socialization (pp. 180-200). Newbury Park, CA: Sage. Heller, M., & Barker, G. (1988). Conversational strategies and contexts for talk: Learning activities for Franco-Ontarian minority schools. Anthropology and Education Quarterly, 19, 2046. Hymes, D. (1971). On communicative competence. In C. J. Brumfit & K. Johnson (Eds.), The communicative approach to language teaching (pp. 526). Oxford: Oxford University Press. Klassen, C. (1987). Language and literacy learning: The adult immigrants account. Unpublished masters thesis, Ontario Institute for Studies in Education/University of Toronto, Canada. Krashen, S. (1981). Second language acquisition and second language learning. Oxford: Pergamon. Krashen, S. (1982). Principles and practice in second language acquisition. Oxford: Pergamon. Morgan, B. (1992). Teaching the Gulf War in an ESL classroom. TESOL Journal, 2, 1317. Ng, R. (1987). Immigrant women in Canada: A socially constructed category. Resources for Feminist Research/Documentation sur la recherche feministe, 16, 1315. Ngo, H. (1994, March). From learner to teacher: Language minority teachers speak out. Paper presented at the 28th Annual TESOL Convention, Baltimore, Maryland. Ogbu, J. (1978). Minority education and caste: The American system in cross-cultural perspective. New York: Academic Press. Peirce, B. N. (1989). Toward a pedagogy of possibility in the teaching of English internationally: Peoples English in South Africa. TESOL Quarterly, 23, 40l 420. Peirce, B. N. (1993). Language learning, social identity, and immigrant women. Unpublished doctoral dissertation. Ontario Institute for Studies in Education/University of Toronto, Canada. Peirce, B. N. (1994). Using diaries in second language research and teaching. English Quarterly, 26, 2229. Peirce, B. N., Harper, H., & Burnaby, B. (1993). Workplace ESL at Levi Strauss: Dropouts speak out. TESL Canada Journal, 10, 930. Rockhill, K. (1987). Literacy as threat/desire: Longing to be SOMEBODY. In J. Gaskill & A. McLaren (Eds.), Women and education: A Canadian Perspective (pp. 315331). Calgary, Canada: Detselig Enterprises. Savignon, S. (1991). Communicative language teaching: State of the art. TESOL Quarterly, 25, 261-278. Schumann, J. (1976). Social distance as a factor in second language acquisition. Language Learning, 26, 135143.
30
TESOL QUARTERLY
Schumann, J. (1978). The acculturation model for second-language acquisition. In R. C. Gringas, (Ed.), Second language acquisition and foreign language teaching (pp. 2750). Washington, DC: Center for Applied Linguistics. Scovel, T. (1978) The effect of affect on foreign language learning: A review of the anxiety research. Language Learning, 28, 12942. Simon, R. (1992). Teaching against the grain: Texts for a pedagogy of possibility. New York: Bergin & Garvey. Spolsky, B. (1989). Condition for second language learning. Oxford: Oxford University Press. Stein, P., & Janks, H. (1992). The process syllabus: A case study. Perspectives in Education, 13, 93105 Stein, P., & Peirce, B. N. (in press). Why the Monkeys Passage bombed: Tests, genres, and teaching. Harvard Educational Review. Stern, H. H. (1983). Fundamental concepts of language teaching. Oxford: Oxford University Press. Weedon, C. (1987). Feminist practice and poststructuralist theory. London: Blackwell.
31
POLICY AND PRACTICE IN THE EDUCATION OF CULTURALLY AND LINGUISTICALLY DIVERSE STUDENTS
Views from Language Educators G. Richard Tucker Editor
ix recognized experts in second language education define the agendas and issues critical to education reform for culturally and linguistically diverse learners. The papers are based on remarks delivered at a panel presentation at the annual convention of the American Educational Research Association (AERA) in Atlanta, Georgia in April 1993, Contributors include JoAnn Crandall, Donald Freeman, Else V. Hamayan, Denise McKeon, Merrill Swain, and G. Richard Tucker. 1993, 22 pp., ISBN 0939791-50-1. $9.95 Use order form in this issue.
University of Cambridge
Third Summer Institute in English and Applied Linguistics
Context in Language Learning and Language Understanding

17-28 July 1995 Academic Director: Gillian Brown
Contributors are expected to include:

Henry Widdowson Charles Fillmore Diane Blakemore Eric Kellerman Herbert H. Clark Dick Allwright Sylvia Adamson David Good
For information and application forms: University of Cambridge, Madingley Hall, Madingley, Cambridge CB3 8AQ, UK Fax: + 44 1954 210677 Tel: + 44 1954 210636
Student Reactions to Teacher Response in Multiple-Draft Composition Classrooms*

DANA R. FERRIS
California State University, Sacramento
Research in L1 and L2 student writing has suggested that teacher response to student compositions is most effective when it is given on preliminary rather than final drafts of student essays (Freedman, 1987; Krashen, 1984). One area of research in L1 and L2 composition is the assessment of student reactions to the feedback they receive from their teachers (Cohen & Cavalcanti, 1990; Hedgcock & Lefkowitz, 1994; Leki, 1991; McCurdy, 1992). However, most previous studies of ESL student response to their teachers written comments on their essays have been undertaken in single-draft, rather than multiple-draft, contexts. In this study, 155 students in two levels of a university ESL composition program responded to a survey very similar to the ones utilized by Cohen (1987) and McCurdy (1992) in single-draft settings. The results of the survey indicated that students pay more attention to teacher feedback provided on preliminary drafts (vs. final drafts) of their essays; that they utilize a variety of strategies to respond to their teachers comments; that they appreciate receiving comments of encouragement; and that, overall, they find their teachers feedback useful in helping them to improve their writing. Responses also showed that students had a variety of problems in understanding their teachers comments, suggesting that teachers should be more intentional in explaining their responding behaviors to their students.
s L1 and L2 composition research and pedagogy have evolved over the past several decades, many things have changed. One factor has remained constant, however: the importance of composition teachers roles in providing feedback to their students. Reid (1993) notes that the ESL writing teacher plays several different roles, among
*Results of this study were first presented at the 28th Annual TESOL Convention in Atlanta, GA, April 1993.
33
them coach, judge, facilitator, evaluator, interested reader, and copy editor (p. 217). Whatever a particular teachers orientation(s) toward responding to student writing, it is clear that teachers response is important to both instructor and students. Research investigating various aspects of ESL writing instruction has demonstrated that students expect and value their teachers feedback on their writing (Cohen & Cavalcanti, 1990; Hedgcock & Lefkowitz, 1994; McCurdy, 1992). Further, the amount of time and effort teachers spend in providing written and/or oral feedback to their students suggests that teachers themselves feel that such response is a critical part of their job as writing instructors.
DOES TEACHER FEEDBACK HELP?

Despite the perceived importance of the role of the teacher in responding to student writing, research in both L1 and L2 student writing provides very little evidence that such feedback actually helps the students writing improve (see Leki, 1990, for a review). In particular, Knoblauch and Brannon (1981) contrasted various types of teacher response in L1 writing (e.g., oral vs. written, explicit vs. implicit, praise vs. criticism), concluding that none of these different response modes had much impact on subsequent student writing. Similarly, Hillocks (1986) in investigating L1 teacher response concludes that teacher comment has little impact on student writing (p. 165). In L2 research, many studies of teacher response have arrived at similar conclusions. Such research criticizes the responding behaviors of teachers, on the grounds that
ESL writing teachers misread student texts, are inconsistent in their reactions, make arbitrary corrections, write contradictory comments, provide vague prescriptions, impose abstract rules and standards, respond to texts as fixed and final products, and rarely make content-specific comments or offer specific strategies for revising the text. (Zamel, 1985, p. 86)
However, some scholars have questioned the conclusions drawn in these reviews. For example, Leki (1990) points out that if research has failed to establish that annotations on student papers help them improve their writing, it may well be that the problem is not the annotation but the entire teaching environment (p. 63). In addition to Lekis comments about the important of the overall pedagogical context, several researchers have noted that comments on intermediate drafts which are to be subsequently revised are more useful in facilitating student improvement than feedback on final drafts (Freedman, 1987; Hillocks, 1986; Knoblauch & Brannon, 1981; Krashen, 1984).
34 TESOL QUARTERLY
Berger (1991) reviews both L1 and L2 studies on the influence of teacher, peer, and self-feedback on helping students to revise and improve their writing, noting that several L2 studies (Chaudron, 1983; Zhang, 1985; Zhang & Halpern, 1988) appear to demonstrate that teacher feedback on preliminary drafts did help students to revise effectively. A more recent study (Fathman & Whalley, 1990) demonstrated that students revisions improved in overall quality and in linguistic accuracy when they received comments and/or corrections on both the content and form of their essays.
WHAT DO STUDENTS THINK ABOUT TEACHER FEEDBACK?

A related area of research in teacher response to student writing has examined students preferences about and reactions to their teachers written commentary. Leki (1990) reviewed L1 studies on this topic, noting that L1 students reported not paying much attention to teacher commentary, not understanding it, or feeling some hostility about teachers attempts to take over their ideas and writing. Several different studies have surveyed and/or interviewed ESL and/ or foreign language writing students to obtain their feedback on feedback (Hedgcock & Lefkowitz, 1994). These studies have pursued two general, related lines of inquiry: (a) studies of student preferences regarding teacher feedback (Hedgcock & Lefkowitz, 1994; Leki, 1991; Radecki & Swales, 1988) and (b) studies of student response to feedback they have already received (Cohen, 1987; Cohen & Cavalcanti, 1990; McCurdy, 1992). In the first group of studies, students have been asked about the type(s) of feedback they prefer to receive (both form and substance). Although the earlier two studies (Leki, 1991; Radecki & Swales, 1988) found that the students generally preferred extensive comments on grammar rather than content, the more recent study (Hedgcock & Lefkowitz, 1994) reported a more complex finding: Foreign language students paid more attention to form, whereas ESL students were as interested in teacher feedback on content as they were in sentence-level comments and corrections. The authors suggest that this result may be due to the fact that whereas foreign language students use L2 writing as a form of language practice, ESL students must use their writing skills for all of their academic endeavors (i.e., beyond the language classroom). In the second group of studies, students were asked about their perceptions of what their teachers actually focused on in responding to student essays and to discuss their own subsequent actions: Did they reread their papers when returned? Did they pay attention to their
STUDENT REACTIONS TO TEACHER RESPONSE 35
teachers comments? What strategies did they use to apply their teachers feedback to subsequent writing assignments? Did they have trouble understanding any teacher commentary, and if so, what did they do to resolve these problems? The earliest of these studies (Cohen, 1987) reported rather discouraging findings: Although most of the students claimed to have reread their papers and attended to their teachers comments, a full 20% did not. Further, the students in general reported a limited repertoire of strategies for processing teacher feedback (p. 65). Most students claimed that they merely made a mental note (p. 63) of their teachers feedback. Cohen concludes that his results suggest that the activity of teacher feedback as currently constituted and realized may have a more limited impact on the learners than the teachers would desire (p. 66). Two studies following Cohens (1987) research (Cohen & Cavalcanti, 1990; McCurdy, 1992) reported more positive results: The students in general were happy with the feedback they received, claimed that they paid attention to it, and found it helpful. Although McCurdys study found that the students reported a variety of problems in understanding their teachers feedback, they also utilized a great variety of strategies (e.g., asking the teacher for help, looking up corrections in a grammar book) to resolve difficulties and respond to the teacher feedback. Of all of the previous studies of L2 student reactions to teacher feedback, only one (Hedgcock & Lefkowitz, 1994) was carried out in contexts in which revision and multiple drafting were consistently required. It makes sense that student attention to and preferences regarding teacher feedback would differ in a pedagogical setting in which multiple drafting is required: Because students must rethink and revise previously written essay drafts, they are more likely to pay close attention to their teachers advice on how to do so than in a situation in which they are merely receiving a graded paper with comments and corrections to apply to a completely new essay assignment. Along these lines, Leki ( 1991), who found that students had very strong preferences for extensive error correction on their papers, suggested that a follow-up questionnaire at the end of a semester of writing using a process, multiple-draft approach might have been useful in order to see if students attitudes toward error are changed by an approach which does not emphasize error (p. 210). In order to build on the previous research and extend it to reflect the widespread use of process-oriented ESL composition pedagogy, the present study applies the methodology used by Cohen (1987) and McCurdy (1992) to a multiple-draft setting. The first two research questions, therefore, are the same ones posed by Cohen (1987, p. 59); the third reflects the present context:
36 TESOL QUARTERLY
1. What does teacher feedback tend to cover and in what form is it presented? 2. How much of this feedback do students process, how do they go about doing this, and what forms of feedback might be difficult for them to interpret? 3. What (if any) differences are there in teacher feedback and what is student response to it between preliminary and final drafts?
METHOD
Subjects
The subjects in this study were 155 students taking ESL writing classes at California State University, Sacramento. Nearly all of them were immigrant (not international) students. Because the surveys were taken anonymously, it was not possible to get specific data about their linguistic backgrounds; however, most of the students in the ESL program come from Pacific Rim nations (especially Vietnam and Hong Kong) or Mexico. The majority of the students had taken at least one previous composition class at the university, so they were well acquainted with process-oriented techniques.
Pedagogical Context
The students were enrolled in various sections of two separate courses: English 2A (a prebaccalaureate course that comes before the ESL freshman composition course) or English 2B (a sheltered freshman composition course). English 2A is a credit/no-credit course; students must pass a programwide exit examination to move to English 2B. English 2B is a regular graded course with no exit examination. In both courses, all instructors use a multiple-draft syllabus (at least three drafts per assignment for English 2A and two drafts for English 2B). In addition, it is program policy that teachers focus on content and organization in their feedback on first drafts, saving grammatical and mechanical concerns for final drafts. Further, the grading rubrics that are used for the two classes (with which both teachers and students are familiar) make it clear that content and rhetorical issues are given greater weight than sentence-level concerns in determining the final grade on an essay.
Data Collection
The students were surveyed during the Spring and Fall 1992 semesters using the questionnaire developed by McCurdy (1992) but slightly
adjusted for the multiple-draft context.1 A copy of the questionnaire is provided in the Appendix. The surveys were taken in class, following a standard protocol, during the tenth week of a 15-week semester, so the students had been through at least two multiple-draft essay cycles with their current teacher and were well into a third cycle. The students represented 9 different teachers and 13 separate sections (7 of English 2A and 6 of English 2B). The teachers agreed to administer the survey to their students on the condition that both they and their students remained anonymous. 175 completed surveys were returned (a response rate of about 85%), but 20 of those taken during the Fall semester were excluded from analysis because the students indicated that they had already completed the survey during the previous semester. Thus, 155 surveys were analyzed, of which 89 were completed by English 2A students, and 66 by English 2B students.
Analysis
The quantitative items on the questionnaire (14, 1011) were tallied and summed. In order to examine differences between students responses about preliminary versus final drafts, a Wilcoxon MatchedPairs Signed-Ranks test was run on Questions 1 through 4. This test was also used to compare the responses to Questions 10 and 11.2 For the structured-response items (Questions 59), the responses were collated and grouped in order to examine various trends in the responses.
RESULTS AND DISCUSSION Were There Significant Differences in Student Responses About Preliminary and Final Drafts?
For all items in Questions 14, the Wilcoxon test (measuring differences in student reactions to teachers comments on preliminary and final drafts) showed statistically significant differences. These results are summarized in Tables 1 and 2. Further, the differences in responses to Questions 10 and 11 were also statistically significant (see Table 3 below).
1
McCurdys questionnaire was based on the instruments used in Cohen (1987) and Cohen and Cavalcanti ( 1990). The statistical package SPSS for Windows, Release 6.0 (Norusis, 1993), was used for all tests.
38
TESOL QUARTERLY
TABLE 1
Questions 1 and 2: Differences Between Preliminary and Final Drafts (N = 155)
Note. All numbers may not add up to 155 due to missing values. Percentages may not add up to 100% due to rounding. * p < .005 (Wilcoxon Matched-Pairs Signed-Ranks test)
Did Students Reread Their Papers and Attend to Their Teachers Comments?
For Questions 1 and 2, in which students were asked to respond to questions about how much of their papers and the teachers comments they read and paid attention to, there were significant differences between the preliminary and final drafts. As Table 1 shows, if the all and most categories are combined, students were more likely to reread of their essays and pay attention to their teachers comments on the earlier draft(s) than on the final draft. In addition, the proportion of students who claimed to have reread all or most of their papers and to have read and attended to all or most of their teachers comments was greateron both drafts (preliminary and final)in this study than in two prior studies which asked students only about their responses to single drafts (Cohen, 1987; McCurdy, 1992). Two observations can be made about these results. First, in this multiple-draft setting, students appeared to take their own work and their teachers feedback quite seriously. In particular, student writers paid great attention to their preliminary drafts and to their teachers comments on them (as seen by the high number of all or most responses). Given that they were required to revise these drafts at least once (and more in some cases), it is not surprising that the vast majority of students reported paying close attention to their own drafts and to their teachers responses to their papers. Second, even on final drafts, which the students typically did not rewrite, the students in this study reread more of their papers and paid more attention to their teachers
comments than did the students in previous studies undertaken in single-draft settings. A possible explanation for this is that, having expended the effort to write several drafts and to revise according to their teachers suggestions on their preliminary drafts, students had more interest in knowing how their efforts had been received by their teachers.
What Types of Feedback Did Students Report Receiving, and How Much Attention Did They Pay to Each Type?
Questions 3 and 4 assessed the students perceptions of the nature and amount of their teachers feedback and how much attention the students paid to specific types of feedback (on organization, content, grammar, vocabulary, and mechanics). The responses to these two questions are summarized in Table 2. On Question 3, for all categories, when the a lot and some categories were combined, most students felt they received more comments on their first/second drafts and significantly fewer comments of all types on their final drafts. This of course makes sense in the multiple-draft context: Because the students were required to revise their earlier drafts, teachers would naturally give them some direction for doing so, writing fewer comments of any type on final drafts which would not be rewritten. As Table 2 illustrates, the students felt they received the most comments on grammar, followed (in this order) by organization, content, mechanics (defined for the students as spelling, punctuation and capitalization), and vocabulary. This order held for both preliminary and final drafts and is identical to the one reported by McCurdy (1992). The top ranking for grammar is consistent with the previous studies cited above on student response to teacher feedback. However, in the most recent research (Hedgcock & Lefkowitz, 1994; McCurdy, 1992), students reported that content and organization (in addition to grammar) were very important to their instructors. This finding differs from the results of earlier studies (Cohen, 1987; Leki, 1991; Radecki & Swales, 1988), in which students reported that instructors focused mainly on grammar and/or that they themselves preferred to receive feedback on grammar, rather than content. The trends seen in these two recent studies, together with the results of the present study, suggest that although ESL writing instructors still pay a great deal of attention to grammar, they may be focusing more on students ideas and organization in their written feedback than they did in the past, and that they are communicating these shifting priorities to their students. One issue arising from the responses to Question 3 is why the students felt they received so much feedback on grammar and mechanics on first drafts, on which the teachers ostensibly were only responding
40 TESOL QUARTERLY
TABLE 2
Questions 3 and 4: Differences Between Preliminary and First Drafts (N = 155)
Frequency of Response Survey Question Organization Question 3* Preliminary Drafts Final Drafts Question 4* Preliminary Drafts Final Drafts Content/Ideas Question 3* Preliminary Drafts Final Drafts Question 4* Preliminary Drafts Final Drafts Grammar Question 3* Preliminary Drafts Final Drafts Question 4* Preliminary Drafts Final Drafts Vocabulary Question 3* Preliminary Drafts Final Drafts Question 4* Preliminary Drafts Final Drafts Mechanics Question 3* Preliminary Drafts Final Drafts Question 4* Preliminary Drafts Final Drafts Note. All numbers may not add up to 155 due to missing values. Percentages may not add up to 100% due to rounding. * p <.005 (Wilcoxon Matched-Pairs Signed-Ranks test) A lot No. (%) Some No. (%) A little No. (%) None No. (%) N/A No. (%)
STUDENT REACTIONS TO TEACHER RESPONSE
41
to their content and organization. (In fact, one composition teacher was so upset by her students responses to this question that she refused to submit her set of surveys after having administered them.) There are at least two explanations for this result. First, some of the teachers may have made comments/corrections on grammar points on preliminary drafts, despite the program policy which discouraged them from doing so (but which was not enforced in any official way). Second, the students may have been confused as to what their teachers actually did on the various drafts because they were relying on their memories to complete the survey. Although this may simply reflect an inevitable limitation of this type of research, it may also indicate that teachers need to be more intentional about explaining their responding practices (e.g., content on first drafts, grammar on later drafts) to their students, a point that is discussed elsewhere in this article. After reflecting on their teachers perceived response priorities, the students were asked in Question 4 how much attention they paid to the different types of feedback they received on each draft. In general, the students said they paid more attention to comments of all types on preliminary drafts than on final drafts. Although they still reported paying the most attention to teacher comments about their grammar (67% said they paid a lot of attention to grammar comments on preliminary drafts), attention to content-oriented feedback was not far behind (63%). Thus, the results here indicate that the students perceived the importance of focusing on content, particularly in preliminary drafts. On final drafts, students attention to comments on vocabulary and mechanics was quite similar to the attention given to content and organization. A possible explanation for this is that teacher commentary on students ideas tends to be very text-specific, relating primarily to the content of the paper currently under consideration. Students therefore may not see much relevance in teacher commentary about their ideas on final drafts, reasoning that they will be moving on to a new topic anyway. On the other hand, feedback on vocabulary and mechanics may be perceived by students as transcending the specifications of a particular assignmentinformation that they could apply to any future writing project.
How Did Students Perceive Themselves as Learners and Writers?

The final two questions of the survey (10 and 11) asked the students to assess themselves as learners and as writers. The majority (about 70%) felt that they were excellent or good learners, but only good or fair
42 TESOL QUARTERLY
TABLE 3
Questions 10 and 11: Students Perceptions of Their Own Abilities* (N = 155)
Self Rating Survey Question Question 10 Learning Question 11 Writing Excellent No. (%) Good No. (%) 93 (60) 67 (43) Fair No. (%) Poor No. (%) 4 (3) 9 (6)
18 (12) 2 (1)
32 (21) 68 (44)
Note. All numbers may not add up to 155 due to missing values. Percentages may not add up to 100% due to rounding. *Difference Between Perceived Learning and Writing Abilities: p < .005 (Wilcoxon MatchedPairs Signed-Ranks test)
writers (Table 3). Although it is difficult to determine exactly why the students felt this way, it may simply reflect the fact that these ESL students, the great majority of whom are computer science, engineering, or business majors, feel more confident of their general academic abilities than of their writing proficiency. Many ESL students at this particular university feel great anxiety about their writing, especially because all students must pass a timed essay examination to graduate. However, these responses may also be indicative of the insecurity most people feel about writing: One might find similar patterns of response to Questions 10 and 11 among the native-English-speaking student population.
What Did Students Report Doing in Response to Teacher Feedback?

A second set of questions on the survey (59) called for more qualitative responses from the students. A summary of the comments given in response to these questions is provided in Figure 1. Questions 5 and 7 asked students to describe their strategies for addressing their teachers feedback. On early drafts, many of the students reported going to an outside source (instructor, tutor, friends, grammar book, dictionary) for help in understanding or responding to their teachers comments. On final drafts, fewer of the students said they consulted an outside source, whereas about 50% of the students said that they try to make corrections themselves, think about the teachers comments, or do nothing, rather than taking any further steps to respond to the feedback.
The results of this study are encouraging in this regard because Cohen (1987) noted that L2 student writers in his study reported a limited range of strategies for responding to and dealing with problems with their teachers feedback. The finding that students utilize a variety of resources to deal with teacher commentary may be attributed to the multiple-draft design of the writing classes, requiring students to grapple with their teachers feedback in order to revise successfully and to the collaborative nature of the writing classes, making students more willing to ask friends and tutors for help. These external factors may have influenced the students strategies for responding to their teachers feedback because they were more convinced of the necessity of doing so and more familiar with resources outside the writing class available to help them.
Did Students Report Difficulties in Understanding Teachers Feedback?

On Question 6, the students were asked if they ever had any problems with their teachers feedback, and if so, to give specific examples of such problems. Nearly 50% said that they never had any problems understanding their teachers comments; another 11% said that they sometimes did but couldnt think of any specific examples. Although these results appear to indicate that more than 50% of the students did have problems with their teachers comments, it should be noted that the survey did not require students to say whether they always had trouble understanding their teachers, but rather if they ever did. In other words, a student who had only one problem on one draft over the whole semester would be counted in that 50%. Seen in this light, it seems encouraging that nearly half of the students reported never having trouble understanding the teachers comments. Of the students who did report specific problems, 13 students (9%) complained about not being able to read their teachers handwriting. In the remaining responses, many of the specific examples mentioned by students related to grammar corrections of various typesproblems with both specific grammar terms (fragment, verb tense) and symbols used to indicate a grammatical error (abbreviations, arrows, and circles). In addition, several students mentioned that their teachers questions about content confused them (too general or too specific), and some said that although they understood the teachers comments, they didnt always agree with them. In particular, one student remarked, My writing style is not American and the teacher seems to like American style. Its hard to change my style.
44
TESOL QUARTERLY
FIGURE 1
Number of Responses to Structured-Response Questions (5-9)
5. Describe what you do after you read your instructors comments and corrections. a. First Drafts: b. Final Drafts: Ask teacher for help (13) Nothing (7) Make corrections myself (61) Ask teacher for help (10) Ask friends for help (11 ) Make corrections myself (29) Check grammar book (34) Ask friends for help (6) Think about/remember mistakes (6) Check grammar book (28) Ask tutor for help (34) Think about/remember mistakes (44) Check dictionary (18) Ask tutor for help (22) Check dictionary (13) 6. Are there ever any comments or corrections that you do not understand? If so, can you give any examples? No (79) Yes; no examples given (17) Cant read teachers handwriting (13) Understand but sometimes disagree (4) Grammar terms, abbreviations, and symbols (16) Word choice (2) Comments about ideas or organization (3) Questions that are too general/too specific (3) 7. What do you do about those comments or corrections that you do not understand? Nothing (12) Ask my instructor to explain them (78) Look corrections up in a grammar book or dictionary (11) Ask a tutor for help (13) Ask friends/classmates/family for help (17) Try to fix it myself (22) 8. Are any of your instructors comments positive? If so, can you give an example? Yes (no example given) (35) No, rarely, it depends (8) They are all positive because they are all helpful (5) Comments on content/ideas (53) Comments on organization/rhetorical structure (26) Comments on grammar/vocabulary/writing (30) Negative comments (22) 9. Do you feel that your instructors comments and corrections help you to improve your composition writing skills? Why or why not? Yes (145) I know what to avoid/improve next time (41) I know where my mistakes are (36) Helps me to improve my writing skills (23) Helps me to think more clearly/make more sense/find more ideas (11) Good comments build my confidence (3) Helps me to get better grades/pass essay exams (5) I respect my instructors opinion (3) Challenges me to try new things (2) No (8) I need more help to correct my errors (5) Some help; some dont (2) Instructors comments are too negative and discouraging (3) *Note. Because subjects could write more than one comment on any question and some wrote nothing on some questions, numbers do not add up to 155 and percentages are not calculated.
45
Did Students Say They Received Positive Feedback?

The most interesting set of comments was in response to Question 8: Does your teacher ever give you any positive comments on your papers? Although 35 students said that they did receive positive comments, but didnt provide any examples, many other students seemed to have very specific recollections of positive comments they had received. Many of the students were able to quote their teachers affirming comments word for word (e.g., Your essay is poetic and has wonderful images). Of the specific positive comments reported, 53 dealt with the content or ideas of the essay, 26 with the organization or rhetorical structure, and 30 with the language (grammar, vocabulary, or writing). These student recollections are especially interesting given the quantitative results already discussed: Despite the students perceptions that teachers feedback dealt more with grammar than with anything else and that they themselves paid more attention to grammar corrections, it was the teachers positive comments on their ideas and organization that students remembered most specifically. However, there were some 22 specific comments reported by the students as positive in the survey that appeared to be critical or even negative (e.g., You need to work harder, Some parts of your essay dont make sense, Off topic.). A possible explanation for this may lie in the apparent interpretation of the word positive by some of the students as helpful; five students specifically noted that all of their teachers comments were positive because all comments helped them to improve their writing. In contrast, a few students reported that they never or rarely received positive comments from their teachers. Several wrote rather bitterly that their teachers comments were all negative and that this fact depressed them and decreased their motivation and self-esteem.
Did Students Feel That Their Teachers Feedback Was Helpful?

The above observations were reinforced in the responses to Question 9: Do you think that your teachers feedback has helped you improve your writing? The response to this question was overwhelmingly affirmative: 145 (93.5%) students felt that their teachers feedback had indeed helped them improve as writers because it helped them know what to improve or avoid in the future, find their mistakes, and clarify their ideas. Overall, the students seemed to respect their teachers opinions and appreciate their efforts and attention. Eight students, however, responded negatively to this question, saying that they wished they had received more help/feedback than they actually got, that the
46 TESOL QUARTERLY
teachers comments were too negative, and that they were not specific enough. Nonetheless, in general the students assessments of the value of their teachers feedback were overwhelmingly positive.
CONCLUSIONS Summary of Research Findings

As discussed above, previous L1 studies examining students reactions to their teachers comments have yielded discouraging results: Students disregard, misunderstand, and sometimes disagree with their teachers feedback. However, like earlier studies of L2 students reactions to their teachers feedback (Cohen & Cavalcanti, 1990; Hedgcock & Lefkowitz, 1994; McCurdy, 1992), this study showed that ESL writing students in general take their teachers feedback quite seriously and pay a lot of attention to it. In assessing the effects of a multipledraft setting on student reactions to teacher feedback, this study yielded the following clear results: 1. Students reported rereading their papers more often and paying more attention to teacher feedback on earlier drafts than on final drafts (Tables 1 and 2, Questions 1, 2, and 4). 2. Although students reported receiving more and paying the most attention to comments on grammar than any other aspects of their papers, they also indicated that they received many comments on the content and organization of their essays and that they took such feedback very seriously (Table 2, Questions 3 and 4). 3. Students reported seeking help from outside sources (instructor, tutor, other students, grammar book, or dictionary) to respond to their teachers suggestions and to clarify points of confusion, particularly those on preliminary drafts (Figure 1, Questions 5 and 7). 4. Although almost 50% of the students reported having no problems understanding their teachers comments, other students noted specific problems with the teachers feedback on grammar (both terminology and symbols used), with their teachers use of questions to respond to content, and with reading the teachers handwriting (Figure 1, Question 6). 5. Many of the students reported having received positive comments from their teachers and had vivid memories of their teachers encouraging remarks. Some, however, gave specific examples of negative or critical remarks as being positive, whereas a few wrote that they never received any positive comments and that they found this discouraging (Figure 1, Question 8).
6. An overwhelming majority of the students (93.5%) felt that their teachers feedback had helped them to improve their writing (Figure 1, Question 9). If teacher feedback is to be successful in helping students revise and improve their writing, a necessary prerequisite is that students at least read and take seriously those comments. Though this study does not investigate a causal link between teacher feedback and student improvement, it does suggest that the students pay more attention to the feedback on earlier drafts. The results, therefore, can be argued to support the suggestions of L1 and L2 researchers (e.g., Hillocks, 1986; Leki, 1990, 1991) that teacher feedback on preliminary drafts of student work may be more effective than responses to final drafts. However, these findings should not be interpreted as indicating that students find teacher commentary on final drafts unnecessary because a convincing majority of them also said that they reread all or most of their final papers and read all or most of their teachers comments on their final drafts.
Implications for Pedagogy

The results also suggest the following pedagogical applications for ESL writing teachers.
The Role of Grammar Feedback

Students reported receiving and paying the most attention to feedback on grammar, content, and organization, in that order. In their qualitative responses, they also said that they felt that their teachers comments helped them avoid future mistakes, improve their grammar, and clarify their ideas. These responses should inform teachers in two ways: (a) Many scholars over the years (e.g., Krashen, 1984; Zamel, 1985) have criticized ESL writing teachers for excessive attention to students grammar problems. The results of this study suggest that students both attend to and appreciate their teachers pointing out their grammar problems, a finding which echoes that of earlier studies of L2 students opinions about teacher feedback (e.g., Leki, 1991; Radecki & Swales, 1988). (b) At the other extreme, some teachers may feel that their students do not value anything but grammar feedback, an intuition supported by the findings of Cohen (1987), Leki (1991) and Radecki and Swales (1988). The students responses in this survey, however, show that they also value teachers suggestions about their ideas and organization, a result consistent with the findings of Hedgcock and Lefkowitz (1994) and McCurdy (1992).
48 TESOL QUARTERLY
The Need to Clarify Responding Behaviors

In reviewing L1 research, Cohen (1987) notes that even in a course with an enlightened, process-oriented teacher, the students may still misinterpret the teachers comments (p. 58). The specific problems reported by the students in this study in understanding their teachers comments and corrections suggest that teachers should be quite intentional about explaining their responding behaviors (Zamel, 1985) to their students. Teachers can do this by taking time early in the term to explain their overall philosophy of responding (as well as specific strategies and/or symbols or terminology used) to the students. They can reinforce this by discussing general trends in their responses with the class and by allowing class time for students to read, ask questions about, and respond in their journals to teacher feedback each time a marked draft is returned to the students. Many ESL students, absent this encouragement, might not ask questions, fearing that it might show disrespect by appearing to criticize their teacher for being unclear.
Positive Comments Versus Constructive Criticism

The students responses to Questions 8 and 9 indicated how valuable they found positive comments, remembering many specific examples and expressing some bitterness when they felt they had not received any praise. At the same time, about 15% seemed to find all comments positiveeven ones that teachers might consider critical or negative. This suggests that teachers should not abandon constructive criticism but should place it side-by-side with comments of encouragement.
Directions for Future Research

This study measures student reaction to teacher feedback in multiple-draft contexts, as opposed to single-draft situations. But in some classes (particularly those utilizing portfolio assessment), students are allowed unlimited opportunities to rewrite their essays for higher grades. In this context, students would most likely pay even greater attention to teacher comments on final (i.e., graded) drafts because they are given the opportunity to continue working on them. With the increase in the use of portfolio grading in ESL writing classes, future research should investigate the value to students and effectiveness of teacher response in a pedagogical context in which papers are never done until the student decides they are or until the end of the course (whichever comes first). In addition, this study did not investigate a connection between student perceptions of their teachers commentary and the actual reSTUDENT REACTIONS TO TEACHER RESPONSE 49
spending behaviors of the teachers. Though Cohen and Cavalcanti (1990) did examine this link by surveying a small sample in a singledraft context, this research should be replicated on a larger scale and in a multiplee-draft setting. An additional extension of this research, piloted by Cohen (1987) and suggested by Hedgcock and Lefkowitz (1994), would be to relate the preferences and strategies of students with regard to teacher feedback to their overall writing achievement. Despite its limitations, this study, together with the previous research on ESL students reactions to and perceptions of their teachers feedback, should provide some encouragement to todays composition slaves (Hairston, 1986, p. 117): Students do attend to, grapple with, and appreciate the efforts their teachers make in responding to their writing. Most importantly, this study, together with two other recent studies on student response to teacher feedback (Hedgcock & Lefkowitz, 1994; McCurdy, 1992) indicate that the priorities of processoriented writing instructionmultiple-drafting, emphasis on content, and willingness to utilize a variety of strategies (including collaboration with others) to solve problems and respond effectively to teacher feedbackare being understood and accepted to some degree by the ESL composition students. In other words, the students survey responses may have been a direct reflection of the priorities they thought their instructors were already observing (Hedgcock & Lefkowitz, 1994, p. 155)and which the students themselves were internalizing.
ACKNOWLEDGMENTS
This research was supported in part by funding from the Faculty Professional Development Mini-Grant Program at California State University, Sacramento (CSUS). I would like to thank Sandra McKay and two anonymous TESOL Quarterly reviewers for their specific and incisive comments on earlier versions of this article. However, any errors or omissions are my sole responsibility. I would also like to thank the teachers and students in the English Department at CSUS for their participation in the survey. Finally, I am indebted to Pamela McCurdy for her cooperation and assistance as I replicated her study.
AUTHOR
Dana Ferris is Assistant Professor of English at California State University, Sacramento, where she is Coordinator of the MA TESOL program. Her interests include the description and teaching of ESOL writing, general literacy issues, and teacher education.
50
TESOL QUARTERLY
REFERENCES
Berger, V. (1991). The effects of peer and self feedback. CATESOL Journal, 3, 2135. Chaudron, C. (1983, March). Evaluating writing: Effects of feedback on revision. Revised version of a paper presented at the 17th Annual TESOL Convention, Toronto, Canada. (ERIC Document Reproduction Service No. ED 227 706) Cohen, A. (1987). Student processing of feedback on their compositions. In A. L. Wenden & J. Rubin (Eds.), Learner strategies in language learning (pp. 5769). Englewood Cliffs, NJ: Prentice Hall. Cohen, A., & Cavalcanti, M. (1990). Feedback on compositions: Teacher and student verbal reports. In. B. Kroll (Ed.), Second language writing: Research insights for the classroom (pp. 155177). New York: Cambridge University Press. Fathman, A., & Whalley, E. (1990). Teacher response to student writing: Focus on form versus content. In B. Kroll (Ed.), Second language writing: Research insights for the classroom (pp. 178190). New York: Cambridge University Press. Freedman, S. (1987). Response to student writing. Urbana, IL: National Council of Teachers of English. Hairston, M. (1986). On not being a composition slave. In C. W. Bridges (Ed.), Training the new teacher of college composition (pp. 11724). Urbana, IL: National Council of Teachers of English. Hedgcock, J., & Lefkowitz, N. (1994). Feedback on feedback: Assessing learner receptivity to teacher response in L2 composing. Journal of Second Language Writing, 3, 141163. Hillocks, G., Jr. (1986). Research on written composition: New directions for teaching. Urbana, IL: ERIC Clearinghouse on Reading and Communication Skills and the National Conference on Research in English. Knoblauch, C., & Brannon, L. (1981). Teacher commentary on student writing: The state of the art. Freshman English News, 10, 14. Krashen, S. (1984). Writing: Research, theory, and application. Oxford: Pergamon Press. Leki, I. (1990). Coaching from the margins: Issues in written response. In B. Kroll (Ed.), Second language writing: Research insights for the classroom (pp. 57-68). New York: Cambridge University Press. Leki, I. (1991). The preferences of ESL students for error correction in collegelevel writing classes. Foreign Language Annals, 24, 203218. McCurdy, P. (1992, March). What students do with composition feedback. Paper presented at the 27th Annual TESOL Convention, Vancouver, B.C. Norusis, M. (1993). SPSS for windows base system users guide, release 6.0. Chicago: SPSS Inc. Radecki, P., & Swales, J. (1988). ESL student reaction to written comments on their written work. System, 16, 355365. Reid, J. (1993). Teaching ESL writing. Englewood Cliffs, NJ: Regents Prentice Hall. Zamel, V. (1985). Responding to student writing. TESOL Quarterly, 19, 79102. Zhang, S. (1985). The differential effects of source of corrective feedback on ESL writing proficiency. Occasional Paper No. 9. Honolulu, HI: Department of English as a Second Language, University of Hawaii at Manoa. Zhang, S., & Halpern, P. (1988, January). The effects of corrective feedback on the discourse quality and linguistic accuracy of ESL compositions. Paper presented at the meeting of the Hawaii Educational Research Association, Honolulu, HI.
51
APPENDIX
Composition Survey
1. How much of each composition do you read over again when your instructor returns it to you? 1st/2nd Drafts None of it____ Some of it____ All of it____ Most of it____ Final Drafts {the one that receives a grade/score) Some of it____ None of it____ Most of it____ All of it____ 2. How many of the instructors comments and corrections do you think about carefully? 1st/2nd Drafts None of them____ Some of them____ Most of them____ All of them____ Final Drafts None of them____ Some of them____ Most of them____ All of them____ 3. How many of the comments and corrections involve: Some A little A lot 1st/2nd Drafts Organization Content/Ideas Grammar Vocabulary Mechanics (e.g., punctuation, spelling) A little Some A lot Final Drafts Organization Content/Ideas Grammar Vocabulary Mechanics (e.g., punctuation, spelling) None
None
4. If you pay attention to what your instructor wrote, how much attention do you pay to the comments and corrections involving: A little Not Applicable None Some A lot 1st/2nd Drafts Organization Content/Ideas Vocabulary Mechanics (e.g., punctuation, spelling) Not Applicable A little None A lot Some Final Drafts Organization Content/Ideas Grammar Vocabulary Mechanics (e.g., punctuation, spelling) 5. Describe what you do after you read your instructors comments and corrections (e.g., Do you look up the corrections in a grammar book? See a tutor? Rewrite your paper?) 1st/2nd Drafts Final Drafts 6. Are there ever any comments or corrections that you do not understand? If so, can you give any examples? 7. What do you do about those comments or corrections that you do not understand? 52 TESOL QUARTERLY
8. Are any of your instructors comments positive? If so, can you give an example? 9. Do you feel that your instructors comments and corrections help you to improve your composition writing skills? Why or why not? 10. How would you rate yourself as a learner? Excellent____ Good____ Fair____ Poor____ Poor____
11. How would you rate your skills in writing compositions? Excellent______ Good_____ Fair____
53
On the Teachability of Communication Strategies

ZOLTN DRNYEI
Etvs University, Budapest
Because a significant proportion of real-life L2 communication is problematic, L2 learners might benefit from instruction on how to cope with performance problems. Such instruction could include the specific teaching of communication strategies, which involve various verbal and nonverbal means of dealing with difficulties and breakdowns that occur in everyday communication. Opinions on the teachability of such strategies, however, vary widely, and several researchers have questioned the validity of strategy training. This article first describes what communication strategies are and provides an overview of the teachability issue, discussing the arguments for and against strategy instruction, and suggests three possible reasons for the existing controversy. After this the results of a study aimed at obtaining empirical data on the educational potential of strategy training are presented. The findings point to the possibility of developing the quality and quantity of learners use of at least some communication strategies through focused instruction.
n the 1970s, four studies prepared the ground for the study of communication strategies (CSs), a new area of research within applied linguistics: Selinkers (1972) classic article on interlanguage introduced the notion of strategies of L2 communication. Vradi (1973, but published in 1980) and Tarone (1977; also Tarone, Cohen, & Dumas, 1976) elaborated on Selinkers notion by providing a systematic analysis of CSs, introducing many of the categories and terms used in subsequent CS research. Savignon (1972) reported on a pioneering language teaching experiment involving a communicative approach, which, for the first time, included student training in CSs (or, as she termed them, coping strategies). Since these early studies, much research has been done to identify and classify CSs (for reviews, see Bialystok, 1990; Cook, 1993; Poulisse, 1987); however, far less attention has been paid to the question of whether these strategies could be integrated
55
into second or foreign language teaching programs. This article addresses this issue. I will first describe what communication strategies are and what role they play in communicative competence, then I will give an overview of the controversy that exists in the literature over their teachability. Finally, the results of a quasiexperimental study will be presented, involving a strategy training classroom project conducted with Hungarian EFL learners to obtain data on the effectiveness of such instruction. The results include comparisons of the learners strategy use and speech rate before and after the training in both the treatment and the control groups, as well as measures of attitudes toward the training program.
STRATEGIC COMPETENCE AND COMMUNICATION STRATEGIES

Some people can communicate effectively in an L2 with only 100 words. How do they do it? They use their hands, they imitate the sound or movement of things, they mix languages, they create new words, they describe or circumlocute something they dont know the word forin short, they use communication strategies. Because they lack basic grammar and vocabulary in the target language, their communicative success relies entirely on their ability to communicate within restrictions (Savignon, 1983, p. 43) by using strategies, that is, on their strategic competence. The importance of strategic competence in communication has been widely recognized since Canale and Swain (1980) included it as a major component in their well-known construct of communicative competence, defining it as verbal and nonverbal strategies that may be called into action to compensate for breakdowns in communication due to performance variables or to insufficient competence (p. 30). Complete agreement has not been reached on the definition of CSs, but one working definition many researchers accept is that CSs are a systematic technique employed by a speaker to express his [or her] meaning when faced with some difficulty (Corder, 1981, p. 103). This definition, in accordance with Canale and Swains (1980) and Frch and Kaspers (1983a) conceptualizations, posits problem orientedness and systematicness/consciousness as central features of CSs. Other researchers, however, have conceived CSs in a broader sense by also including attempts to enhance the effectiveness of communication (Canale, 1983, p. 11). It has been generally accepted that CSs are not unique to L2 speakers because communication problems occur and are tackled in L1 commu56 TESOL QUARTERLY
nication as well (see Bongaerts & Poulisse, 1989). There is, however, disagreement concerning the range of these strategies, in particular whether to include interactive strategies that are used when miscommunications (Gass & Varonis, 1991) occurfor example, repair mechanisms and the negotiation of meaningor whether the term communication strategies should be restricted to devices speakers use when they have difficulties in verbalizing a mental plan for lack of linguistic resources (see Cook, 1993; Frch & Kasper, 1984; Vradi, 1992; Yule & Tarone, 1991). This article focuses on the latter category only. It follows that the taxonomies offered by various researchers vary somewhat (for a review, see Bialystok, 1990; Poulisse, 1987) but as Bialystok (1990) remarks, the variety of taxonomies proposed in the literature differ primarily in terminology and overall categorizing principle rather than in the substance of the specific strategies. If we ignore, then, differences in the structure of the taxonomies by abolishing the various overall categories, then a core group of specific strategies that appear consistently across the taxonomies clearly emerges (p. 61). In Figure 1 I have collected a list and descriptions of the CSs I consider most common and important in this core group, based on Vradi (1973), Tarone (1977), Frch and Kasper (1983a), and Bialystok (1990). In the latter half of the 1980s, researchers at Nijmegen University (Netherlands) criticized the existing topologies of CSs as being product oriented, focusing on the surface structures of underlying psychological processes and thus resulting in a proliferation of different strategies of ambiguous validity (Kellerman, 1991; Poulisse, 1987; see also Cook, 1993). The alternative they proposed instead, a process-oriented classification of Css, is presented in Figure 2. Following the basic principles for classifying Css established by Vradi (1973), Tarone (1977), and Frch and Kasper (1983a), the first two strategies in Figure 1 are usually referred to as avoidance or reduction strategies as they involve either an alteration, a reduction, or complete abandonment of the intended message. Strategies 311 are normally termed achievement or compensatory strategies as they offer alternative plans for the speakers to carry out their original communicative goal by manipulating available language, thus compensating somehow for their linguistic deficiencies. The strategies suggested by the Nijmegen group (see Figure 2) also fall under this category. Strategy 12 is an example of stalling or time-gaining strategies. These strategies are functionally different from the strategies mentioned above because they are not actually used to compensate for any linguistic deficiencies but rather to gain time and to keep the communication channel open at times of difficulty. It must be pointed out that commuCOMMUNICATION STRATEGIES 57
FIGURE 1
CSs Following Traditional Conceptualizations
Avoidance or Reduction Strategies 1. Message abandonmentleaving a message unfinished because of language difficulties. 2. Topic avoidanceavoiding topic areas or concepts which pose language difficulties. Achievement or Compensatory Strategies 3. Circumlocutiondescribing or exemplifying the target object or action (e.g., the thing you open bottles with for corkscrew). 4. Approximationusing an alternative term which expresses the meaning of the target lexical item as closely as possible (e.g., ship for sail boat). 5. Use of all-purpose wordsextending a general, empty lexical item to contexts where specific words are lacking (e.g., the overuse of thing, stuff, make, do, as well as using words like thingie, what-do-you-call-it). 6. Word-coinagecreating a nonexisting L2 word based on a supposed rule (e.g., vegetarianist for vegetarian). 7. Use of nonlinguistic meansmime, gesture, facial expression, or sound imitation. 8. Literal translationtranslating literally a lexical item, an idiom, a compound word or structure from L1 to L2. 9. Foreignizingusing a L1 word by adjusting it to L2 phonologically (i.e., with a L2 pronunciation) and/or morphologically (e.g., adding to it a L2 suffix). 10. Code switchingusing a L1 word with L1 pronunciation or a L3 word with L3 pronunciation in L2. 11. Appeal for helpturning to the conversation partner for help either directly (e.g., What do you call . . . ?) or indirectly (e.g., rising intonation, pause, eye contact, puzzled expression). Stalling or Time-gaining Strategies 12. Use of fillers/hesitation devicesusing filling words or gambits to fill pauses and to gain time to think (e.g., well, now let me see, as a matter of fact).
FIGURE 2
CSs as Conceptualized by the Nijmegen University Group
1. Conceptual strategies manipulating the target concept to make it expressible through available linguistic resources. (a) Analytic strategies specifying characteristic features of the concept (e.g., circumlocution). (b) Holistic strategies using a different concept which shares characteristics with the target item (e.g., approximation). 2. Linguistic/code strategies manipulating the speakers linguistic knowledge. (a) Morphological creativity creating a new word by applying L2 morphological rules to a L.2 word (e.g., grammatical word coinage). (b) Transfer from another language.
58
TESOL QUARTERLY
nication maintenance strategies of this type have not been included in the most well-known taxonomies put forward by Tarone, Frch and Kasper, Bialystok or the Nijmegen group. Several other researchers, however, have highlighted the significance of using fillers and hesitation devices as a conscious means to sustain communication in the face of difficulties (Canale, 1983; Canale & Swain, 1980; Ellis, 1985; Haastrup & Phillipson, 1983; Hatch, 1978; Rost, 1994; Rubin, 1987; Savignon, 1972, 1983). In Hatchs (1978) words, learners should be told to use whatever fillers they can to show the Native Speaker that they really are trying . . . . The most important thing of all has to be dont give up (p. 434). Canale (1983) specifically listed the use of pause fillers (p. 25) among the CSs making up strategic competence. Haastrup and Phillipson (1983) included in their taxonomy a set of strategies which they termed Strategies aimed at solving retrieval problems (p. 144) (e.g., er now I have to think), which appear to be similar to the ones we are talking about here. Rost (1994) also mentions using conversational fillers to keep the conversation going in his list of communication strategies. The question, then, is whether it is justifiable to include stalling strategies among CSs or not. Frch and Kasper (1983b) considered any filled pause (lexical or nonlexical alike) to be temporal variables of speech performance rather than CSs.1 According to them, therefore, fillers and hesitation devices are not CSs. On the other hand, it was Frch and Kaspers definitions of problem orientedness and consciousness as criteria of CSs, which originally prompted me to include stalling strategies as CSs: The conscious use of communication maintenance fillers and gambits appears to satisfy both criteria. From another perspective, Tarone (1980) distinguished between production and communication strategies, the former referring to general attempts to use the linguistic system efficiently and clearly, the latter being used more specifically to negotiate meaning by offering alternative means to communicate ones message. In this system, stalling strategies fall under production and not communication strategies. Although this distinction makes sense, I believe that it is difficult to draw the line exactly between the two types of strategy; for example, in its rigid application, this framework would restrict CSs primarily to achievement strategies; avoidance strategies would fall short of qualifying as real CSs because by using them ones meaning is not so much negotiated as reduced. In sum, there is clearly a need to provide a thorough analysis and typology of all the cognitive strategies that speakers use to enhance communication, but this goes beyond the scope of this study. This
1
However, Frch and Kasper (1983b) also pointed out that the exact functions of the various types of pauses are still far from being well-described (p. 215). 59
COMMUNICATION STRATEGIES
article uses the term communication strategies broadly to cover a wide range of communication-enhancing devices, including stalling strategies. It may be useful to point out that Tarone (1980) mentions learning strategies as a third type of strategy in her classification. These strategies are defined by Oxford (1990) as actions taken by second and foreign language learners to control and improve their own learning (p. ix). Even though such strategies are functionally different from communication-enhancing strategies, the distinction is not so clear at a closer glance. A great deal of language attainment takes place through taking an active part in actual communication, and CSs help learners to do so and thus (a) to obtain practice, and (b) to gain new information by testing what is permissible or appropriate. In fact, Tarone ( 1980) points out that in actual use all CSs may serve learning purposes; for the same reason, Oxford (1980) included compensation strategies as one of the six main classes in her system of learning strategies.
THE TEACHABILITY CONTROVERSY

The teachability of CSs has been a source of considerable controversy in the past decade. Whereas strong theoretical arguments reject the validity and usefulness of specific CS training, practical considerations and experience appear to support the idea. A brief summary of the problem and the arguments follows. Tarone (1981) points out that CSs, rather than being part of linguistic knowledge, are descriptive of the learners pattern of use of what he/she knows as he/she tries to communicate with speakers of the TL [target language] (p. 63). What is more, most researchers would agree that strategic competence develops in the speakers L1 and is freely transferable to target language use (see Bongaerts & Poulisse, 1989; Bongaerts, Kellerman, & Bentlage, 1987; Kellerman, Ammerlaan, Bongaerts, & Poulisse, 1990; Paribakht, 1985). This means that most adult language learners already have a fairly developed level of this competence, involving a repertoire of applicable CSs, regardless of their level of L2 proficiency (see Bialystok & Kellerman, 1987). If, therefore, there is no new linguistic knowledge involved and the cognitive processes are familiar from the L1, what then is the point in teaching these strategies? As Kellerman (1991) concludes, there is no justification for providing training in compensatory strategies in the classroom . . . . Teach the learners more language and let the strategies look after themselves (p. 158). After providing a comprehensive overview of strategy use and language processing, Bialystok (1990) argues that communicative strate60 TESOL QUARTERLY
gies are reflections of underlying psychological processes, and therefore it is unlikely that focusing on surface structures will enhance strategy use or the ability to communicate. Her conclusion is very similar to Kellermans: The more language the learner knows, the more possibilities exist for the system to be flexible and to adjust itself to meet the demands of the learner. What one must teach students of a language is not strategy, but language (p. 147). Canale and Swain (1980) also believe that CSs are most likely to be acquired in real-life communication and not developed through classroom practice. The arguments above are well-founded. Still, many other researchers maintain that strategy training is possible and desirable (e.g., Brooks, 1992; Chen, 1990; Frch & Kasper, 1983a, 1986; Haastrup & Phillipson, 1983; Paribakht, 1986; Rost, 1994; Rost & Ross, 1991; Savignon, 1972, 1983, 1990; Tarone, 1984; Tarone & Yule, 1989; Willems, 1987). The sources of this seeming contradiction, I believe, lie in the following three observations: 1. Most of the arguments on both sides are based on indirect evidence. 2. There is variation within CSs with regard to their teachability. 3. The notion of teaching allows for a variety of interpretations.
Indirect Evidence
Very little systematic strategy training research has been conducted thus far to test the teachability of CSs. I share Bialystoks (1990) view that there is little empirical research investigating the pedagogy of CSs, so descriptions and evaluations of any procedure are somewhat speculative (p. 149). Most arguments concerning the teachability issue are based on indirect or inconclusive evidence, but it must be noted that some of these data actually appear to confirm the validity of strategy training. Some studies did investigate the potential usefulness of the specific training of some CSs. These were, however, either too narrow in scope to be generalizable (i.e., focusing only on one strategy), or did not follow rigorous experimental research methods. Wildner-Bassett (1986) provides evidence, for example, that explicit instruction can increase both the quality and quantity of time-gaining fillers used by students. Frch and Kasper (1986) and Tarone and Yule (1989) report on four different classroom projects that successfully incorporated strategy training into foreign language instruction. Rost (1994) conducted a questionnaire survey among teachers of conversation-based L2 classes, in which they were asked to indicate to what extent they considered certain (primarily interfactional) communication strategies
COMMUNICATION STRATEGIES 61
to be useful and teachable. Several strategies, including using conversational fillers, were considered highly teachable. There are also some indications in the literature that learners who have been exposed to certain L2 input do improve their strategic competence. Tarone (1981) reports on a study by Piranian investigating learners of Russian, in which learners who had had some extracurricular exposure to Russian were found to use strategies more often and more effectively than their peers whose Russian experience was limited to the classroom. Raupach (1983) had similar findings with a group of learners of French who had spent a term in France: Whereas the interviews following the stay abroad showed no appreciable progress in the learners command of grammatical structures, there generally was a considerable change in the use of communication strategies (p. 207). Bialystok (1983) found that those subjects who had travelled widely and spoke more than two foreign languages proved to be superior in their L2 strategy use. There is also some evidence that students in classroom settings which offer more natural input (such as immersion classes) tend to develop a higher level of strategic competence (see Tarone, 1984) than students in ordinary classrooms, who tend to use only a limited number of mostly unsophisticated CSs (see also Willems, 1987).
Variation Within CSs

The range of strategies researchers include when they talk about communicative strategies varies from study to study. Most references in the literature to the teaching of CSs involve generalizations (either in favor of or against teaching them) based on one or two strategy types, and the current study, though attempting to investigate a range of strategies, is no exception to this. This approach is obviously not ideal, as some strategies (such as message abandonment) are clearly not desirable to teach, whereas some others (e.g., circumlocution or appeal for help), as we will see below, are not only useful and desirable, but also involve certain core words and structures, which lend themselves readily to classroom instruction. This implies that the question of whether communication strategies in general are teachable or not may be too simplistic, and this maybe partly responsible for the controversial answers given to it.
Various Interpretations of the Notion of Teaching

Those who argue against teaching CSs claim, in broad terms, that there is no need to do so because learners are already familiar with them from their L1. This argument, however, is based on a narrow
62 TESOL QUARTERLY
interpretation of teaching, namely that of passing on new information, whereas in the L2 literature, teaching is often used in a broader sense, for example when we speak about teaching L2 reading skills to learners who can already read in their L1. A broader interpretation of teaching would involve the following six (interrelated) procedures, all relevant to strategy training. 1. Raising learner awareness about the nature and communicative potential of CSs by making learners conscious of strategies already in their repertoire, sensitizing them to the appropriate situations where these could be useful, and making them realize that these strategies could actually work. The importance of conscious attention in the learners internalization process in general is highlighted by Schmidt (1990) in his review of what cognitive psychology tells us about learning and memory. From a cognitive perspective, the main role of instruction is to orient the learners and focus their attention on a given topic. Frch and Kasper (1986) also emphasize the need to increase learners metacommunicative awareness (p. 187) with respect to strategy use. In fact, most definitions of CSs include (potential) consciousness as a major feature and, as they also point out, this implies that these strategies can be influenced by teaching (Frch & Kasper, 1984, p. 47). 2. Encouraging students to be willing to take risks and use CSs, that is, to manipulate available language without being afraid of making errors (Frch & Kasper, 1986; Yule & Tarone, 1990). Willems (1987) also argues that very often we need to make it clear to learners that for some strategies, their innate tendency to use them in free speech activities is quite a natural urge and nothing to be frowned upon (p. 356). It must be noted that Bialystok and Kellerman (1987) agree that the use of CSs should be encouraged, but they do not consider this part of teaching them: It is one thing to encourage their use (and create the conditions in which they can be used) and quite another to actively teach communication strategies in the classroom (p. 172). This is a good example of the fact that some of the teachability controversy stems from the different interpretations of what teaching involves. 3. Providing L2 models of the use of certain CSs through demonstrations, listening materials and videos, and getting learners to identify, categorize, and evaluate strategies used by native speakers or other L2 speakers. A variation of this structured inductive approach, described by Frch and Kasper (1986), is when conversations between the students and native speakers are recorded on video, and after viewing their own recordings, students analyze their own strategy use. 4. Highlighting cross-cultural differences in CS use might involve various degrees of stylistic appropriateness associated with CSs (e.g., in some languages particular CSs may be seen as indications of bad
style), differences in the frequency of certain CSs in the speakers L1 and L2, as well as differences in the verbalization of particular CSs. 5. Teaching CSs directly by presenting linguistic devices to verbalize CSs which have a finite range of surface structure realizations. According to McLaughlin (1990), verbal tasks are hierarchically structured and in order to realize a higher order goal, each of the component skills needs to be executed. This would imply that being familiar with a strategy in L1 might be an insufficient condition for efficient strategy use in L2 if certain lower order components are missing or not automatized properly. Tarone and Yule (1989) point out that circumlocution, for example, requires certain basic core vocabulary and sentence structures to describe properties (e.g., shape, size, color, texture) and function. They provide examples like top side, bowl-shaped, triangular, on the rim, circular, square. Drnyei and Thurrell (1992) consider the automatization of basic structures such as its a kind of/sort of the thing you use for. . . . its what/when you . . . , its something you do/say when . . . , necessary for circumlocution. They also provide a list of common fillers and hesitation devices which come in handy when learners wish consciously to buy time (e.g., well, actually, as a matter of fact, the thing is . . . . how shall I put it... ), as well as a set of ways to appeal for help (e.g., What do you call it/someone who . . . . Whats the word for . . . ). One good way of collecting such sets is by asking the learners to perform strategies in their L1 and then trying to find L2 equivalents for the structures and core lexis they used. 6. Providing opportunities for practice in strategy use appears to be necessary because CSs can only fulfil their function as immediate first aid devices if their use has reached an automatic stage. My experience in L2 teaching and CS training suggests that this automatization will not always occur without specific focused practice (see also Willems, 1987). Again, Kellerman (1991) acknowledges the possible usefulness of situational classroom practice of strategies in order to help learners overcome inhibitions arising from having to operate in the L2, but does not consider this part of actual strategy teaching since such exercises would be designed to help learners perform their competence, rather than build it up (p. 160).
Communication Strategy Training Versus Learning Strategy Training

It may be interesting to compare the six categories listed above to experiences gained from learning strategy training programs (for an overview, see Chamot, 1990; OMalley & Chamot, 1990; Oxford, 1990; Wenden, 1991). OMalley and Chamot (1990) emphasize that learning
64 TESOL QUARTERLY
strategy training should be direct, that is students should be apprised of the goals of strategy instruction and should be made aware of the strategies they are being taught (p. 184). This emphasis on directness is very similar to Oxfords (1990) and Wendens (1991) emphasis on informed training. As Oxford summarizes:
Research shows that strategy training which fully informs the learner (by indicating why the strategy is useful, how it can be transferred to different tasks, and how learners can evaluate the success of this strategy) is more successful than training that does not. (p. 207)
Thus, learning strategy training is found to be most efficient if it is explicit (direct, informed), which I have argued to be the case for CSs as well. The components of direct training of learning strategies, according to the above authors, include awareness training (Oxford, 1990, p. 202) offering a general introduction to the concept of learning strategies and strategy training; identification of the strategies students are already using; encouragement of strategy use in general; direct explanation of the use and importance of new strategies; initial demonstration, naming and modeling of the new strategy by the teacher; guided in-class practice of the new strategy followed by a cyclical review; exploration of the significance of the strategy and the evaluation of the degree of success with it; student identification of additional strategies and their potential applications; and, finally, the transfer of the new strategies to new tasks. Many of the above elements show a remarkable similarity to the CS training components listed earlier. There are two components of CS training absent herethe highlighting of cross-cultural differences in CS use and the actual teaching of linguistic devicesand this is because they are closely associated with the verbal nature of CSs.
THE INVESTIGATION: RESEARCH QUESTIONS AND DESIGN

Research Questions
In order to obtain empirical data on the teachability of CSs, we conducted a strategy training course and assessed the effects of the treatment using pre- and posttests and comparing the results with those obtained from control groups. We were interested in how strategy training affected some qualitative and quantitative aspects of strategy use as well as the rate of delivery of speech. We also wanted to find out how language proficiency affected the results and what
students affective dispositions were toward such training. Thus, we formulated five research questions: 1. Does the training of a specific strategy increase the frequency of the use of this strategy by the students? 2. Does the training of a specific strategy improve the quality (efficiency) of this strategy in actual language use? 3. Does strategy training have a direct impact on the students speech rate? 4. Is the success of strategy training related to the students initial level of language proficiency? 5. What are the students attitudes toward strategy training and the usefulness of CSs?
Strategies Investigated
The research focused on the training of three CSs and offered both awareness and practice activities. These strategies were: (a) topic avoidance and replacement, (b) circumlocution, and (c) using fillers and hesitation devices. By selecting three different types of strategies, we intended to increase the range of our training program. We assumed that including topic avoidance and replacement skills in the training could improve learners fluency along the lines of the old slogan, Language learners should say what they can and not what they want to, or along the lines of a variation on this slogan, Language learners should be encouraged to say what they can, rather than retreat silently from what they cant. 2 Circumlocution is often seen as the most important achievement strategy, and most of the existing strategy training activities focus on it. The ability to use fillers and hesitation devices plays an important role in helping a person to remain in the conversation and gain time to think; we have found in the past that teaching fillers brings about an improvement in students fluency. Research Design The study had a quasiexperimental design (i.e., it involved intact EFL learner groups) and included a treatment group and two types of control group: In the first type, students received no treatment at all but followed their regular EFL curriculum; in the second, general conversational training was given without any specific strategic focus.
2
I am grateful to George Yule for suggesting this variation on the slogan.
66
TESOL QUARTERLY
METHOD
Subjects
Subjects were 109 students (72 girls and 37 boys), aged 1518, studying English in 8 class groups in 5 different secondary schools in Hungary. Table 1 presents a summary of the subjects according to schools and class groups in the treatment and control groups. I selected these two schools for the treatment group because I was supervising two theses on communication strategies, so their classes were used for the treatment. The three schools for the control groups were selected because I had some personal contacts there with teachers who were ready to participate in the project. The selection of the schools and teachers was intentional, and we tried to control for as many other variables as possible. The 5 schools were of the same type, gimnzium (similar to British grammar schools), providing general instruction and preparing students for further studies in higher education. They were all respectable but not particularly famous or elite schools. The 6 teachers involved in the project were in the same age group (2530), having had between 25 years of teaching experience. Students in all 8 groups followed a similar EFL curriculum (the Hungarian national curriculum), using coursebooks published in Great Britain. Group sizes ranged from 13 to 18 (people who were absent during the pre- or posttests were not included in the investigation), which is the usual size for EFL class groups in Hungarian secondary schools (for a more detailed description of the EFL teaching situation in Hungary, see Drnyei, 1992; Medgyes, 1993). Because the research also involved the investigation of the effect of L2 proficiency on strategy use, we selected classes of different English proficiencies to ensure sufficient variation. All the students had been studying English between 1.5 and 3.5 years and had received between TABLE 1
Subjects Class Groups, Schools, and Group Types
Treatment Group (n =53)
Control Groups (n =56) No-Treatment Group Conversational Training Group School 5
School 1
School 2
School 3
School 4
Class 8 Class 4 Class 5 Class 6 Class 7 Class 2 Class 3 Class 1 Teacher 1 Teacher 2 Teacher 2 Teacher 2 Teacher 3 Teacher 4 Teacher 5 Teacher 6 16 12 11 14 12 12 15 17
67
200 and 480 English lessons; their EFL proficiency ranged from preintermediate to postintermediate (about 1+ to 2+ on the U.S. Foreign Service Institute scale). Both the treatment and control groups included some higher and some lower level classes.
Description of the CS Training Program

The experiment consisted of a 6-week strategy training program, embedded in the pupils official secondary school English course. The 4 class groups receiving treatment (n = 53) were taught by 2 teachers following exactly the same syllabus. The strategy training took place in three lessons each week, lasting for about 2040 minutes each time. The teaching material was based on the techniques described by Drnyei and Thurrell (1991), supplemented with awareness-raising discussions and feedback. An attempt was made to cover all the six types of CS teaching procedures listed above. In order to learn to use topic avoidance and replacement strategies, students were taught to go off the point, evade answers, and steer the conversation in a given direction. First the teachers provided demonstrations of the strategies, then students were asked to perform these in their L1. In the next stage, students were given time to prepare their manoeuvres in English and after the performance their achievement was discussed; later during the course, an increasing amount of improvisation was required. The activities focusing on circumlocution involved comparing various dictionary definitions and analyzing the structure of effective ones. Students were then given various tasks in which they had to describe objects and later more abstract notions, to extend definitions using long relative clauses, and play games such as Call my Bluff. The training of the use of fillers involved first collecting and classifying fillers, then inserting fillers into dialogues, lengthening dialogue turns as much as possible by adding sequences of fillers, expressing hesitation explicitly by using fillers, and matching fillers with different emotions and moods. The Appendix contains a selection of the most typical activities used to teach the three strategies. The program had a cyclic content design with each of the three teaching topics recurring and being further elaborated in every third lesson. The rest of the English lessons were typical foreign language classes, including a balanced teaching of integrated skills, using standard British coursebooks such as Access to English (Coles & Lord, 1975) or Headway (Soars & Soars, 1987).
Control Groups
Of the 8 class groups in the sample, 4 served as control groups (n = 56). These were divided into two parts. Two groups (n = 24) re68 TESOL QUARTERLY
ceived no treatment at all but followed their regular EFL curriculum; students were not even told that an experiment was in progress but were only asked to participate twice (with an interval of 6 weeks) in a recording activity (i.e., the pre- and posttests). In the other 2 control groups (n = 32) subjects were exposed to a conversational training supplement to their normal English classes (similar in length to the strategic supplement of the treatment group), without any specific strategic focus. The conversational training activities included communicative tasks such as role-play, games, and discussions, involving a lot of pairwork and groupwork. Students in these groups were told in advance that they would take part in an interesting communicative experiment.
The Pre- and Posttests

All the students took a written and an oral test before the program and the oral test again after the training (or after 6 weeks in the case of the no-treatment group). The elicited speech was recorded and transcribed. Students in the treatment group also filled out a questionnaire assessing how interesting and useful they had found the training. The written test consisted of the following. 1. The Test of English for International Communication (TOEIC), a standardized multiple-choice test (listening and reading sections) offered by the Educational Testing Service, Princeton, New Jersey. (The control groups were given a shortened version.) 2. The C-test, an integrative pencil and paper test; the particular version used had been validated with Hungarian EFL learners (see Drnyei & Katona, 1992). 3. The oral test consisted of the following. Topic description: Students were given an abstract topic (e.g., vegetarianism, marriage, peace) and were asked to talk about it for 3 minutes. Cartoon description: Students were asked to describe the content of a cartoon strip consisting of three to four pictures. Definition formulation: Students were given five Hungarian words related to school or family life (e.g., child care benefit, school leaving certificate, specialization course) and were asked to provide a definition
69 TESOL QUARTERLY
or an explanation in English. The target words were chosen from a pool of eight words in the pretest and seven words in the posttest.3
Variables in the Survey

Definition Quality The effectiveness of the definitions the students provided was taken to reflect the quality of their use of circumlocutions. Even though the task of giving formal definitions does not fully represent the ability to generate circumlocutions in context, this method was used to control for the number and the topic of circumlocutions, thus ensuring comparability across students. The evaluation of the effectiveness of the definitions required a complex measure: The success of a circumlocution does not depend on its length or the speech rate it is delivered at but rather on whether the listener can identify the target word described. Therefore all definitions produced by the students were transcribed and inserted into Definition-Evaluation Questionnaires for judges who had to guess the key words that the definitions defined and write them in the questionnaire.4 To ensure that one judge did not evaluate more than one definition of a key word (because the task of finding out the item would have been significantly easier the second time), each judge was given only one questionnaire which included definitions of different key words (614 definitions on a questionnaire). This meant, however, that 95 Definition-Evaluation Questionnaires needed to be prepared to include every definition generated by the students in the pre- and posttests. Copies of these 95 questionnaires were distributed to more than 600 English majors at Etvs University, Budapest, who served as the judges. Data were obtained on a minimum of 9 out of the total of 10 definitions a student produced (the missing definitions are explained by the exclusion of some key wordssee Note 3). The evaluation of each definition was based on an average of seven judges guesses. The answers were evaluated on a 3-point scale (wrong = 0, semicorrect = 1, correct = 2). Based on the results, two composite measures, pretraining definition
3
The data for one item in the pretest and two items in the posttest had to be excluded because some of the students did not know their exact meaning or mixed them with other items, 4 The reason we used the written transcripts of orally produced definitions to be evaluated by the judges was largely practical: We did not want one judge to evaluate more than one definition of a key word because the task of finding out the item would have been significantly easier the second time. This meant, however, that hundreds of judges were needed to ensure multiple evaluations of each definition. With such a huge number, we were simply unable to play individually the recorded definitions to the selected judges.
70
TESOL QUARTERLY
quality and posttraining definition quality, were obtained by first averaging the judges quality scores for each definition for each individual and then averaging the definition scores a student obtained in the pretest and the posttest (thus these quality means ranged from 0 to 2, with 0 indicating that none of the definitions a student generated was correctly interpreted by any of the judges, and 2 that all the definitions were understood by every single judge). A third measure, definitionquality gain, was also computed by deducting the figure for pretraining definition-quality from that of posttraining definition-quality. Frequencies of Circumlocutions and Fillers All occurrences of circumlocutions and fillers in the students speech were identified by the author and a group of research assistants. We did not include the circumlocutions from the definition-formulating activity (where the actual task was to generate circumlocutions) unless further circumlocutions were embedded in the definitions. Also, we did not include fillers whose use was not appropriate in English but was rather influenced by L1 interference. The decisions about each occurrence were based on three researchers agreement; in case of different judgments, the issue was discussed until an agreement was reached. Here again frequency gain scores were computed by subtracting the pretest frequency scores from the posttest scores. Speech Rate Although the efficiency of the training of fillers and circumlocutions could be directly evaluated by computing quality and frequency gain scores (see above), the use of the third featured strategy in the training program, topic avoidance/replacement, was only indirectly assessed through the students fluency. The assumption was that topic avoidance/replacement skills have a positive effect on fluency and, therefore, an improvement in the use of this strategy will be reflected in an increase in the fluency measure. There are several ways of conceptualizing fluency (see Schmidt, 1992, for a comprehensive overview); we were particularly interested in one aspect, the ability to fill the time with talk, which contrasts with a characteristic feature of L2 speech (typical of learners whose contact with the L2 is mainly restricted to the L2 classroom) in which the learner keeps grinding to a halt, pauses for lengthy periods, and often gets so lost that the interlocutor loses patience, or a complete communication breakdown occurs. In measuring speech rate, fillers, lexicalized hesitations (e.g., gambits, but not those of hesitation), and repetitions are considered to be part of fluent speech even though such leximic units may be viewed as dysfunctional
intrusions and, as such, may be examples of a lack of fluency from the perspective of other fluency conceptions. Speech rate measures were computed by simply dividing the total number of words a student produced by the length of time of that particular stretch of speech (measured in seconds). Separate coefficients were obtained for the cartoon description and the topic description tasks (but not the definition formulation tasksee above) for both the pre- and the posttests. These were then averaged to form two composite measures: pretraining speech rate and posttraining speech rate, and again a gain score, speech rate gain was also computed. Language Proficiency A general language proficiency measure was computed by adding up the standardized scores of the C-test and the two subtests of TOEIC equally weighted. Perceived Usefulness of Training A Student Questionnaire asked the students to indicate on a 7-point scale how useful they considered the training of each of the three strategies to be. Attitudes Toward the Training Students indicated on a 7-point scale on the Student Questionnaire the extent to which they had enjoyed the course.
Statistical Analyses
The definition-quality gain in the three conditions (treatment group and two kinds of control group) was compared by means of a oneway ANOVA of the gain scores. Student improvement in the use of circumlocutions and fillers in the treatment and control groups was compared by means of the Chi-square test. The significance of speech rate gain was first tested by paired sample t tests separately in the three conditions, then the gain scores were compared using a one-way ANOVA. To detect interrelationships between the students language proficiency, speech rate, and measures related to strategy use, Pearson correlation coefficients were computed.
72
TESOL QUARTERLY
RESULTS AND DISCUSSION

Table 2 presents descriptive statistics of the quality of the students definitions in the pre- and posttests across the three conditions. As can be seen, in the treatment group there is an improvement in the quality of the definitions after the training, whereas in both types of control group the quality score decreases (possibly because the key words in the posttest were somewhat more difficult to define, which makes the increase in the treatment group even more noteworthy).5 In order to test whether these contrasting results were caused by the treatment, a comparison of the gain scores across the three conditions was carried out by means of a one-way ANOVA. The analysis (see Table 2) produced a significant result and the Scheff test showed that the difference between the treatment and the no-treatment conditions was significant. This was not the case between the treatment and the conversational training conditions at the p < .05 level. However, a comparison of the mean gains in these latter two groups indicated that the difference approached significance (F[1,83] = 3.44, p = .067) and we can therefore talk about a trend in favor of the treatment condition (see Hatch & Lazaraton, 1991, p. 232). We may conclude that the CS treatment was successful in improving the quality of the definitions the students generated as confirmed by the difference between the treatment and the no-treatment conditions. The reason for the conversational training group showing no significant difference from the treatment group, but only a trend, might lie in the nature of conversational training: Communicative activities often include information-gap elements, which can be considered indirect practice in strategy use, and this reduces somewhat the difference between the two types of training. Table 3 contains descriptive statistics of the frequency of circumlocutions and fillers in the pre- and posttests as well as the percentage of students who showed a positive change in the use of these strategies in the posttest. The means are the averages of raw frequencies for all the tasks (i.e., students in the treatment group used, e.g., fewer than two fillers on average during the whole of the pretest). As can be seen, in the treatment group the use of both circumlocutions and fillers
5
The key words to be defined were different in the pre- and posttests, so a within-group repeated-measure comparison of definition quality may not be very meaningful because it may reflect the difference between the difficulty levels of the sets of keywords rather than within-group change. For this reason, no paired-sample t test statistics are given in Table 2. Incidentally, these statistics are in accordance with the claim about the superiority of the treatment condition: In the treatment group the difference in definition quality is significant (t =-2.04, df= 52, p <.05), whereas in the two types of control group it is not (t = 1.61, df= 23, p = n.s.; t = .76, df = 31, p = n.s.).
73
TABLE 2
Descriptive Statistics of Definition Quality and One-Way ANOVA of Definition Quality Gain Across the Treatment and Control Groups
M Group Treatment Group (n = 53) Control Group (n = 56) No-treatment group (n = 24) Conversational training group (n = 32) Source Between groups Within groups Total SS 2 106 108 Pretest 1.50 1.49 1.58 1.43 df 1.01 13.99 15.00 Posttest 1.61 1.42 1.46 1.39 MS .51 .13 Gain +.11 .07 .12 .04 F 3.83 Pretest .30 .31 .35 .27
SD Posttest .27 .27 .27 .27
.025
increased, with the increase in the use of fillers appearing to be particularly substantial. In the control groups, on the other hand, there was only a minimal change in the frequency of circumlocutions, whereas the number of fillers actually decreased in the posttest.6 Because of the frequency data, parametric procedures such as the ANOVA were not appropriate here to test whether the changes had been caused by the treatment. Instead, a nonparametric test, the Chi-square test, was applied to compare student improvement in the three conditions, with the number of students who showed a positive change being the dependent and group type the independent variables. With respect to the use of fillers, the results of the comparison are highly significant, indicating that significantly more students in the treatment group showed improvement in their use of fillers (72%) than in the no-treatment group (13%) and the conversational training group (28%). This substantial increase as the function of the treatment is in accordance with Wildner-Bassetts (1986) findings. The Chi-square analysis, however, did not produce significant results with respect to circumlocutions (X2 [2] = 1.53, p = n.s.) in the three conditions (38%, 30% and 25%, respectively). There are two possible reasons for this: (a) The treatment affected not so much the frequency of the circumlocutions but rather their quality (which we have seen before). (b) Circumlocutions are not very common in everyday speech, and even in our tasks which were designed to pose language difficulties to the learners
6
In the no-treatment group the decrease of the number of fillers in the posttest is rather large. This was partly caused by one student, who used a disproportionately large number of fillers (30, mostly wells) in the pretest and only half as many in the posttest.
74
TESOL QUARTERLY
TABLE 3
Descriptive Statistics of the Frequency of Circumlocutions and Fillers and Chi-Square Test on the Percentage of Students Showing a Positive Change Across the Three Conditions
M Group Treatment Group (n = 53) Circumlocution Fillers Control Group (n = 55) Circumlocution Fillers No-treatment group (n = 23) Circumlocution Fillers Conversational training group (n= 32) Circumlocution Fillers
b
Pretest .57 (.80) 1.70 (2.22) .69 (1.05) 1.95 (4.32) .65 (.78) 2.52 (6.26) .72 (1.22) 1.53 (2.06)
2
Posttest .91 (.86) 6.36 (6.33) .69 (.88) 1.11 (2.30) .71 (.86) 1.04 (3.07) .66 (.90) 1.16 (1.55) df 2 2
Gain + .34 +4.66 .00 .84 +.06 1.48 .06 .37

n . s .c .000
Percentage of students showing positive change 38 72 27 22 30 13 25 28
Circumlocution Fillers
a b
1.53 29.23
Standard deviations are given below the means in parentheses. Because of missing data, one student was excluded from the control groups Nonsignificant at the < .05 level.
and thus elicit CSs, the frequency of circumlocutions was rather low. This, coupled with the limited sample sizes, may not have allowed for stable trends to emerge and the results to reach significance. Table 4 contains a comparison of the students pre- and posttest speech rate in the treatment and the control groups, as well as a comparison of the speech rate gains across the three conditions. As can be seen, in the treatment group the improvement in the students speech rate is highly significant. The two types of control group show a different pattern: There is no significant change in the no-treatment group, but students in the conversational training group improved in their speech rate significantly after the training. This latter result was
75
actually expected because the primary purpose of a conversational training supplement to a language course is to improve the students fluency. The one-way ANOVA of the gain scores across the three conditions did not produce a significant result, meaning that there were no differences between any of the three groups. The fact that the treatment and the conversational training conditions did not show any difference was not surprising in view of the significant increase of speech rate in the conversational training group. On the other hand, the lack of a significant difference between the treatment and the no-treatment groups was rather unexpected. One possible explanation may be that even though CSs help smooth out trouble spots in conversation and thus reduce thinking time and increase fluency, the general rate of speech delivery is also a function of other aspects of ones communicative competence we did not control for in our survey. A second possible explanation may be that even if the treatment condition does have a stronger effect on the students speech rate than the no-treatment condition, the resulting difference may not be sufficiently great after a 6-week strategy training supplement (18 x 2040 mins) with such a limited sample, and would require more participating students and/ or a longer training program to reach statistical significance. Table 5 contains correlations obtained in the treatment group beTABLE 4
Paired Sample t tests on Pre- and Posttest Speech Rate and a One-Way ANOVA of Speech Rate Gain in the Treatment and Control Groups
Group Treatment Group (n = 53) Control Group (n = 55)

b
df 52 54 22
Pretest .89 (.33) 1.05 (.33) 1.05 (.36) 1.06 (.32)
Posttest 1.06 (.35) 1.17 (.33) 1.13 (.35) 1.21 (.32) MS .07 .06
Gain +.17 +.12 +.08 +.15 F 1.08
t value 5.14 3.46 1.43 3.40
p .000 .001 n.s.c .002
No-treatment group (n = 23)
Conversational training group (n = 32) 31 Source Between groups Within groups Total
a b c
SS 2 105 107
df .14 6.78 6.92
n.s.
Standard deviations are given below the means in parentheses. Because of missing data, one student was excluded from the control groups. Nonsignificant at the < .05 level. TESOL QUARTERLY
76
tween the students pretraining language proficiency, speech rate, and variables describing their strategy use. It was expected that the students pre- and posttraining speech rate would be related to their language proficiency, that is, better students would be more fluent. This was indeed the case. What is important, however, is that the speech rate gain after the training is unrelated to the students language proficiency, which means that success in the training was not a function of the participants initial language competence. This is further confirmed by the correlations with definition quality, circumlocution, and fillers, where none of the gain scores correlated significantly with pretraining language proficiency (for an analysis of the relationship between language proficiency and strategy use in general, see Poulisse & Schils, 1989). The second column in Table 5 shows correlations between the students speech rate and strategy use before the training. The significant correlations indicate that both the quality and the quantity of the students strategy use were positively related to their speech rate. The correlation between speech rate and fillers is particularly high, implying that more fluent students tended to use time-gaining strategies. With respect to the correlations between variables after the training and correlations between variable-differences, (Columns 3 and 4), fillers again correlate highly significantly with speech rateindicating TABLE 5
Correlations Between Pretraining Language Proficiency y, Speech rate, and Variables Describing Strategy Use in the Treatment Group (n = 53)
Oral performance variables Speech Rate Pretraining Posttraining Gain Definition Quality Pretraining Posttraining Gain Circumlocution Pretraining Posttraining Gain Fillers Pretraining Posttraining Gain
Pretraining language Proficiency
Speech Rate Pretraining Posttraining Gain
* p < .05. ** p < . 01. *** p < .001. COMMUNICATION STRATEGIES 77
TABLE 6
Descriptive Statistics of the Affective Items from the Student Questionnaire
Variable Usefulness of fillers Usefulness of topic avoidance Usefulness of circumlocution Attitudes towards the training
M 5.68 5.87 6.04 5.93
SD 1.14 1.10 1.07 1.04
Lowest value 3 4 2 4
Highest Value 7 7 7 7
Note. The answers were given on 7-point scales ranging from not useful at all/I didnt like them at all (l) to very useful/I liked them a lot (7).
that the mastery of fillers is positively associated with improvement in speech ratebut neither definition quality nor circumlocution do so. This lack of significant correlations could be due to several facts: Because the training also focused on a third strategy, topic avoidance/ replacement (which was assumed to be directly related to speech rate), unremarkable use of definitions/circumlocutions in the posttest may have been compensated for by good topic avoidance skills, which depressed the correlations. Another explanation might be that enhanced use of circumlocution does not directly affect speech rate but rather the quality of general message conveyance, which was not measured. Table 6 contains descriptive statistics of the affective variables in the Student Questionnaire. The high value means indicate that students found the strategies in the training useful, especially circumlocution, and their general attitude toward the training was very favorable.
CONCLUSION
What prompted this study was my realization that a significant proportion of real-life communication in L2 is problematic (Gass & Varonis, 1991), and yet language classes do not generally prepare students to cope with performance problems. I assumed that one educational approach learners might potentially benefit from in developing their coping skills could be the direct teaching of CSs. My own experience, as well as indications in the literature, suggested that it was possible to develop efficient strategy training activities; however, the serious theoretical arguments of researchers questioned the teachability of CSs. Three possible sources of this controversy have been suggested here: (a) The arguments concerning the systematic training of CSs have been typically based on indirect evidence, and some of this evidence actually supports the teachability of strategic
78
TESOL QUARTERLY
competence. (b) There is variation within CSs with respect to their teachability and, therefore, no straightforward answer can be given to the question whether CSs in general are teachable or not. (c) Part of the contradiction stems from different interpretations of ones notion of teaching. To obtain empirical data regarding the potential usefulness of CS instruction, a training experiment was carried out, focusing on three different kinds of strategies. In the treatment group, the posttraining results showed improvement in measures related to both the quality and quantity of strategy use (quality of circumlocutions and the frequency of fillers and circumlocutions). A comparison of the gain scores with those obtained in the control groups provided evidence that the improvement in the quality of circumlocutions and in the quantity of fillers could indeed be attributed to the treatment; however, the same thing could not be confirmed about the quantity of circumlocutions, which was argued to be caused, at least partly, by the low frequency of this strategy in the corpus. As for the students speech rate, it was found that both the quality and the quantity of the students strategy use were positively related to their fluency in the pretest but only fillers affected speech rate in the posttest. With respect to the differences between the pre- and the posttest results, significant within-group gains were found in the groups that received CS and conversational training (but not in the no-treatment group), but a comparison of the three conditions did not have significant results. It was argued that the unexpected lack of significant difference between the treatment and the no-treatment groups may have been caused by the shortness of the training and the limited sample sizes; alternatively, fluency is also determined by other important variables related to communicative competence which were not focused on or controlled for in our study and which may not have changed significantly during the 6-week period of the treatment, thus depressing speech rate gain differences. With respect to the students level of L2 proficiency, the effectiveness of the training was found to be unrelated to the learners EFL competence. This implies that strategy training can be incorporated early as our study shows, even at a preintermediate levelin an L2 teaching syllabus, which is in accordance with Savignons (1972) original recommendation. It was also found that student attitudes toward such training were favorable, indicating that such training activities are relatively safe to use in the classroom. The results presented above are far from conclusive. Only three types of CS were examined, which does not allow for generalizations, and there were quite a few mixed messages. Furthermore, a very
79
simple conception of fluency was applied in our study (words per seconds), which did not take into account any qualitative aspects of fluency, that is, the quality and efficiency of message conveyance. Even bearing the above cautions in mind, the results of the CS training experiment are still promising. Although the experiment was a pilot study in the sense that we could not rely on any established methodology or the experiences of other teachers and researchers, the treatment was successful in improving some of the qualitative and quantitative aspects of strategy use. Future extensions and elaborations of the training program maybe expected to achieve even more marked results, and thus our project appears to provide some support to Tarone and Yules (1989) claim concerning the direct teaching of CSs:
In our suggestions for teaching sociolinguistic skills, we argued for an essentially inductive, integrative approach . . . However, for the purpose of developing communication strategies, we feel that a more focused and even explicitly didactic approach is possible. We differ in our approach from other researchers, who argue that communication strategies cannot be explicitly taught. (p. 114)
The direct approach to teaching CSs might involve the following procedures:
q
q q q q
Raising learner awareness about the nature and communicative potential of CSs Encouraging students to be willing to take risks and use CSs Providing L2 models of the use of certain CSs Highlighting cross-cultural differences in CS use Teaching CSs directly by presenting linguistic devices to verbalize them Providing opportunities for practice in strategy use
Finally, some teachers might have doubts about teaching CSs such as fillers or topic avoidance, language behaviors normally not encouraged in their own L1s. Why then do learners need them? The answer is that they provide the learners with a sense of security in the L2 by allowing them room to manoeuvre in times of difficulty. Rather than giving up their message, learners may decide to try and remain in the conversation and achieve their communicative goal. Providing learners help towards accomplishing this is, I believe, a worthy objective of communicative language instruction.
ACKNOWLEDGMENTS
I would like to thank Marianne Celce-Murcia, Evelyn Hatch, Mary Lee Scott, Sarah Thurrell, Tams Vradi, George Yule, and the anonymous TESOL Quarterly 80 TESOL QUARTERLY
reviewers for their very helpful comments and suggestions on earlier drafts. Also, such a laborious study could not have been carried out without the invaluable help of several students, friends, and colleagues; special thanks are due to Rka Asztalos, Enik Csomay, Andrea Fischer, Mria Gspr, Krisztina Kertsz, Gabriella Komor, Judit Kormos, Nra Kozki, and Sndor Nmeth, who contributed a great deal to the collection and processing of the data. I am also grateful to the dozens of colleagues who helped to administer the definition-evaluation questionnaires in their classes.
THE AUTHOR
Zoltn Drnyei is Associate Professor in the Department of English Applied Linguistics, Etvs University, Budapest, Hungary. His publications include papers in Language Learning, Language Testing, The Modern Language Journal, ELT Journal, and Studies in Educational Evaluation, as well as three ELT books, the latest of which, Conversation and Dialogues in Action (Prentice Hall; co-authored by Sarah Thurrell) was published in 1992.
REFERENCES
Bialystok, E. (1983). Some factors in the selection and implementation of communication strategies. In C. Frch & G. Kasper (Eds.), Strategies in interlanguage communication (pp. 100118). Harlow, England: Longman. Bialystok, E. (1990). Communication strategies. Oxford: Blackwell. Bialystok, E., & Kellerman, E. (1987). Language strategies in the classroom. In B.K. Das (Ed.), Communication and learning in the classroom community (pp. 160 175). Singapore: SEAMEO Regional Language Centre. Bongaerts, T., & Poulisse, N. (1989). Communication strategies in L1 and L2: Same or different? Applied Linguistics, 10, 253268. Bongaerts, T., Kellerman, E., & Bentlage, A. (1987). Perspectives and proficiency in L2 referential communication. Studies in Second Language Acquisition, 9, 171 200. Brooks, F. B. (1992). Can we talk? Foreign Language Annals, 25, 5971. Canale, M. (1983). From communicative competence to communicative language pedagogy. In J. C. Richards & R. W. Schmidt (Eds.), Language and communication (pp. 227). Harlow, England: Longman. Canale, M., & Swain, M. (1980). Theoretical bases of communicative approaches to second language teaching and testing. Applied Linguistics, 1, 147. Chamot, A. U. (1990). Cognitive instruction in the second language classroom: The role of learning strategies. In: J. E. Alatis (Ed.), Georgetown University Round Table on Language and Linguistics (pp. 496513). Washington, DC: Georgetown University Press. Chen, S. Q. (1990). A study of communication strategies in interlanguage production by Chinese EFL learners. Language Learning, 40, 155187. Coles, M., & Lord, B. (1975). Access to English: Getting on. Oxford: Oxford University Press. Cook, V. (1993). Linguistics and second language acquisition. New York: St. Martins Press. COMMUNICATION STRATEGIES 81
Corder, S. P. (1981). Error analysis and interlanguage. Oxford: Oxford University Press. Drnyei, Z. (1992). English teaching in Hungary: How far behind? Studies in Educational Evaluation, 18, 4756. Drnyei, Z., & Katona, L. (1992). Validation of the C-test amongst Hungarian EFL learners. Language Testing, 9, 187206. Drnyei, Z., & Thurrell, S. (1991). Strategic competence and how to teach it. ELT Journal, 45, 16-23. Drnyei, Z., & Thurrell, S. (1992). Conversation and dialogues in action. Hemel Hemstead, England: Prentice Hall. Ellis, R. (1985). Understanding second language acquisition. Oxford: Oxford University Press. Frch, C., & Kasper, G. (1983a). Plans and strategies in foreign language communication. In C. Frch & G. Kasper (Eds.), Strategies in interlanguage communication (pp. 2060). Harlow, England: Longman. Frch, C., & Kasper, G. (1983b). On identifying communication strategies in interlanguage production. In C. Frch & G. Kasper (Eds.), Strategies in interlanguage communication (pp. 210238). Harlow, England: Longman. Frch, C., & Kasper, G. (1984). Two ways of defining communication strategies. Language Learning, 34, 4563. Frch, C., & Kasper, G. (1986). Strategic competence in foreign language teaching. In G. Kasper (Ed.), Learning, teaching and communication in the foreign language classroom (pp. 179193). Aarhus: Aarhus University Press. Gass, S. M., & Varonis, E. M. (1991). Miscommunication in nonnative discourse. In N. Coupland, H. Giles, & J. M. Wiemann (Eds.), Miscommunication and problematic talk (pp. 121145). Newbury Park, CA: Sage. Haastrup, K., & Phillipson, R. (1983). Achievement strategies in learner/native speaker interaction. In C. Frch & G. Kasper (Eds.), Strategies in interlanguage communication (pp. 140158). Harlow, England: Longman. Hatch, E. M. (1978). Discourse analysis and second language acquisition. In E. M. Hatch (Ed.), Second language acquisition (pp. 401435). Rowley, MA: Newbury House. Hatch, E. M., & Lazaraton, A. (1991). The research manual: Design and statistics for applied linguistics. New York: Newbury House. Kehe, D., & Kehe, P. D. (1994). Conversation strategies: Pair and group activities for developing communicative competence. Brattleboro, VT: Pro Lingua Associates. Kellerman, E. (1991). Compensatory strategies in second language research: A critique, a revision, and some (non-) implications for the classroom. In R. Phillipson, E. Kellerman, L. Selinker, M. Sharwood Smith, & M. Swain (Eds.), Foreign/second language pedagogy research: A commemorative volume for Claus Frch (pp. 142-161). Clevedon, England: Multilingual Matters. Kellerman, E., Ammerlaan, T., Bongaerts, T., & Poulisse, N. (1990). System and hierarchy in L2 compensatory strategies. In R. C. Scarcella, E. S. Andersen, & S. D. Krashen (Eds.), Developing communicative competence in a second language (pp. 163-178). New York: Newbury House. McLaughlin, B. (1990). Restructuring. Applied Linguktics, 11, 113128. Medgyes, P. (1993). The national L2 curriculum in Hungary. Annual Review of Applied Linguistics, 13, 2436. OMalley, J. M., & Chamot, A. U. (1990). Learning strategies in second language acquisition. New York: Cambridge University Press. Oxford, R. L. (1990). Language learning strategies: What every teacher should know. New York: Newbury House.
82
TESOL QUARTERLY
Paribakht, T. (1985). Strategic competence and language proficiency. Applied Linguistics, 6, 132146. Paribakht, T. (1986). On the pedagogical relevance of strategic competence. TESL Canada Journal, 3, 5366. Pattison, P. (1987). Developing communication skills. Cambridge: Cambridge University Press. Poulisse, N. (1987). Problems and solutions in the classification of compensatory strategies. Second Language Research, 3, 141153. Poulisse, N., & Schils, E. (1989). The influence of task- and proficiency-related factors on the use of communication strategies: A quantitative analysis. Language Learning, 39, 1548. Raupach, M. (1983). Analysis and evaluation of communication strategies. In C. Frch & G. Kasper (Eds.), Strategies in interlanguage communication (pp. 199 209). Harlow, England: Longman. Rost, M. (1994, March). Communication strategies: Are they teachable? Paper presented at the 28th Annual TESOL Convention, Baltimore, Maryland. Rost, M., & Ross, S. (1991). Learner use of strategies in interaction: Typology and teachability. Language Learning, 41, 235273. Rubin, J. (1987). Learner strategies: Theoretical assumptions, research history and typology. In A. Wenden & J. Rubin (Eds.), Learner strategies in language learning (pp. 1530). Hemel Hemstead, England: Prentice Hall. Savignon, S. J. ( 1972). Communicative competence: An experiment in foreign-language teaching. Philadelphia, PA: The Center for Curriculum Development. Savignon, S. J. ( 1983). Communicative competence: Theory and classroom practice. Reading, MA: Addison-Wesley. Savignon, S. J. (1990). Communicative language teaching: Definitions and directions. In: J. E. Alatis (Ed.), Georgetown University Round Table on Language and Linguistics (pp. 207217). Washington, DC: Georgetown University Press. Schmidt, R. (1990). The role of consciousness in second language learning. Applied Linguistics, 11, 129158. Schmidt, R. (1992). Psychological mechanisms underlying second language fluency. Studies in Second Language Acquisition, 14, 357385. Selinker, L. (1972). Interlanguage. IRAL, 10, 209230. Soars, J., & Soars, L. (1987). Headway intermediate. Oxford: Oxford University Press. Tarone, E. (1977). Conscious communication strategies in interlanguage: A progress report. In H. D. Brown, C. A. Yorio, & R. C. Crymes (Eds.), On TESOL 77 (pp. 194203). Washington, DC: TESOL. Tarone, E. (1980). Communication strategies, foreigner talk and repair in interlanguage. Language Learning, 30, 417431. Tarone, E. (1981). Some thoughts on the notion of communication strategy. TESOL Quarterly, 15, 285295. Tarone, E. (1984). Teaching strategic competence in the foreign-language classroom. In S. J. Savignon & M. S. Berns (Eds.), Initiatives in communicative language teaching (pp. 127136). Reading, MA: Addison-Wesley. Tarone E., & Yule, G. (1989). Focus on the language learner. Oxford: Oxford University Press. Tarone, E., Cohen, A. D., & Dumas, G. (1976). A closer look at some interlanguage terminology: A framework for communication strategies. Working Papers on Bilingualism, 9, 7690. Vradi, T. ( 1973). Strategies of target language learner communication: Message adjust-
83
ment. Paper presented at the 6th Conference of the Rumanian-English Linguistics Project, Timisoara. Published in IRAL, 18, 1980, 5971. Vradi, T. (1992). [Review of the books Communication strategies: A psychological analysis of second-language use and The use of compensatory strategies by Dutch learners of English. ] Applied Linguistics, 13, 434440. Wenden, A. (1991). Learner strategies for learner autonomy. Hemel Hempstead, England: Prentice Hall. Wildner-Bassett, M. (1986). Teaching and learning polite noises: Improving pragmatic aspects of advanced adult learners interlanguage. In G. Kasper (Ed.), Learning, teaching and communication in the foreign language classroom (pp. 163 178). Aarhus: Aarhus University Press. Willems, G. (1987). Communication strategies and their significance in foreign language teaching. System, 15, 351364. Yule, G., & Tarone, E. (1990). E1iciting the performance of strategic competence. In R. C. Scarcella, E. S. Andersen, & S. D. Krashen (Eds.), Developing communicative competence in a second language (pp. 179194). New York: Newbury House. Yule, G., & Tarone, E. (1991). The other side of the page: Integrating the study of communication strategies and negotiated input in SLA. In R. Phillipson, E. Kellerman, L. Selinker, M. Sharwood Smith, & M. Swain (Eds.), Foreign/ second language pedagogy research; A commemorative volume for Claus Frch (pp. 162171). Clevedon, England: Multilingual Matters.
84
TESOL QUARTERLY
Appendix
Examples of Strategy Training Activities Used in the Study Topic Avoidance and Replacement
1. Avoiding giving information: The teacher addresses a student with a question that asks for specific information, for example, How old are you? The student must respond in two or three sentences without actually giving that particular information, for example, Well, thats an interesting question. Isnt it strange how people always feel that they need to know the age of a person? 2. Going off the point: Students are told that no matter what their question is, they must steer the conversation to a given topic, for example, judo. If the question is, for example, Does your grandmother own a pet? the answer might be something like this: Yes, my grandmother keeps an enormous Alsatian dog because it makes her feel safe when shes at home alone. When she was younger, of course, she didnt need a dog because she was extremely fit and active, and right up to the age of 60 she attended judo classes . . . .
Circumlocution
3. Comparing dictionary definitions: In small groups students look up entries for a given word in monolingual dictionaries and compare and discuss the differences. Then they are asked to prepare a perfect definition for the word in question by editing/compiling the dictionary definitions. 4. Challenging the definition: Students work in pairs. Each pair is given the name of an object, which they must define using a relative clause. Each pair in turn reads out their definition, while the other pairs check whether it is precise enough. If it is notthat is, if a pair can find another object the definition suitsthey get a point, and for another point they must give a more specific definition. Of course, this new definition is also open to challenge. After students have gained some competence in creating definitions, the task is made more difficult by giving them abstract notions (e.g., friendship, peace) instead of objects to describe. 5. Calling my bluff: Students are in groups of three. Each group is given a card with one very difficult word and its definition on it. Students must invent two convincing but false definitions of the same word. Then they read out the three definitions for the other groups to decide which is the real one.
Fillers and Hesitation Devices

6. Adding fillers: The teacher takes a short excerpt (two-three utterances) from a dialogue in the class textbook and puts it on the board. Students are divided into groups of two or three; each group in turn must add one filler to the dialogue, which the teacher then inserts into the text on the board. You may want to specify that each filler can be used only once. If a group fails to provide an extra filler, or offers one to be inserted at an inappropriate place, they drop out. 7. Composing nonsense dialogues: In pairs, students compose short nonsense dialogues that consist almost entirely of fillers; they may use names of cities, for example, as content words. For example: A: You know, I thought maybe London. B: Well, I see what you mean, and dont get me wrongthats very Chicagobut actually, as a matter of fact, I was thinking more along the lines of Montreal if you see what I mean. A: Really? But thats Istanbul! (For further strategy teaching ideas the reader is referred to the following publications: Drnyei & Thurrell, 1991, 1992; Kehe & Kehe, 1994; Pattison, 1987; Savignon, 1983; Tarone, 1984; Tarone & Yule, 1989; Willems, 1987.)
85
Interpretation Tasks for Grammar Teaching

ROD ELLIS
Temple University
Grammar teaching has traditionally consisted of giving learners opportunities to produce specific grammatical structures. Such an approach may prove ineffective because it does not take account of how learners acquire grammatical structures (e.g., Krashen, 1982). This article examines an alternative approach to grammar teaching one based on interpreting input. This approach emphasizes helping learners to notice grammatical features in the input, comprehend their meanings, and compare the forms present in the input with those occurring in learner output. The rationale for the approach is discussed as are the principles for designing interpretation tasks for grammar teaching.
lthough applied linguists now largely agree that L2 classroom acquisition occurs when learners participate in interaction that affords comprehensible input and output (Krashen, 1985; Long, 1983; Pica, 1992; Swain, 1985), they have also recognized that higher levels of grammatical competence require direct intervention in interlanguage development. A case has been made for supplementing activities designed to focus learners attention on message conveyance with activities that also require a focus on form (Ellis, 1993a; VanPatten, in press; White, 1987). How, then, should this be done? What kinds of grammar teaching will work best for acquisition? Traditionally, grammar teaching has been conducted by means of activities that give learners opportunities to produce sentences containing the targeted structure. These activities can consist of mechanical pattern-practice drills of the kind found in the audiolingual method or situational grammar exercises in which the target structure is contextualized in terms of some real or imaginary situation (see Ur, 1988, for examples). The underlying assumption of both types of activity is that having learners produce the structure correctly and repeatedly helps them learn it. This traditional approach faces a number of problems. First, second
87
language acquisition (SLA) research (e.g., Ellis, 1989; Pienemann, 1984) has shown that learners pass through a number of stages on route to acquiring the ability to produce a target language structure and that grammar teaching often does not alter this sequence. Teaching learners to produce a target structure that they are not ready to produce may not work. Second, asking learners to produce grammatical structures they find difficult and then correcting them when they make mistakes may increase their anxiety and result in a psychoaffective block to learning anything (Krashen, 1982). An alternative approach to grammar teaching is to design activities that focus learners attention on a targeted structure in the input and that enable them to identify and comprehend the meaning(s) of this structure. This approach emphasizes input processing for comprehension rather than output processing for production and requires the use of what I have termed interpretation tasks to replace traditional production tasks (Ellis, 1993b). This article describes and illustrates interpretation tasks for grammar teaching. I will begin, however, with a brief examination of the psycholinguistic rationale for a comprehension-based approach to grammar teaching.
A PSYCHOLINGUISTIC RATIONALE
Figure 1 presents a model of L2 acquisition (see Ellis, 1990, 1993a). This model, designed to address the role of formal instruction in acquisition, is based on a distinction between implicit and explicit L2 knowledge. Implicit knowledge is typically manifest in some form of naturally occurring language behavior (e.g., a conversation). It is intuitive and, therefore, exists in unanalyzed form. It can be abstract and structured (i.e., rule based) or chunklike (i.e., formulaic). Explicit knowledge typically manifests itself in some form of problem-solving activity (e.g., a sentence transformation exercise), but it can also be accessed in natural language use that allows time for monitoring, as represented by A in Figure 1. Explicit knowledge is held consciously and is stored in analyzed form. Unlike implicit knowledge, therefore, it is reportable. The model is a weak-interface model. That is, it hypothesizes that explicit knowledge of L2 items and structures may convert directly into implicit L2 knowledge (see B in Figure 1) but, as the dotted lines are intended to suggest, usually does not. This position is grounded in research which indicates that learners do not bypass developmental sequences (which I assume to reflect implicit knowledge) as a result
88 TESOL QUARTERLY
FIGURE 1
A Model of L2 Acquisition Incorporating a Weak Interface Position
Key: = primary processes ---------- = secondary processes
of practicing target structures (Ellis, 1989; Pienemann, 1984, 1989; Schumann, 1978). In addition to this direct relationship between explicit and implicit L2 knowledge, the model hypothesizes an indirect relationship, and it is this that is most important. The model proposes that explicit L2 knowledge facilitates implicit L2 knowledge in two principal ways, as shown in C in Figure 1. First, it helps learners notice linguistic properties of the input they otherwise might not notice. Input is processed by means of top-down strategies designed to derive the message content with maximum efficiency by utilizing context cues and also by bottom-up strategies with which the learner attends to and attempts to decode specific L2 items and structures. My claim is that bottom-up processing is a necessary condition of L2 acquisition (i.e., no noticing, no acquisition) and that if learners possess explicit knowledge of a specific feature, they are better able to engage in bottom-up processing. In other words, explicit knowledge helps learners obtain intake (i.e., to process grammatical information for short-term and maybe medium-term memory). This hypothesis owes much to Faerch and Kaspers (1986) views concerning the importance of bottom-up processing for acquisition and to Schmidts (1990, 1993) ideas about the role of consciousness in language learning. Faerch and Kasper (1986) suggest that whereas effective listening inGRAMMAR INTERPRETATION TASKS 89
volves the use of top-down processes, where learners utilize contextual information and existing knowledge to understand what is said, the acquisition of new linguistic forms may require the use of bottom-up processing, where learners pay attention to forms that are problematic to them. Schmidt argues that no learning is possible without some degree of consciousness. He distinguishes between intentionality and attention, arguing that while the intention to learn is not always crucial to learning, attention (voluntary or involuntary) to the material to be learned is (Schmidt, 1992, p. 209). Neither Faerch and Kasper nor Schmidt are suggesting that learners attend consistently to form when they are communicating. Clearly that is not possible if communication is to proceed smoothly. At certain points, however, their attention may be directed away from comprehending for meaning to attending to and subjectively noticing specific language forms. Second, intake is also enhanced when learners carry out a second operation-comparing what they have noticed in the input with what they currently produce in their own output. This kind of cognitive comparison 1 is hypothesized to help learners identify what it is that they still need to learn. It can serve two functions: It can help learners notice the gap (Schmidt & Frota, 1986, p. 310) between the input and their own output, and it can give the learner evidence that an existing hypothesis regarding a target language structure is the correct one. In other words, cognitive comparisons serve as a mechanism for disconfirming or confirming hypotheses in implicit knowledge. This model allows us to identify a number of processes2 involved in learning and using grammatical features. Interpretation
This is the process by which learners endeavor to comprehend input and in so doing pay attention to specific linguistic features and their meanings. It involves noticing and cognitive comparison and results in intake.
Integration
Integration occurs when learners are able to incorporate intake into their developing interlanguage systems (i.e., their implicit knowledge). Not all intake is so accommodated, as learners are only able to incorporate features
1
The term cognitive comparison replaces the term noticing the gap used in previously published versions of this model (see Ellis, 1993a). This is because this term better captures the fact that learners need to notice when their own output is the same as the input as well as when it is different. 2 VanPatten (in press) has identified a very similar set of processes. He refers to them as input processing, accommodation and restructuring, and monitoring, acress, retrieval, speech accommodation.
90
TESOL QUARTERLY
for which they are ready.3 Integration may also be accompanied by restructuring (McLaughlin, 1990). That is, the incorporation of new linguistic material may cause learners to reorganize the information in their existing interlanguage systems. Production
Production typically relies on implicit knowledge (cf. Krashens Monitor Model), but this can be supplemented by explicit knowledge through monitoring (see A in Figure 1). Production does not serve as the primary means for acquiring new linguistic knowledge although it may help learners to gain mastery over features that have already entered their interlanguage (i.e., it can lead to greater accuracy).
One implication of this model for pedagogy is that grammar teaching might usefully focus on interpretation. As VanPatten (in press) puts it:
Given the important role of input and input processing in second language acquisition, it is reasonable to wonder whether or not explicit instruction in grammar that involves a focus on input is more appropriate than traditional approaches to grammar instructin where learners are engaged in production.
Although the model also affords other roles for grammar teaching (e.g., consciousness raising to develop learners explicit knowledge and production practice to help learners use already learned features more accurately), it suggests that teachers might profitably try to focus learners attention on noticing and understanding specific grammatical features in input, as it is by this means that the acquisition of new features gets started. Before we consider how this might be done, however, we will briefly examine what empirical evidence there is in favor of interpretation-based grammar teaching.
SOME EMPIRICAL EVIDENCE

Empirical support for an input-based approach to teaching grammar can be found in the early studies of the comprehension approach (see Winitz, 1981). A general finding of these studies was that beginning learners who were exposed to input they were required to comprehend but not asked to produce outperformed learners following a more traditional, production-based program in tests of listening and reading comprehension (as might be expected) and did as well and often better
3
There are various ways of explicating what is meant by readiness. Pienemann and Johnston (1987), for example, suggest that the acquisition of developmental grammatical features is only possible if learners have developed the prerequisite processing operations.
GRAMMAR INTERPRETATION TASKS
91
in tests of speaking and writing. A good example of a comprehensionbased approach is Total Physical Response (TPR) (see Asher, 1977). This method follows a structural syllabus but does not involve production practice, at least in the early stages. Instead learners are asked to perform actions to demonstrate their understanding of commands that have been specially contrived to teach the structures. Asher has conducted a number of studies (e.g., Asher, Kusudo, & de la Torre, 1974), involving both children and adults, to evaluate the effectiveness of TPR in comparison to other, production-based methods, in particular, the audiolingual method. The results he reports demonstrate that TPR leads not only to better comprehension and production but also to enhanced motivation and greater persistence in language learning.4 In a review of comprehension-based approaches, Gary ( 1978) identifies four main advantages: (a) a cognitive advantage (i.e., better L2 learning), (b) an affective advantage (i.e., the avoidance of the stress and embarrassment that often accompanies trying to produce sentences in front of others), (c) an efficiency advantage (i.e., a comprehensionbased approach works equally well with low and high aptitude learners), and (d) a utility advantage (i.e., teaching listening skills helps a learner become functional in using the L2 and also enables a learner to continue their language study independently of the teacher). Although evaluation studies of comprehension-based approaches to language teaching demonstrate their effectiveness in promoting overall L2 proficiency, they do not show (or try to show) that comprehending input enables learners to acquire specific grammatical features. A number of recent studies, however, provide evidence of just this. Doughty (1991) investigated the effects of instruction on adult learners acquisition of relative clauses. The instruction took the form of a computer-assisted reading lesson, based on a text specially designed to include examples of the target structure. One group received help in understanding the text by means of expansions or clarifications of sentences containing relative clauses. A second group received explicit instruction on relative clauses. A third, control group just read the sentences. Doughty found that the first and second groups improved in their ability to produce relative clauses to a significantly greater extent than the control group. She also found that the first group outperformed both the second and third groups in a test that measured overall comprehension of the passage. In other words, the meaningoriented instruction directed at making sentences containing the target
4
Ashers studies evaluating TPR should be treated with some caution. They only examined beginning learners and typically did not include follow-up tests. It is not certain, therefore, whether TPR is equally effective with more advanced learners or whether the advantages are long term.
92
TESOL QUARTERLY
structure comprehensible seemed to work best because it led to both acquisition of the target structure and to better overall comprehension. VanPatten and Cadierno (1993) compared traditional productionoriented practice with listening practice that required learners to process ,specially contrived input. The study involved university-level Spanish learners and focused on object-verb-subject word order and clitic object pronouns in Spanish. They found that the learners who were asked to process input by means of interpretation-based grammar tasks outperformed those taught by means of production-based practice on a test that measured comprehension of the target structures and, more surprisingly, did just as well on a test that measured ability to produce the target structures accurately. These results were repeated in follow-up tests administered 1 month later. VanPatten and Cadierno suggest that whereas the production-based instruction only contributed to explicit knowledge, the comprehension-based instruction created intake which the learners were able to integrate into their interlanguage systems (i.e., it led to implicit knowledge). A somewhat similar study was carried out by Tuz (1992) on Japanese university students studying general English. In this case the target structure was word order with psychological verbs such as like, attract, and disgust (see next section). Both groups made use of a set of pictures depicting events involving psychological verbs, similar to those found in Activity 1 of the materials provided in the Appendix. In the case of the control group, the pictures were used as stimuli for sentence production, whereas in the experimental group, they were used to practice comprehension of sentences containing psychological verbs. The results of this study were even more striking than those of VanPatten and Cadierno. Again, the learners receiving the comprehensionbased instruction outperformed those receiving the production-based instruction on a comprehension test of the structure, but, in addition, they also outperformed them on a production test. The interpretation tasks used in this study enabled the learners to develop the kind of knowledge needed to both comprehend and produce the target structure and did so to a much greater extent than the production tasks. Unfortunately, the study had no follow-up test, so it is not possible to say to what extent this advantage was maintained over time. The research to date, therefore, suggests that comprehension-based instruction not only results in greater overall proficiency but is also more effective in enabling learners to acquire specific grammatical structures. One caveat is in order, however. The tests used in both the early studies of comprehension-based language teaching and in the later studies investigating specific grammatical structures were of the kind that allowed for the use of explicit L2 knowledge through monitoring. VanPatten and Cadiernos (1993) claim that comprehensionGRAMMAR INTERPRETATION TASKS 93
based instruction results in implicit L2 knowledge is speculative, therefore. Before we can be sure of this, we need to investigate whether the knowledge obtained through comprehension-based instruction can be used in spontaneous communication, where there is little opportunity to employ explicit knowledge through monitoring. VanPatten (in press) reports that a study currently in progress indicates that it is.
DESIGNING INTERPRETATION TASKS

Interpretation tasks have the following goals.
1. To enable learners to identify the meaning(s) realized by a specific grammatical feature (i.e., to help them carry out a form-function mapping). In this case, the goal is grammar comprehension, to be distinguished from what might be termed message comprehension, which can take place without the learner having to attend to the grammatical form. For example, on hearing the sentence: Id like three bottles please. a learner may be able to understand that bottles is plural in meaning without noticing the -s morpheme or understanding its function. 2. To enhance input (Sharwood Smith, 1993) in such a way that learners are induced to notice a grammatical feature that otherwise they might ignore. In other words, interpretation tasks are designed to facilitate noticing. 3. To enable learners to carry out the kind of cognitive comparison that has been hypothesized to be important for interlanguage development. Learners need to be encouraged to notice the gap between the way a particular form works to convey meaning in the input and how they are using the same form or, alternatively, how they convey the meaning realized by the form when they communicate. One way of fostering this is to draw learners attention to the kinds of errors that learners typically make.
Interpretation tasks can be devised as sequences of activities that reflect these three operations. That is, in the first instance, learners are required to comprehend input that has been specially contrived to induce learners to attend to the meaning of a specific grammatical structure, followed by a task that induces learners to pay careful attention to the important properties of the target feature, and finally by a task that encourages the kind of- cognitive comparison learners will have to perform ultimately on their own output. This proposal is, in fact, not so different from an earlier proposal of Ingram, Nord and
94 TESOL QUARTERLY
Dragt (1975). They suggested that the development of listening fluency required learners to pass through three phases: (a) the Decoding Phase, when learners were invited to respond to stimuli by selecting from alternative answers, (b) the Auditory-Response Phase, where learners were required to anticipate what was going to be said, and (c) the SelfMonitoring Phase, where learners were asked to identify errors or incongruities. (a) and (c) resemble the processes of noticing and erroridentification. However, (b) appears to involve the kind of top-down processing involved in message comprehension rather than the bottom-up processing needed for grammar comprehension. Two factors are important in selecting target structuresproblematicity and learnability. Problematicity can be determined by examining samples of the learners output in order to determine (a) which grammatical structures are not yet being used (i.e., the forms have not been acquired) and, also, more crucially, (b) the forms that are being used but incorrectly because their target function(s) has not yet been acquired. This will call for some kind of error analysis (Corder, 1974). The problems so identified become candidates for instruction, the final selection of which will need to take account of learnability. This concerns whether the learner is able to integrate new grammatical information into the interlanguage system. In the case of problems resulting from lack of knowledge of target forms it will be very difficult to decide when a particular group of learners are ready to acquire a specific new form. However, if the new learning required is that of assigning a different function to an already acquired form, learnability may be less of a problem. The best candidates for interpretation tasks, therefore, may be strictures for which the form is known but the meaning(s) realized by the form is not5. Many learners, for example, will be familiar with the simple form of regular verbs (e.g., come/comes) but not yet use this form to express general truths (e.g., Iron rusts if it gets wet.) or futurity (e.g., I fly to Tokyo next week.). A good example of a problematic structure for many intermediate learners is what Burt (1975) has referred to as psychological predicate constructions. Tuzs (1992) study demonstrated that Japanese learners do have considerable difficulty in both comprehending and producing sentences with such verbs. Burts article suggests that this difficulty may be one that learners with other L1s also experience. A psychological
5
It may not be necessary to totally exclude problems resulting from ignorance of form, however, if the instructional aim is not to effect a change in the learners interlanguage (i.e., long-term L2 mernory) but the lesser one of facilitating intake (i.e. short-term memory). In this case, learnabilty may not be an issue and the choice of target structures can be determined solely on the grounds of problematicity. It remains to be seen, however, whether learnability is only an issue where integration is concerned (as I suspect) or whether it also applies to noticing and comprehending.
95
verb (e.g., love, prefer, bore, and worry) is one that refers to some affective state. These typically occur in transitive constructions in which one noun phrase functions as expenenter and the other as a stimulus. The following might be considered the unmarked order:6 Experiencer + verb + stimulus (e.g., Mary loves cats.) whereas the more marked order is: Stimulus + verb + experience (e.g., Mary worries her mother.) The learning problem arises in the marked order. Learners overgeneralize the unmarked pattern, thus misunderstanding sentences that take the marked order. The above sentence, for example, may be understood as Mary worries about her mother. It can also result in production errors, as when a learner says: * He doesnt worry the cat. when intending to say that the cat doesnt worry him. Burt suggests that psychological predicate constructions are an example of global grammar in that they affect overall sentence organization and seriously interfere with communication. As such, they are prime candidates for instruction. To overcome the problem that they pose, learners need to (a) recognize that psychological verbs fall into two classes according to the order of the noun phrases that function as experience and stimulus and (b) discover which verbs belong to which class. Given that intermediate learners will already have acquired a knowledge of transitive constructions and will already be using many psychological verbs, the problem can be considered to be primarily one of function rather than form. An interpretation task for teaching marked psychological verbs is included in the Appendix. This begins with an activity designed to practice students comprehension of sentences containing a number of psychological verbs, some common and some not so common. In this activity, students are required to assess the truthfulness of a set of sentences in relation to pictures. The input is oral. For example, the students hear a sentence such as: She loved his hairstyle.
6
The grounds for considering experiencer + verb + stimulus unmarked are (a) some of the most common psychological verbs in English function in this way (e.g., like, enjoy, want), and (b) these verbs do not easily permit the alternative pattern (e.g., The book was wanted by Mary). However, psychological verbs that permit the stimulus + verb + experiencer pattern also easily permit the alternative pattern (e.g., Mary worries JohnJohn worries about Mary).
96
TESOL QUARTERLY
and evaluate it in relation to a picture which shows a woman looking admiringly at a young man with an exotic hairdo. The sentences are contrived in such a way that there are pronominal clues as to the correct meaning. For example, the pronoun his in the above sentence indicates that the sentence is about a mans hairstyle, not a womans. In this way, students can arrive at the correct interpretation of sentences even if they are not sure of which group a particular verb falls in. Another feature of Activity 1 is that learners are allowed to request repetition of sentences. This is to encourage the process of negotiating input, which a number of researchers (e.g., Long, 1983) have hypothesized is important for comprehension and acquisition. This activity is designed to have students grapple initially with meaning while encouraging them to pay attention to the syntactic relations between words. The second activity is more analytic. It focuses students attention on the experience in sentences containing both unmarked and marked psychological verbs. In this case the input is written so as to allow time for students to reflect on the sentences. They are asked to draw arrows to show who or what experiences the feeling described by the verb. For example if the verb is like the arrow will need to go from the subject of the sentence to the verb: Sometimes people like dogs. whereas if the verb is disgust, the arrow will need to go from the object of the sentence to the verb: Sometimes people disgust dogs. This activity has a consciousness-raising function. That is, it seeks to make students aware of the grammatical difference between psychological verbs such as like and disgust. It can be extended by other consciousness-raising activities (see Ellis, 1994). For example, students might be asked to classify the verbs in the sentences they are exposed to into two groups according to whether the experience is the grammatical subject or object. The teacher might also like to provide an explicit explanation of the difference between the two verb groups. The third activity requires attention to both the target form and the meaning of a set of sentences. VanPatten (in press) distinguishes between referential and affective or learner-centered activities. The former call for an objective interpretation of sentences, whereas the latter ask for a more personalized response. Thus, although Activities 1 and 2 are referential in nature, Activity 3 is learner centered. The students, for example, are asked to reveal something about their personal responses to attributes of women and men. If they read a sentence such as: Tall women frighten me.
GRAMMAR INTERPRETATION TASKS 97
and evaluate it as true, partly true, or not true for them. If time permits, the teacher can use the students responses in this activity to carry out a survey of what types of men/women the students feel positive and negative about. The final activityActivity 4focuses students attention on the difference between the correct way of using marked psychological verbs and the incorrect way. This is done by means of a dialogue which the students listen to. They hear an imaginary language learner (Koji) attempt to explain his reaction to different types of women to a nativespeaking friend. However, he has not yet learned that with marked psychological verbs the experience is the grammatical object rather than the subject. The result is that he produces such sentences as: * I frighten tall women. when he means to say: Tall women frighten me. His friend helps him by rephrasing the sentences correctly. The students task is to identify the incorrect sentences Koji produces and work out what he should have said.7 This task illustrates a number of general principles for the design of interpretation tasks in general. These are: 1. Learners should be required to process the target structure, not to produce it. 2. An interpretation activity consists of a stimulus to which learners must make some kind of response. 3. The stimulus can take the form of spoken or written input. 4. The response can take various forms (e.g., indicate true-false, check a box, select the correct picture, draw a diagram, perform an action) but in each case the response will be either completely nonverbal or minimally verbal. 5. The activities in the task can be sequenced to require first attention to meaning, then noticing the form and function of the grammatical structure, and finally error identification. 6. As a result of completing the task, the learners should have arrived at an understanding of how the target form is used to perform a particular function or functions in communication (i.e., they must have undertaken a form-function mapping).
7
It can be argued that learners will need subsequent opportunities to try to use psychological verbs in communicative production tasks (e.g., information-gap tasks). This is when they need to be encouraged to pay close attention to their own output. One way in which this might be achieved is through focused communication tasks (see Nobuyoshi & Ellis, 1993).
98
TESOL QUARTERLY
7. Learners can benefit from the opportunity to negotiate the input they hear or read. 8. Interpretation tasks should require learners to make a personal response (i.e., relate the input to their own lives) as well as a referential response. 9. As a result of completing the task, learners should have been made aware of common learner errors involving the target structure as well as correct usage. 10. Interpretation grammar teaching requires the provision of immediate and explcit feedback on the correctness of the students responses. However, the extent to which each principle is essential in the sense that it contributes to task-effectiveness or- affects learning outcomes remains to be seen.
CONCLUSION
A number of applied linguists (e.g., Krashen 1982; Prabhu 1987) have argued in favor of what I term a zero position where grammar teaching is concerned.8 That is, they have proposed that attempts to teach grammar should be abandoned and learners allowed to develop their interlanguages naturally by engaging in communication in the L2. This position is motivated by research showing that learners progress along a natural sequence of development for grammatical structures, which direct instruction is unable to circumvent. This article has proposed an approach to grammar teaching that is compatible with how learners learn grammar. Interlanguage development can be more readily influenced by manipulating input than output, an approach that requires interpretation tasks that cause learners to attend to specific grammatical properties in the input, to identify and understand the meanings they convey, and to compare the form-function mappings of the target language with those that characterize the interim stages of learners own interlanguage development. Interpretation tasks offer teachers the chance to intervene directly in interlanguage development. But they do not guarantee that their intervention will be successful because intake may not become part of implicit L2 knowledge. Nor is it the case that all grammar teaching should be comprehension based. There may be a role for other forms
8
Krashen (1982) does allow for some grammar teachingfor what he terms subject matter. This, however, has a very limited place and is only for students who are interested in the study of language per se (p. 119).
99
of grammar teaching, such as consciousness-raising (Ellis, 1994) and perhaps, also, traditional production-based instruction as a way of improving learners accuracy in the use of target language grammatical forms they have already acquired. Interpretation tasks are proposed as just onealbeit a highly promising oneof several ways of tackling grammar instruction. Finally, the emphasis this article has placed on grammar teaching is not meant to suggest that there is no room for tasks that invite learners to make a free selection from whatever current linguistic resources are available to them (e.g., information-gap tasks). A complete language program will include a variety of tasks that invite both a focus on form and a focus on message conveyance.
ACKNOWLEDGMENTS
This article was first presented at the 28th Annual TESOL Convention, Baltimore, March 1994. I would like to thank two anonymous TESOL Quarterly reviewers for their helpful comments.
AUTHOR
Rod Ellis is Professor of TESOL in the Department of Curriculum, Instruction, and Technology in Education, College of Education, Temple University. He has published widely in the general area of second language acquisition and, more particularly, in the application of research and theory to language teaching.
References
Asher, J. (1977). Learning another language through actions: The complete teachers guidebook. Los Gatos, CA: Sky Oaks Publications. Asher, J., Kusudo, J., & de la Torre, R. (1974). Learning a second language through commands: The second field test. Modern Language Journal, 58, 2432. Burt, M. (1975). Error analysis in the adult EFL classroom. TESOL Quarterly, 9, 5363. Corder, S. (1974). Error analysis. In J. Allen & S. Corder (Eds.), The Edinburgh course in applied linguistics, Vol. 3 (pp. 122154). London: Oxford University Press. Doughty, C. (1991). Second language instruction does make a difference: Evidence from an empirical study on SL relativization. Studies in Second Language Acquisition, 13, 431469. Ellis, R. (1989). Are classroom and naturalistic acquisition the same? A study of the classroom acquisition of German word order rules. Studies in Second Language Acquisition, 11, 305328.
100
TESOL QUARTERLY
Ellis, R. (1990). Instructed second language acquisition. Oxford: Blackwell. Ellis, R. (1993a). Second language acquisition and the structural syllabus. TESOL Quarterly, 27, 91-113. Ellis, R. (1993b). Interpretation-based grammar teaching. System, 21, 6978. Ellis, R. (1994, March). Metalinguistic knowledge and second language pedagogy. Paper presented at the Annual Conference of the American Association of Applied Linguistics, Baltimore. Faerch, C., & Kasper, G. (1986). The role of comprehension in second language acquisition. Applied Linguistics, 7, 257274. Gary, J. (1978). Why speak if you dont need to? The case for a listening approach to beginning foreign language learning. In W. Ritchie (Ed.), Second language acquisition research (pp. 185199). New York: Academic. Ingram F., Nord, J., & Dragt, D. (1975). A program for listening comprehension. Slavic and East European Journal, 19, 1 1 0 . Krashen, S. (1982). Principles and practice in second language acquisition. Oxford: Pergamon. Krashen, S. (1985). The input hypothesis. London: Longman. Long, M. (1983). Native speaker/non-native speaker conversation and the negotiation of comprehensible input. Applied Linguistics, 4, 126141. McLaughlin, B. (1990). Restructuring. Applied Linguistics, 11, 113128. Nobuyoshi, J., & Ellis, R. (1993). Focused communication tasks. English Language Teaching Journal, 47, 203210. Pica, T. (1992). The textual outcomes of native speaker-nonnative speaker negotiation: What do they reveal about second language learning. In C. Kramsch & S. McConnell-Ginet (Eds.), Text and context: Cross-disciplinary perspectives on language study (pp. 198237). Lexington, MA: D. C. Heath. Pienemann, M. (1984). Psychological constraints on the teachability of languages. Studies in Second Language Acquisition, 6, 186214. Pienemann, M. (1989). Is language teachable? Psycholinguistic experiments and hypotheses. Applied Linguistics, 10, 5279. Pienemann, M., & Johnston, M. (1987). Factors influencing the development of language proficiency. In D. Nunan (Ed.), Applying second language acquisition research (pp. 45141). Adelaide, Australia: Adult Migrant Education Program/ National Curriculum Resource Center. Prabhu, N. S. (1987). Second language pedagogy. Oxford: Oxford University Press. Schmidt, R. ( 1990). The role of consciousness in second language learning. Applied Linguistics, 11, 129158. Schmidt, R. (1992). Awareness and second language acquisition. Annual Review of Applied Linguistics, 13, 206-226. Schmidt, E., & Frota, S. (1986). Developing basic conversational ability in a second language: A case-study of an adult learner. In R. Day (Ed.), Talking to learn: Conversation in second language research (pp. 237-326). Rowley, MA: Newbury House. Schumann, J. ( 1978). The pidginization process: A model for second language acquisition. Rowley, MA: Newbury House. Sharwood Smith, M. (1993). Input enhancement in instructed SLA: Theoretical bases. Studies in Second Language Acqusiition, 15, 165179. Swain, M. (1985). Communicative competence: Some roles of comprehensible input and comprehensible output in its development. In S. Gass & C. Madden (Eds.), Input in second language acquisition (pp. 235-252). Rowley, MA: Newbury House. Tuz, L. (1992). Comparison of two grammar teaching options: Comprehension-based
101
instruction vs. production-based instruction. Unpublished manuscript, Temple University Japan. Ur, P. (1988), Grammar practice activities: A practical guide for teachers. Cambridge: Cambridge University Press. VanPatten, B. (in press). Grammar teaching for the acquisition-rich classroom. Foreign Language Annals. VanPatten, B., & Cadierno, T. (1993). Explicit instruction and input processing. Studies in Second Language Acquisition, 15, 225259. White, L. (1987). Against comprehensible input: The input hypothesis and the development of second language competence. Applied Linguistics, 8, 95110. Winitz, H. (Ed.). (1981). The comprehension approach to foreign language instruction. Rowley, MA: Newbury House.
102
TESOL QUARTERLY
Appendix
Activity 1: Comprehending
Listen to the sentences and decide whether they describe the pictures below. If you think they describe the picture put a check in the blank next to the picture. If you think they do not, put a cross. If you like you can ask the teacher to repeat a sentence. 1. She appreciated his singing.
2. His present offended her.
103
3. Her driving impressed him.
4. He deplored her laziness.
104
TESOL QUARTERLY
Activity 2: Paying Attention

Draw arrows to show who or what experiences the feeling described by the verb in these sentences. Use a dictionary to check the meanings of any verbs you do not know. Examples: Sometimes people like dogs. Sometimes people disgust dogs. 1. Mary worries her mother. 2. Cats bother Mary. 3. John prefers dogs. 4. Few politicians impress people. 5. Jane loves smart men. 6. Poor people envy rich people. 7. Sometimes teachers amuse their students. 8. Rabbits like children. 9. Sometimes men disappoint women. 10. Dolores mourns her father.
Activity 3: Responding Personally

Respond to each of these sentences with: True Partly true Not true 1. Tall women frighten me. 2. Women who can cook impress me. 3. Smartly dressed women impress me. 4. Very clever women overwhelm me. 5. Quiet women interest me. 6. Talkative women bore me. 7. Argumentative women confuse me. 8. Women with a sense of humor charm me.
Activity 4: Whats the Difference?

Listen to Randy talk to his Japanese friend Koji. Can you work out what Koji should have said? Listening text: Randy: You know something. I dont really like tall women. I get a bit scared by them. Koji: Yeah, I am the same. I frighten tall women. Randy: Sorry? Koji: I frighten tall women. Randy: Oh, you mean you get frightened by tall women. Koji: Yeah. And clever women too. I overwhelm clever women. Randy: I know what you mean. They overwhelm me too. Koji: But the worst are argumentative women. I confuse them. Randy: They confuse you? Koji: Uh? I mean I get confused by them. Randy: They dont worry me. I like a good argument. Koji: And the next worst is talkative women. I bore them. Randy: You bore them. Or they bore you. I think you mean they bore you. Koji: Yeah, they bore me.
105
The Role of Lexical Aspect in the Acquisition of Tense and Aspect

KATHLEEN BARDOVI-HARLIG DUDLEY W. REYNOLDS
Indiana University
This article presents the results of a study investigating the acquisition of the simple past tense, identifies areas of difficulty, and presents an acquisitionally based approach to instruction for the problematic areas. The study, a cross-sectional investigation of 182 adult learners of English as a second language at six levels of proficiency, showed that the acquisition of the past tense in English is not a unitary phenomenon, but that it proceeds in stages. These stages are determined by the meaning of verbs as they relate to the expression of action and time, what we will term lexical aspect. These findings show that the acquisition of tense by classroom language learners follows the same sequences of development (with instruction) that have been observed in the acquisition of adult learners and in children without instruction. In early stages, learners often do not use the past tense where it is preferred by native speakers, indicating an undergeneralization of the meaning of the past in the learner grammar. We present an approach to instruction aimed at increasing the use of the past to balance contextualized examples through the use of authentic text and focused noticing exercises to encourage the learners toward a more targetlike association of form and meaning.
his article examines the role of lexical aspect in determining the pattern of acquisition of the past tense by adult learners of English as a second language. Lexical aspect, one facet of verbal semantics, refers to the inherent temporal makeup of verbs and predicates. Temporal characteristics, such as whether a verb or verb phrase describes an action with inherent duration like talk and sleep, or is punctual like recognize and notice, or has elements of both duration and culmination like build a house and paint a picture, have been found to influence the acquisition of tense. To date, however, most studies have been largely anecdotal, relying on the spontaneous production of very few learners. This article presents a large cross-sectional study confirming that lexical
107
aspect influences acquisition even in classroom language learners and concludes with pedagogical suggestions basing classroom practice on acquisitional evidence.
BACKGROUND
Before presenting the theoretical framework for the study, a brief overview of the English tense/aspect system is in order. English marks both tense, the location of an event in time (Comrie, 1985), and aspect, ways of viewing the temporal constituency of a situation (Comrie, 1976, p. 3). In Example 1, John sings (present) and John sang (past) show a difference in tense. In Example 2, John sang (simple past) and John was singing (past progressive) show a contrast in grammatical aspect, although both are in the past tense. 1. Tense a. John sings. b. John sang. 2. Grammatical Aspect a. John sang. b. John was singing. Grammatical aspect is also sometimes called viewpoint aspect (Smith, 1983, p. 480) because the choice between progressive and simple, for example, often reflects the speakers view of the action. A single verb may show contrasting grammatical aspect as in Example 2, but its inherent lexical aspect does not change. In these sentences, sing has intrinsic duration whether in simple past or past progressive. A different predicate sing a song has both duration, the singing of the song, and a specific endpoint, the completion of the song (i.e., when there is no more song to sing). These differences are captured in the Vendler (1967) classification of lexical aspect which is based on a classification system traced back to Aristotle. The classification was first employed for second language acquisition (SLA) research by Andersen (1991), and we follow him in adopting it as the framework for this study. In the Vendler framework, there are four lexical aspectual classes: states, activities, accomplishments, and achievements. They can be distinguished by three features (Table 1): punctual, which distinguishes predicates that can be thought of as instantaneous or as a single point (begin to sing) from those with duration (sing a song); telic, which distinguishes predicates with endpoints (sing a song) from those without (sing); and dynamic, which distinguishes dynamic verbs (e.g., play, read a book, wake up) from stative verbs (e.g., seem and know).
108 TESOL QUARTERLY
TABLE 1
Semantic Features of Aspectual Categories
Lexical Aspectual Categories Features punctual telic dynamic States Activities + Accomplishments + + Achievements + + +
Note. From Anderson, R. W. (1991). Developmental sequences: The emergence of aspect marking in second langauge acquisition. In T. Heubner & C. A. Ferguson (Eds.), Second language acquisition and linguistic theories (p. 31l). Amsterdam, Netherlands: John Benjamins. Adapted by permission.
The most familiar division falls between the stative verbs and the nonstative, or dynamic verbs (activities, accomplishments, and achievements). This is represented by the feature [dynamic] in Table 1. States persist over time without change. Examples of state verbs include seem, know, need, want, and be (as in be tall). 1 Activity verbs have inherent duration in that they involve a span of time, like sleep and snow. They have no specific endpoint, as in I studied all week, and as such are atelic. Examples of activity verbs include rain, play, walk, and talk. Achievements are distinguished from the other dynamic verbs by the feature [punctual]. Achievement verbs capture the beginning or the end of an action (Mourelatos, 1981) as in The race began or The game ended, and can be thought of as reduced to a point (Andersen, 1991). Examples of achievement verbs include arrive, leave, notice, recognize, and fall asleep. Accomplishment verbs share features with activity verbs [ punctual] and achievement verbs [+ telic]. Like activity verbs, they have inherent duration, as in build a house or paint a painting. Like achievement verbs, they have a goal or an endpoint. In build a house, for example, the endpoint is the completion of the house, in read a book, the completion of the book. The classes of achievement and accomplishment verbs can be grouped together as telic verbs, known as events (Mourelatos, 1981).2
These and other examples of predicates representative of lexical aspectual classes are from Dowty (1979). We use the terms state verbs, activity verbs, and so on to refer to the members of the lexical aspectual classes. However, as noted earlier, the relevant unit is generally considered to be the predicate or verb phrase as in be tall, sing a song, or read a book. 2 There are also other analyses of lexical aspectual classes which are based on Vendler categories that might be of interest to the reader (e.g., Binnick, 1991; Quirk, Greenbaum, Leech & Svartvik, 1985). The literature on lexical aspectual classes has developed diagnostic tests that distinguish the aspectual categories from each other. (See especially Vendler, 1967; Dowty, 1979). One such test is the in + time phrase/ for + time phrase test which distinguishes activity verbs from accomplishment and achievement verbs. Activity verbs are acceptable with adverbial phrases such as for ten minutes but unacceptable with in ten minutes
1
THE ACQUISITION OF TENSE AND ASPECT
109
Acquisition Studies Related to Lexical Aspectual Classes

Most of the research on the acquisition of tense and aspect by learners of a second language in the framework of lexical aspect has investigated uninstructed learners (e.g., Andersen, 1991; Robison, 1990), and many studies have examined languages other than English (e.g., FrenchTrvise, 1987; Vronique, 1987; GermanKlein, 1986; von Stutterheim & Klein, 1987; SpanishAndersen, 1991). This crosslinguistic work suggests that the distribution of tense/aspect morphology in learner language may be influenced by lexical aspect. Andersens (1989, 1991) studies of two untutored children learning Spanish as a second language showed that the preterite was used early in punctual verbs (achievements) and spread gradually to accomplishments, activities, and finally to states. The use of imperfect moved from states, to activities, accomplishments, and finally to achievements. Robisons (1990) study of a very low proficiency adult untutored learner of English showed that punctual verbs were significantly more likely to show past tense marking than durative verbs (12% vs. 5.1% use of past in past time contexts) and that durative verbs were more likely to show -ing (e.g., I workin) than punctual verbs (20.6% vs. 5.1 %). Although the spread of Spanish tense/aspect morphology is not directly applicable to English, the principle is: Both studies suggest that lexical aspect determines the distribution of verbal morphology in untutored learners. Not only has this pattern been observed in SLA but it has also been reported in child (first) language development. Lexical aspect has been reported to play a role in the development of verbal morphology in the child language of English (Bloom, Lifter, & Hafitz, 1980), Italian, (Antinucci & Miller, 1976), French (Bronckart & Sinclair, 1973), and Greek (Stephany, 1981). (For a comprehensive review and interpretation of L1 studies, see Andersen & Shirai, 1994; Shirai, 1991.) Preliminary evidence suggests that instructed ESL learners also show the influence of lexical aspectual class on tense/aspect. In a study of the relation of form and meaning in interlanguage verbal morphology, Bardovi-Harlig (1992) reported that punctual verbs showed higher use of simple past than durative verbs. Although the number of learners (N = 135) was large in comparison to the studies of untutored learners, the relatively small number of verbs tested (three activity and two achievement verbs) makes the results only suggestive. Further evidence that classroom language learners show the influ(e.g., ]ohn slept for an hour/*in an hour). Accomplishment and achievement verbs are acceptable with in -phrases but unacceptable with for -phrases (e.g., John built a house in a year/*for a year). 110 TESOL QUARTERLY
ence of lexical aspect in their acquisition of tense comes from a study of English as a second language (ESL) and French as a foreign language (FFL), which examined narratives written by learners in a film retell task (Bardovi-Harlig & Bergstrm, in press). However, such naturalistic production is difficult to interpret for three reasons. First, the number of verb phrases in unguided production is not balanced across lexical aspectual classes. For example, achievement verbs far outnumbered the verbs in the other lexical aspectual classes, accounting for one half of all the verbs used by both groups. Second, the number of different verbs in the stative class is often limited. As might be expected, be is the most widely used stative verb in the ESL sample, occurring in 64% (90/140) of the stative sample. Finally, although Bardovi-Harlig and Bergstrom examined the production of 23 learners in each of the ESL and FFL environments, their study investigated four groups of learners cross-sectionally, which resulted in a fairly small number of learners at a particular level of proficiency (in the largest group n = 7, in the smallest, n = 4). The present study is designed to address some of the limitations of previous work by testing a reasonably large number of learners at multiple levels of proficiency on a variety of predicates, balanced across lexical aspectual classes. The goal of the study is to determine whether adult classroom language learners, like their untutored child and adult counterparts, exhibit the influence of lexical aspect in their acquisition of tense and grammatical aspect and to propose a pedagogical approach to facilitate acquisition. We thus test the following hypothesis:
Lexical aspect will influence the acquisition of simple past tense.
In a secondary hypothesis, we predicted that the effect of lexical aspect could be enhanced or diminished by the introduction of adverbs of frequency. Investigating the influence of adverbs of frequency provides additional evidence for the association of tenses with specific restricted meanings or environments in the learner grammar. Such an effect was anticipated on the basis of Bardovi-Harligs (1992) observation that the adverb of frequency usually triggered present tense use in past time contexts.
METHOD
Subjects
A cross-sectional study was conducted testing 182 adult learners at six levels of proficiency from beginning to advanced. All learners were enrolled in the Intensive English Program, Center for English
THE ACQUISITION OF TENSE AND ASPECT 111
Language Training at Indiana University. Placement was determined independently of this project by the battery of placement exams used by the program (which include composition, reading, grammar, and listening comprehension). There were 15 native languages represented in the sample: Arabic (48), Korean (46), Japanese (37), Spanish (18), Chinese (12), Portuguese (5), Thai (5), Italian (2), and Russian (2), with one speaker each of Bahasa Indonesian, Burmese, Catalan, German, Hebrew, and Turkish. The distribution of subjects by level is as follows: Level 1, 31; Level 2, 34; Level 3, 27; Level 4, 29; Level 5, 27; Level 6, 34.3 A control group of 29 native speakers of American English was also tested. All native speakers (NS) were graduate students at Indiana University.
Materials
Learners were given 32 short passages which contained 62 test items and 26 distracters testing verb forms not under investigation here. The passages varied in length from one sentence to five sentences and established time reference through the use of time adverbial or verb tense. Learners were given the base form of the verb and asked to supply the missing word or words in the blank. The target for each test item was determined by the native speaker responses. All verbs were tested in the third person singular environment so that overt morphological marking would be obligatory in the present as well as in the past. A sample test item is given in Example 3.
3. Last night John (work) ___very hard. He (write) ___ two papers and (finish)___ all of his grammar homework.
Broken down by lexical aspectual class, the 62 items testing the use of simple past tense included 14 achievements, 11 accomplishments, 12 activities, and 10 states. The effect of adverbs of frequency was tested on 9 additional activity verbs and 6 state verbs. Vocabulary was restricted to familiar lexical items and was checked by program teachers for appropriateness. (See Appendix A for a classification of all test items.)
ANALYSIS AND RESULTS

A distributional analysis of the responses was performed for each of the items and for each lexical aspectual class. A distributional analysis
3
Levels 16 are the first six levels of a seven-level program. Each instructional term is 7 weeks long and consists of 175 hours of instruction. Levels 1 and 2 may be thought of as beginning learners; Levels 35 as low to high intermediate; and Level 6 as low-advanced.
112
TESOL QUARTERLY
classifies all the verb forms the learners supply for each context and gives the percentage of the responses for each form supplied at each level (Bardovi-Harlig, 1992).4 Responses were grouped into categories according to verbal morphology: past, which included simple past tense forms and regularized past tense forms such as telled; 5 nonpast, which included simple present such as tells, and uninflected base forms such as (she) tell, (base only occurred where there was also a high occurrence of simple present); progressive, which included a verb+ -ing with no auxiliary, present progressive, and past progressive; perfect, which included all perfect forms; and miscellaneous forms, which included all remaining forms. The results presented in Tables 24 were figured in the following way. For each learner, the number of past, nonpast, and progressive responses to verbs in each lexical aspectual class was tabulated. Taking the use of verbal morphology with activity verbs as an example (see Table 3), first the number of past responses to the 12 activity verbs was tallied. The same was done for nonpast and progressive. This produced a usage score for each of the three main form types for each learner. The level percentages reported in Tables 24 represent the average of the usage scores for the learners in each level. The standard deviations indicate the range of variation among individuals at a given level. The results show that the acquisition of past tense is not a unitary phenomenon occurring simultaneously in all contexts. We find clear evidence that lexical aspectual class influences the sequence of acquisition of the past tense. This section presents the findings related to the distribution of the simple past, then examines the alternative forms used by the learners.
The Use of Simple Past

Learners produced 8,554 responses to the cloze passages without adverbs of frequency. The results showed that achievement and accomThe minimum TOEFL score required for entry in Level 4 is 400. The ideal scores for entry into Levels 5 and 6 are 450 and 500, respectively. This task was administered at the end of a term so that learners had completed the level by which they are identified. 4 We present the distribution of tense morphemes in the forms supplied (we have adjusted for incomplete answers by omitting them from the data). This has the greatest impact on Level 1 and 2 who had more difficulty in completing the task than students at other levels. This analysis brings these data into line with studies of free production data in which there is no such thing as an unanswered item (Bardovi-Harlig & Bergstrm, 1993; Robison, 1990; Schumann, 1987). 5 The regularized past tense forms were most frequent at Level 1 where learners scored 68.5% appropriate use with the regularized forms on achievements and accomplishments combined and 58.4% without them. Other levels showed virtually no use of regularized forms.
113
plishment verbs exhibit high levels of appropriate use of simple past even at the lowest levels of proficiency (see Table 2). Because the use of past with achievement and accomplishment verbs is similar throughout, with the possible exception of Level 1, we group them under the heading of events following Mourelatos (1981). (See Figure 1.) For event verbs, learners show approximately 80% appropriate use of simple past as early as Level 2. In contrast, activity verbs show much lower appropriate use of simple past. In Level 2, activity verbs show simple past in only 65.1% of the sample. The use of simple past stays low with activity verbs between 53.6% and 68.3% until Level 5. It is not until the advanced level of proficiency, Level 6, that learners show 82% appropriate use of simple past with activity verbsa rate reached for event verbs by high beginning learners in Level 2. Although the gap between event verbs and activity verbs for the appropriate use narrows considerably by Level 6, event verbs still show slightly more than a 10% advantage. State verbs, like activity verbs, show low rates of appropriate use of past, in Levels 13, but show higher rates than activity verbs in Levels 4 and 5, and the same rate as activity verbs by Level 6. An illustration of the pattern is provided in Figure 1. The results indicate that lexical aspectual class influences the use of simple past tense on this task. They further indicate that level of proficiency influences tense use. A MANOVA procedure (repeated measures) was performed to determine whether there was a statistically TABLE 2
The Use of Simple Past by Lexical Aspectual Class and Level in Percentage of Responses
Events Level 1 2 3 4 5 6 NS Note. ( ) = SD 114 TESOL QUARTERLY States Activities Accomplishments Achievements
FIGURE 1
Distribution of Simple Past by Lexical Aspect
Events
Activities
States
significant difference in past tense use across the lexical aspectual classes and across level of proficiency. Both lexical aspectual class and level were found to be significant ( F [3,525] = 102.47, p < .01; F [5,175] = 12.51, p < .01, respectively). The interaction between lexical aspectual class and level is also significant ( F [15,525] = 4.37, p < .01). Because of the relatively low rates of appropriate use of simple past with activity and state verbs we can say that learners undergeneralize the simple past; they do not use the simple past everywhere they could (or everywhere native speakers do). In the following sections, we examine the factors that contribute to the undergeneralization of simple past in the case of activity and state verbs.
Activity Verbs
Examination of the alternatives to simple past used by the learners also reveals the influence of lexical aspect. In the case of activity verbs
which were tested in contexts such as Example 4, the single main competitor to simple past in activity verbs is the progressive. The number given in parentheses indicates the percentage of native speaker responses that supplied the form shown.
4. After that the weather was nice so we (swim) swam (100%) in the ocean and (ride) rode (100%) bicycles along the beach.
The use of progressive is high in Level 1 at 24.6% of the responses (see Table 3). Progressives at this level consist of progressive forms swimming (14.1%) followed by present progressuch as sleeping and progressives sive (7.5%) and past progressive (3.0%). These emergent are rare after Level 2, and in Level 3 and above, the category consists primarily of past progressive usage. The rate of use of progressive forms with activity verbs drops from Level 1 to Levels 2 and 3. In Level 4, the use of progressive increases to 26.3%. At this level, the use of past progressive is the predominant form of progressive. (Level 4 was also a level in which the past progressive was addressed in instruction.) Finally, in Level 6 learners reach 82.0% use of simple past with activity verbs. We hypothesize that the use of progressive is higher in activity verbs than in any other lexical aspectual class because the meaning of the progressive action in progress at the moment (Andersen & Shirai, 1994, p. 148) is compatible with the inherent meaning of the lexical aspectual class. TABLE 3
The Distribution of Tense-Aspect Markers in Activity Verbs With and Without Adverbs of Frequency in Percentage of Responses
Activities w/o Frequency Adverbs Level 1 2 3 4 5 6 NS Past Nonpast Prog
Activities w/ Frequency Adverbs Past Nonpast Prog
Note. ( ) = SD
116 TESOL QUARTERLY
State Verbs
The responses to items containing state verbs favored a different alternative to the simple past. Ten items tested state verbs using a variety of lexical verbs as in Example 5.
5. Last night everything (seem) seemed (100%) very quiet and peaceful.
The main competing form to simple past in state verbs is the nonpast (see Table 4). The use of simple present, which Andersen and Shirai (1994) interpret as meaning continued existence (p. 148), is consistent with the enduring quality of state verbs. Learners do not use progressive forms with state verbs, which means that the past progressive is not being grossly overgeneralized. Instead, it is restricted to dynamic verbs and, specifically, to activity verbs. The exception to this is that state verbs show a modest use of progressive in Level 1 (7.3% of the responses), but this is only one third the use of progressive (24.6%) found in activity verbs at the same level of proficiency. The use of progressive with activity verbs, and its negligible use with state verbs, and the corresponding use of nonpast with state verbs and its much lower use with activity verbs provides further evidence that lexical aspectual class influences the learners use of verbal morphology. TABLE 4
The Distribution of Tense-Aspect Markers in Stative Verbs With and Without Adverbs of Frequency in Percentage of Responses
Statives w/o Frequency Adverbs Level 1 2 3 4 5 6 NS Note. ( ) = SD THE ACQUISITION OF TENSE AND ASPECT Past Nonpast Prog
Statives w/ Frequency Adverbs Past Nonpast Prog
117
Adverbs of Frequency
The meaning of past tense in the learner grammar was further investigated by testing activity and state verbs in the environment of adverbs of frequency as in Examples 6 and 7. The adverbs of frequency employed in the elicitation tasks included such common adverbs as always, often, usually, and everyday. (See Appendix A for a complete list.)
6. When George lived in Peru he (play) played (100%) soccer everyday. 7. When George was away at school he usually (seem) seemed (100%) happy, but really he often (feel) felt (97%) sad.
Nine additional activity verbs and six state verbs were tested in the environment of adverbs of frequency, yielding a total of 2,730 responses. With the introduction of adverbs of frequency in the environment of activity verbs, the appropriate use of simple past stays nearly constant (see Table 3). However, the competing forms change in the environment of the adverbs. The use of nonpast (i.e., simple present tense and base forms) increases noticeably, becoming the chief competitor to the appropriate use of simple past. The competing form in state verbs is always the nonpast (see Table 4). Unlike the case of activity verbs, when adverbs of frequency occur in the environment of state verbs, the rate of appropriate use of simple past tense falls and the use of nonpast increases noticeably, more than doubling in some cases. The occurrence of adverbs of frequency in the environment of state verbs enhances the association of state verbs with nonpast and the appropriate use of simple past drops. In contrast, in activity verbs, adverbs of frequency again introduce a nonpast reading; but in this case, nonpast replaces progressive as the most used alternative, the nontargetlike progressive being more susceptible than the targetlike use of past. The increase in the use of nonpast with adverbs of frequency in past-tense contexts shows that learners do not recognize such environments as environments for the simple past, revealing another way in which the distribution of past is undergeneralized in the grammars of some learners. The responses indicate that some learners associate the concept of present so strongly with adverbs of frequency that this association overrides contextual cues that establish the past tense. It is only at the advanced level of proficiency, where learners show approximately 80% appropriate use of past, that adverbs of frequency have little effect.
DISCUSSION
With respect to the hypothesis, the results show that learners treat event verbs (achievements and accomplishments) as best case examples
118 TESOL QUARTERLY
of past tense carriers at all levels of proficiency but show lower use of past with activity and state verbs. Thus, it appears that tutored learners, like untutored learners, are sensitive to lexical aspectual class with respect to tense use, not only at beginning stages of acquisition, but at higher levels of proficiency as well. We observed three stages in the acquisition of the simple past. In the first stage, event verbs show higher use of past than nonevent verbs (activity or state verbs). In the next stage (at about Level 4) state verbs begin to show higher use of past than activity verbs. Finally, activity verbs how the same rate of use of past as state verbs. The results show that the use of simple past is undergeneralized. The early use of simple past with event verbs suggests that learners find telic verbs to be the best case examples of past-tense carriers. State and activity verbs each show a different competitor, with state verbs showing high use of nonpast and activity verbs showing high use of progressive. The use of tense/aspect morphology with certain lexical aspectual classes reflects the inherent meaning of the verbs. Andersen (1991) links this to Bybees observation that inflections are more naturally attached to a lexical item if the meaning of the inflection has direct relevance to the meaning of the lexical item (p. 318). As learners move away from using verbal morphology in accord with lexical aspect, toward marking tense uniformly across lexical aspectual classes, they move toward a targetlike use of tense. Second, in addition to activity and state verbs, some learners find adverbs of frequency to be an unlikely environment for the simple past. This provides additional support that learners associate the notion of habitual action with present tense, whereas the native speaker responses show that for them, the notion of habitual action is dissociated from tense, occurring with past or present. The use of nonpast forms increases with the presence of adverbs of frequency in both activity and state verbs, but their use is greater in state verbs where there is no secondary competition from progressive forms. The use of nonpast in the environment of adverbs of frequency provides additional evidence that learners associate tenses with specific meanings which are undergeneralized compared to the target language associations. The results show that the learners have difficulty maintaining tense continuity established by the past-tense context in the environment of adverbs of frequency suggesting that learners associate the notion of habitual action, represented by adverbs of frequency, with the concept of present habitual. The learner undergeneralizations can be represented as subsets of the relevant rules of target grammar as in Figure 2. The acquisitional sequence found in this study and in the studies of uninstructed learners (of both L1 and L2) may have at least two
FIGURE 2
Distribution and Meaning of Past in L2 and NS Grammars
A. Environments of Past Form Usage
B. Meaning Associated with Past Form
potential sources. The first is that the influence of lexical aspectual class may be an acquisitional universal. A potential additional source is that the input to the learner contains what Andersen has called a distributional bias. Andersen (1990) hypothesizes that if two forms occur in the same environment, but if one seems to be more common, the learner can misperceive the meaning and distribution of a particular form that he discovers in the input (p. 58). Such could be the case with the use of simple past with event and activity verbs when a learner singles out event verbs as the best carriers of simple past. Whatever the source of the acquisitional sequence observed here for classroom language learners and elsewhere for untutored learners, there is evidence that learners follow the one-to-one principle proposed by Andersen (1984, 1990). This principle states that the emergent grammar of a learner associates one meaning with one form. Thus, the original meaning will be more limited than the final association. Andersen cites the initial association of past with accomplishment achievement verbs as an example of the one-to-one principle. In contrast to early stages of acquisition, the mature target language does not maintain a strict one-to-one relationship between form and the limited meaning of punctual or telic. Although the early stages of acquisition of past tense suggest that simple past morphology is initially
120 TESOL QUARTERLY
associated with the concept of past punctual action, native speaker use of simple past is not distributionally restricted. Its association with verbs of all lexical aspectual classes suggests a more general meaning of prior. Thus, there is pressure on the learner grammar to move toward broader meaning, as reflected by the occurrence of simple past in all lexical aspectual classes.6 There is evidence from this study, however, that learners have difficulty moving beyond the one-to-one principle with respect to past tense usage for quite some time. Reflecting on the import of acquisitional sequences to pedagogy, Andersen (1990) concludes:
Perhaps a one-form-one-meaning relation is inevitable as a first entry into a language. If so, a major goal of foreign language research should be to discover what form-meaning relations learners perceive and incorporate into their interlanguage. The assumption is that acquisition-directed language pedagogy should work within such natural tendencies. (p. 52)
The next section presents a pedagogical approach offering input to learners which is designed to help them acquire a more targetlike use of simple past tense.
Pedagogical Treatment
Using the acquisitional data to provide an assessment for instruction, we find that learners even at advanced levels of proficiency show low rates of appropriate use of simple past tense with activity and state verbs and low rates of appropriate use of past in the environment of adverbs of frequency. The past tense is a building block for other tenses both formally and conceptually (Bardovi-Harlig, 1994) and its appropriate use is pedagogically expected of intermediate and advanced learners. The low rates of appropriate use can be attributed to undergeneralizations in the learner grammar, and the acquisitional sequence indicates the needed area of instruction. Second language acquisition theory determines the means of instruction. This pattern of undergeneralization presents an interesting case for current hypotheses concerning learner grammars and the usefulness of classroom instruction. When the learner grammar forms a subset of the target grammar, learners are thought to revise their grammars on the basis of positive evidence. Positive evidence is any input which shows a learner what sentences, constructions, or combinations are possible in a language (Sharwood Smith, 1991; White, Spada, Lightbown, & Ranta, 1991). Positive evidence contrasts with negative
6
Andersen (1990) calls this the multifunctionality principle. The multifunctionality principles has two parts, associating one meaning to multiple forms and one form to multiple meanings. We are only concerned with the second case here.
121
evidence which provides information to a learner that a particular form is not allowed in the language.7 Although there have been a number of theoretical and pedagogical discussions which have dealt extensively with negative evidence, or various types of corrective feedback (White, 1991; White, 1992; White et al., 1991; Schwartz& GubalaRyzak, 1992), the important role that positive evidence could play in expanding undergeneralizations in learner grammars has not received much attention. The presentation of positive evidence is hypothesized to help learners whose grammar could be represented by the inner circles of Figure 2 to expand their grammars to incorporate the formmeaning associations in the outer circle. However, such a change requires that learners notice a difference between the input and their own production in order to revise their interlanguage rules (Schmidt, 1990, 1992). The following sections address the presentation of positive evidence and the implementation of focused noticing in the classroom. We draw on examples from an experimental unit which is under development in the Intensive English Program, Center for English Language Training, Indiana University. The unit is not intended to teach all of tense and aspect but to address the undergeneralizations in the learner grammars. We have used the unit with high beginners (Level 2) and intermediate learners (Level 4) whose scores on the cloze tasks reported in the previous section show high formal accuracy of simple past and appropriate use of past with event verbs but who still show low scores with activity and state verbs. The unit is centered on the presentation of contextualized examples of tense use in authentic language. As a source of input, we used a reading based on a narrative account from National Public Radios All Things Considered about a man working with the homeless in Washington, DC. We selected a narrative as an example text because of the rich variety of tenses to be found in a narrative. However, any genre of text could be used. The general organization of the unit follows the order of providing examples of past tense usage through the narrative, then focused noticing exercises. We repeat this cycle twice. Although the presentation of positive evidence is under the control of the teacher, noticing is entirely under the control of the learner.
Positive Evidence
The goal of the presentation of positive evidence is to give learners examples of how the target language works. In this case, our goal was
7
In a classroom, negative evidence may take the form of correction or corrective feedback but may more broadly include misunderstandings or lack of communication in classroom and other settings. (See Schachter, 1986.)
122
TESOL QUARTERLY
to provide learners with examples of past tense use with activity and state verbs, environments in which the simple past is under used by learners. Thus, the goal is to adjust for the distributional bias found in the input.8 We selected the text carefully to provide examples of past tense with adverbs of frequency (I never saw that man again; I always resented them) and examples of past tense with activity verbs (He slept; I sat there for a while . . . and talked to them) and state verbs (he was dirty; he smelled; he looked very clean). As for activity verbs, the text had examples of both simple past and past progressive (e.g., walked and was walking). Our goal in the case of activity verbs was to add the use of simple past to the learner grammar not replace the use of past progressive entirely. Once the appropriate input is identified, the next step is to help the learners notice it.
Focused Noticing
It is possible for positive evidence to occur in natural input but go unnoticed by a learner (Schmidt, 1992; Sharwood Smith, 1991; White et al., 1991). In such cases instruction can facilitate acquisition by increasing the learners awareness. Such an approach to instruction is found in input enhancement (Sharwood Smith, 1991), a refinement of grammatical consciousness raising (Sharwood Smith, 1981; Rutherford, 1987) which Sharwood Smith (1991) defined as a deliberate focus on the formal properties of language with a view to facilitating the development of L2 knowledge (p. 118). We have expanded the concept of input enhancement to include not only form in the strictest sense of formal (i.e., grammatical) accuracy but also to include formmeaning associations as evidenced in the appropriate use of tense/ aspect morphology. (See also Ellis, 1994.) Although providing positive evidence is an important and necessary step in instruction, Schmidts (1990, 1992) position holds that learners must notice the difference between their own grammars and the target grammar. To encourage learners to notice such a difference, their attention may be focused on the use of the target form-meaninguse associations through exercises. Each point in the presentation of positive evidence should have a corresponding noticing exercise. In the examples which follow, learners are encouraged to focus on the use of past with activity verbs, specifically the contrast between the simple past and the past progressive. We begin by having learners identify occurrences of each of the tense/aspect forms (Example 8),
8
A clear case of distributional bias concerns the presentation of present tense and adverbs of frequency. The low rate of past tense and the strong competition from the simple present may be a reflection of the fact that in many textbooks there are no examples of adverbs of frequency with tenses other than simple present.
123
and then an occurrence of the two forms (Example 9). (An excerpt from the narrative is provided in Appendix B).
8. Find a sentence with only one verb in the simple past. Write it on the line below. 9. Find a sentence with one simple past verb and one past progressive verb. Write it on the lines below.
We also focus on activity verbs in the text which occur in both past progressive and simple past forms. Students are given the past progressive forms and then asked to find the corresponding simple past forms.
10. Kerwin [the narrator] used the following verbs in the past progressive. He also used them in the simple past. Find the simple past forms, write them beside the past progressive and write the line where you found them. Look at the way the past progressive and the simple past are used. Can you tell a difference in meaning? Past Progressive Line Simple Past was walking (lines 12) was doing (lines 5556)
In this exercise the learners attention is directed to the fact that there is not a one-to-one relation between a lexical aspectual class and tense/aspect morphology by giving the learners contextualized examples of both tense/aspect forms. Following this exercise, class discussion centers on the differences in meaning of the verbs in past and past progressive as used in the text at hand. Additional exercises would follow emphasizing the use of past with state verbs, adverbs of frequency, and combinations thereof.
Beyond Positive Evidence and Focused Noticing

In addition to using authentic text to present positive evidence and using focused noticing exercises, we also use a range of production tasks to provide contextualized practice. At each stage in the unit, the learners complete controlled and free production activities. The unit begins with a written summary of the reading. The lessons which focus on the use of past with adverbs of frequency include a topically appropriate essay: When you were in high school what did you usually do? Were there some things that other teenagers always did that you rarely or never did? The topic provides a context in which past tense verbs can be used with adverbs of frequency. Next, instruction focuses on the use of simple past with activity verbs. Using the original reading passage, the learners attention is directed to activity verbs used in both the simple past and past progres124 TESOL QUARTERLY
sive, as discussed above. A cloze passage that is a condensed version of the original text provides contextualized practice in the use of the simple past and the past progressive. A final writing assignment is given which offers practice in reporting situations as either completed (requiring past) or in progress at the time of another event (suggesting past progressive). This was set up through the instructions (Write a paragraph. Use the sentence below to start it. Try to tell both what was happening around you and what you did.) and the opening line taken from the first sentence of the authentic narrative (I was walking down the street one day last week.). It is important to note that input enhancement is not a method of language teaching in and of itself. Input enhancement may vary along two dimensions: explicitness and elaboration (Sharwood Smith, 1981, 1991). Sharwood Smith represents this as four possibilities (more or less explicit combining with more or less elaboration). The unit we developed was less explicit, providing no formal rules, but more elaborate, providing contextualized examples and repeated opportunities to notice over four days of instruction. (Compare White, 1992, and White & Trahey, 1993, for a description of an input flood.) It should be noted, however, that we took advantage of the fact that all learners in our program are exposed to the names for tenses both from instruction and materials and that we used grammatical labels in the noticing exercises. Thus, input enhancement can be adapted for use in a variety of instructional methods. The invariant components to input enhancement for our approach to the past tense are the presentation of positive evidence and the focused noticing. Two other advantages of using positive evidence should be noted: longevity and efficiency. We expect that the effects of positive evidence to be long lasting. Once learners notice the difference between their grammars and the target grammar through instruction, unmodified input in the form of written texts of various types, conversation, and other aural input such as news broadcasts provide support that in English the simple past is distributed across lexical aspectual categories. As a result, we expect the gains in the use of past made through instruction to be retained after the instruction ends.9 In fact, that is the pattern reported by Harley (1989) on written retention tests (composition and cloze) administered to a group of sixth-grade French immersion students 3 months after they had received experimental instruction through a series of functionally based lessons in the distribution of the French tense/aspect system. A similar pattern has been
9
Preliminary tests on this unit with small numbers of students support this claim. Of the five classes which we have tested, all improved immediately after instruction and, more importantly, maintained their higher scores or showed continued improvement when tested one month after instruction (see Bardovi-Harlig & Reynolds, 1994).
125
found in the acquisition of questions where learners of question forms continued to develop after the specific instructional period was over (Spada & Lightbown, 1993; White et al., 1991). We hypothesize that maintenance of improved scores or even additional improvement is a characteristic of the presentation of positive evidence which may distinguish it from negative evidence. Because positive evidence, by virtue of the fact that it is found in all naturally occurring language input, is available to the learner after instruction (and because negative evidence is thought to be unavailable), learners maintain high levels of appropriate use, even improving in some cases. Thus, the difference in presenting negative and positive evidence in instruction is that the presentation of negative evidence is an isolated occurrence limited to the classroom. In contrast, the presentation of positive evidence through authentic language samples and focused noticing is reinforced by the ambient language inside and outside the classroom. Once a learner is assisted in noticing a particular characteristic of the language, natural input supports the instruction. Such a case may be found in the acquisition of questions mentioned earlier (Spada & Lightbown, 1993; White et al., 1991). Questions are high frequency items in classrooms, and the learners seemed to be able to take advantage of their presence in classroom talk as positive evidence (even when the instructional focus on questions had ceased) leading to continued 10 development of question forms in learner language. Thus, we expect positive evidence to have long-term effects. Finally, the use of positive evidence is pedagogically efficient. If we were to approach instruction of the past from the point of view of negative evidence, or error correction, we would find that learners supply a variety of alternatives to the simple past with activity and state verbs and with adverbs of frequency. Thus, such correction, whether explicit or not, would have to address a number of verb forms. However, the use of positive evidence is pedagogically efficient because all learners have the same target form and thus the positive evidence is relevant to all the learners in a class regardless of their individual hypotheses about tense marking.
CONCLUSION
The results of the study reported here demonstrate that lexical aspect plays an important role in the use of past tense by instructed adult learners of English as a second language. The use of past tense by classroom language learners shows similar patterns of distribution
10
We thank an anonymous reviewer for bringing the case of questions to our attention.
126
TESOL QUARTERLY
to those of untutored learners of English. Moreover, on the basis of available data, comparison to learners of other languages such as Spanish suggests that the influence of lexical aspect may be common to language acquisition in general, part of what VanPatten (1990) calls the core of SLA (p. 25). Further research is needed to determine how the influence of lexical aspect is realized in different target languages. This article also demonstrates the importance of observing acquisitional sequences for the purposes of instruction. Through research, we identified areas of difficulty in the acquisition of the tense/aspect system. We found that the teaching of past tense for achievement and accomplishment verbs is much less necessary, whereas the teaching of simple past with activity verbs and with adverbs of frequency is clearly warranted, and indications are that it is worthwhile. Based on SLA theories, we have suggested that the presentation of positive evidence will help learners broaden the undergeneralizations in their grammars. We argue that input enhancement which includes focused noticing as well as positive evidence provides learners with an awareness which helps input to become intake even outside the classroom.
ACKNOWLEDGMENTS
We thank Harry L. Gradman, Director of the Center for English Language Training, and Susan Greer and Marlin Howard, Directors of the Intensive English Program, for their assistance in coordinating the elicitation tasks and their continued support of the project. We thank Dr. Camillia Majd Jabbari for her statistical analysis of the data. We would also like to thank the anonymous reviewers for their helpful suggestions. An earlier version of this paper was presented in 1993 at the 27th Annual TESOL Conference in Atlanta, GA. This study was supported by a grant to the first author from the National Science Foundation (BNS 8919616).
THE AUTHORS
Kathleen Bardovi-Harlig is Associate Professor of TESOL and Applied Linguistics at Indiana University. Her research and teaching interests include SLA, interlanguage pragmatic, and teacher education. Her work on the acquisition of tense and aspect has appeared in Language Learning, Applied Psycholingutitics, and edited collections. Dudley W. Reynolds, Associate Instructor at the Center for English Language Training, is a PhD candidate specializing in SLA and discourse pragmatic at Indiana University. His dissertation investigates repetition as a pragmatic device in the writing of ESL learners. He has published in World Englishes and the ELT Journal. THE ACQUISITION OF TENSE AND ASPECT 127
REFERENCES
Anderson, R. W. (1984). The one to one principle of interlanguage construction. Language Learning, 34, 7795. Andersen, R. W. (1989). La adquisicon de la morfologa verbal [The acquisition of verb morphology]. Lingstica, 1, 90142. Andersen, R. W. (1990). Models, processes, principles and strategies: Second language acquisition inside and outside the classroom. In B. VanPatten & J. F. Lee (Eds.), Second language acquisitionForeign language learning (pp. 4578). Clevedon, England: Multilingual Matters. Andersen, R. W. (1991). Developmental sequences: The emergence of aspect marking in second language acquisition. In T. Huebner & C. A. Ferguson (Eds.), Second language acquisition and linguistic theories (pp. 305324). Amsterdam, Netherlands: John Benjamins. Andersen, R. W., & Shirai, Y. (1994). Discourse motivations for some cognitive acquisition principles. Studies in Second Language Acquisition, 16, 133156. Antinucci, F., & Miller, R. (1976). How children talk about what happened. Journal of Child Language, 3, 167189. Bardovi-Harlig, K. (1992). The relationship of form and meaning: A cross-sectional study of tense and aspect in the interlanguage of learners of English as a second language. Applied Psycholinguistics, 13, 2 5 3 2 7 8 . Bardovi-Harlig, K. (1994). Reverse-order reports and the acquisition of tense: Beyond the principle of chronological order. Language Learning, 44, 243282. Bardovi-Harlig, K., & Bergstrom, A. (in press). The acquisition of tense and aspect in SLA and FLL: A study of learner narratives in English (SL) and French (FL). Canadian Modern Language Review. Bardovi-Harlig, K., & Reynolds, D. W. (1994). Improving tense usage through input enhancement. Unpublished Indiana University manuscript. Binnick, R. I. (1991). Time and the verb: A guide to tense and aspect. New York: Oxford University Press. Bloom, L., Lifter, K., & Hafitz, J. (1980). Semantics of verbs and development of verb inflection in child language. Language, 56, 386412. Bronckart, J. P., & Sinclair, H. (1973). Time, tense and aspect. Cognition, 2, 107 130. Comrie, B. (1976). Aspect. Cambridge: Cambridge University Press. Comrie, B. (1985). Tense. Cambridge: Cambridge University Press. Dowty, D. (1979). Word meaning and montague grammar. Dordrecht, Netherlands: Reidel. Ellis, R. (1994, March). Interpretation tasks for grammar teaching. Paper presented at the 28th Annual TESOL Conference, Baltimore, MD. Harley, B. (1989). Functional grammar in French Immersion: A classroom experiment. Applied Linguistics, 10, 331359. Klein, W. (1986). Second language acquisition. (rev. ed.). (Bohuslaw Jankowski, Trans.) Cambridge: Cambridge University Press. (Original work published 1984) Mourelatos, A. P. D. (1981). Events, processes, and states. In P. Tedeschi & A. Zaenen (Eds.), Tense and aspect. (Syntax and Semantics 14) (pp. 191212). New York: Academic Press. Quirk, R., Greenbaum, S., Leech, G., & Svartvik, J. (1985). A comprehensive grammar of the English language. London: Longman. Robison, R. (1990). The primacy of aspect: Aspectual marking in English interlanguage. Studies in Second Language Acquisition, 12, 315330. 128 TESOL QUARTERLY
Rutherford, W. (1987). Second language grammar: Learning and teaching. New York: Longman. Schachter, J. (1986). Three approaches to the study of input. Language Learning, 36, 211225. Schmidt, R. (1990). Consciousness, learning and interlanguage pragmatic. University of Hawaii Working Papers in Linguistics, 9, 213243. Schmidt, R. (1992). Awareness and second language acquisition. Annual Review of Applied Linguistics, 13, 206-226. Schumann, J. (1987). The expression of temporality in basilang speech. Studies in Second Language Acquisition, 9, 2141. Schwartz, B., & Gubala-Ryzak, M. (1992). Learnability and grammar reorganization in L2A: Against negative evidence causing the unlearning of verb movement. Second Language Research, 8, 138. Sharwood Smith, M. (1981). Consciousness raising and the second language learner. Applied Linguistics, 2, 159168. Sharwood Smith, M. (1991). Speaking to many minds: On the relevance of different types of language information for the L2 learner. Second Language Research, 7, 118132. Shirai,Y. (1991 ). Primacy of aspect in language acquisition: Simplified input and prototype. Unpublished doctoral dissertation, University of California, Los Angeles. Smith, C. S. (1983). A theory of aspectual class. Language, 59, 479501. Spada, N., & Lightbown, P. M. (1993). Instruction and the development of questions in L2 classrooms. Studies in Second Language Acquisition, 15, 205224. Stephany, U. (1981). Verbal grammar in modern Greek early child language. In P. S. Dale & D. Ingram (Eds.), Child language: An international perspective (pp. 4557). Baltimore, MD: University Park Press. Trvise, A. (1987). Toward an analysis of the (inter) language activity of referring to time in narratives. In C. W. Pfaff (Ed.), First and second language acquisition processes (pp. 225251). Cambridge, MA: Newbury House. VanPatten, B. (1990). Theory and research in second language acquisition and foreign language learning: On producers and consumers. In B. VanPatten & J. F. Lee (Eds.), Second language acquisition/Foreign language learning (pp. 1726). Clevedon, England: Multilingual Matters. Vendler, Z. (1967). Verbs and times. In Z. Vendler, Linguistics and philosophy (pp. 97121). Ithaca, NY: Cornell University Press. (Reprinted from Philosophical Review, 1957, Vol. 66, 143160) Vronique, D. (1987). Reference to past events and actions in narratives in L2: Insights from North African Learners French. In C. W. Pfaff (Ed.), First and second language acquisition processes (pp. 25227 2). Cambridge, MA: Newbury House. von Stutterheim, C., & Klein, W. (1987). A concept-oriented approach to second language studies. In C. W. Pfaff (Ed.), First and second language acquisition processes (pp. 191205). Rowley, MA: Newbury House. White, L. (1991). Adverb placement in second language acquisition: Some effects of positive and negative evidence in the classroom. Second Language Research, 7, 133161. White, L. (1992). On triggering data in L2 acquisition: A reply to Schwartz and Gubala-Ryzak. Second Language Research, 8, 120137. White, L., Spada, N., Lightbown, P. M., & Ranta, L. (1991). Input enhancement and L2 question formation. Applied Linguistics, 12, 416432. White, L., & Trahey, M. (1993). Positive evidence and preemption in the second language classroom. Studies in Second Language Acquisition, 15, 181204.
129
APPENDIX A
Verbs Used in the Cloze Passages by Lexical Aspectual Class Achievement Verbs
arrive, break, die, discover, drop, explode, fall out, find, happen, kill, lose, start, take off (as in the plane took off), turn (something) off
Accomplishment Verbs
buy, change (apartments), change (the story into a funny one), eat a meal, finish all of the grammar homework, give, go to the class, marry, move (from one apartment to another), rent a video, write two papers
Activity Verbs
dance, eat , go to the same school (as in attend), live (2), ride bicycles, sing, snow, swim, tell stories, work, worry (all weekend)
Stative Verbs
be dangerous, belong, enjoy (2), know, need (2), own, seem, want
Activity Verbs with Frequency Adverbs

(always) study a lot, play soccer (everyday), rain (everyday), (never) clean, (never) cook, (often) cook, (seldom) go out, (sometimes) work, (usually) eat
State Verbs with Frequency Adverbs

(always) look bad, (everyday) smell delicious, (never) want, (often) feel sad, (seldom) taste good, (usually) seem happy
APPENDIX B
Excerpt from the Narrative Text
1 Kerwin: I was walking one night. It was bitterly cold, around Christmas 2 1978, and I was walking. I think I was walking down to the river just to 3 clear my head or to go for a walk. I love to walk. There was a man on 4 the heating vent across the street from the State Department at 21st and 5 E which was only a block from my apartment, and he called out to me. He 6 said he wanted a buck to buy something to eat. I was very irritated with 7 him for calling out after me. I didnt want to be bothered and I didnt 8 believe him either. I thought, well he just wants to get something to 9 drink; and I thought to myself, well Ill fix him. Ill go and get him 10 something to eat and that way hell be frustrated and angry and didnt 11 get what he wanted but at least Ill give him what he asked for. So I 12 went up to my apartment, got him a bowl of soup, got him a sandwich and 13 a cup of tea, and brought it down. I set it down and walked away. I 14 continued my walk, didnt say a word to him, and didnt acknowledge his 15 thanks. I never saw that man again, but I went home that night and I 16 just thought: well, you know that made me feel pretty good. Thats the 17 least I can do. Thats all it takes to make me feel good and to think 18 well you know here Im helping the human being. I can do that. I mean 19 what effort did that take. 20 So I went home, and the next night I went out again with the same 21 type of meal (I think a little bit more but the same type of meal). I 22 just set it down on the heating vent where other people were, and I 23 walked away. They thought I was a little bit crazy bringing out this 24 food and setting it down, but I did. I simply went back to my apartment 25 and that was the end of it. But I kept doing that, and I kept doing it 130 TESOL QUARTERLY
26 night after night after night and eventually got to know some of these 27 people because I was consistent. I went down there. . . . 48 Everything was fine until about three months after Id begun. A man 49 asked to come up to my apartment and to shower and shave. I said, 50 Absolutely not. I said, I live in a dorm. I said, You know, its 51 its graduate housing. I said, We are not allowed to have people in 52 just strangers off the street. He was dirty. He smelled, and he was a 53 little bit inebriated. I said, No. I said, I can bring the food out 54 but I cant have any people up there. I didnt want to, and I was 55 embarrassed if I had been seen with him. Nobody was aware of what I was 56 doing. Nobody in the dorm or anybody else was aware that I was taking 57 this food down, and thats the way I wanted it. So that was fine, but he 58 persisted for three days after that. Finally I said, All right, Glenn. 59 I said, Come up to the apartment. But I said, Youve got to leave as 60 soon as you have your shower and shave. So it was fine. I took him up, 61 and he went into the bathroom. I went into the kitchen. I was washing 62 some dishes, and I came back into the living room about twenty minutes 63 later. There he was in the chair fast asleep. He had come out and he 64 had gone into the chair. Right awaybecause he was so tiredhe had 65 gone to sleep. I had given him some clean clothes and he had put those 66 on. So he looked fine. He looked a lot better than he did on the grates. 67 He slept.
131
Closing the Gap Between Learning and Instruction

DAVID NUNAN
University of Hong Kong
In this paper, I explore the mismatch between the pedagogical intentions and plans of the educational institution, curriculum, teacher, and textbook, and the outcomes as realized through the skills and knowledge that learners take away from instructional encounters. Although there will never be a one-to-one relationship between teaching and learning, there are ways in which teachers and learners and teaching and learning can be brought closer together. In this article, I look at ways of closing the gap in relation to experiential content, learning process, and language content. The theme holding these three disparate domains together is that of learner centredness, and it is this concept which I take as my point of departure.
efore giving my interpretation of the concept of learner centredness, I should like to explain how I became interested in the subject. Many years ago I began to realize that there was a major gap between what I was focusing on as a teacher and what my learners were taking away from the pedagogical opportunities I was providing. I therefore became obsessed with a question so admirably framed by Allwright (1984): Why dont learners learn what teachers teach? (p. 3). This article offers some clues as to where the answer might be found. I also provide practical examples of ways in which the ideas might be applied to teaching situations. However, I should also add that I am not claiming that all of these ideas are necessarily feasible and relevant to all contexts and situations. The context in which any teaching takes place will have a major influence on what is both feasible and desirable. However, the belief that it would never work here, is often used as an excuse for inaction.
THE CONCEPT OF LEARNER CENTREDNESS

The first concept I should like to look at is that of learner centredness. As I have explained elsewhere (see, e.g., Nunan, 1988), a learner133
centred curriculum will contain similar components to those contained in traditional curricula. However, the key difference is that in a learnercentred curriculum, key decisions about what will be taught, how it will be taught, when it will be taught, and how it will be assessed will be made with reference to the learner. Information about learners, and, where feasible, from learners, will be used to answer the key questions of what, how, when, and how well. However, it is a mistake to assume that learners come into the language classroom with a natural ability to make choices about what and how to learn. I believe that there are relatively few learners who are naturally endowed with the ability to make informed choices about what to learn, how to learn it, and when to learn. It is at this point that we need to turn from the concept of learner centredness, to a closely related concept of learning centredness. A learning-centred classroom carries learners toward the ability to make critical pedagogical decisions by systematically training them in the skills they need to make such decisions. Such a classroom is constituted with complementary aims. Whereas one set of aims focuses on language content, the other focuses on the learning process. Learners are therefore systematically educated in the skills and knowledge they will need in order to make informed choices about what they want to learn and how they want to learn. Rather than assuming that the learner comes to the learning arrangement with critical learning skills, the sensitive teacher accepts that many learners will only begin to develop such skills in the course of instruction. Learner centredness is therefore not an all-or-nothing concept; it is a relative matter. It is also not the case that a learner-centred classroom is one in which the teacher hands over power, responsibility, and control to the students from Day 1. I have found that it is usually well into a course before learners are in a position to make informed choices about what they want to learn and how they want to learn, and it is not uncommon that learners are in such a position only at the end of the course. That said, I would advocate the development of curricula and materials which encourage learners to move toward the fully autonomous end of the pedagogical continuum.
THE EXPERIENTIAL CONTENT DOMAIN

In this section, I should like to look briefly at reasons for the gap between teaching and learning in the experiential content domain. This will provide the basis for a more detailed discussion on what we might do about it. I should like to argue that the principal reason for the mismatch
134 TESOL QUARTERLY
between teachers and learners, which gives rise to a disparity between what is taught and what is learned, is that there is a mismatch between the pedagogical agenda of the teacher and that of the learner. While the teacher is busily teaching one thing, the learner is very often focusing on something else. This mismatch has been noted by numerous researchers with an interest in the classroom (see, e.g., Allwright, 1987; Allwright & Bailey, 1991). More recently, Slimani (1992) has sought to determine what individual learners claim to have learned from interactive classroom events (this claimed learning she terms uptake). She further examines what happens in a lesson that can account for this uptake. Her learners were a group of Algerian learners of English as a foreign language who were preparing to undertake engineering studies in English. Slimani found that topics initiated in the classroom by the learners were much more likely to be nominated as having been learned than those nominated by the teacher. In other words, when learners had an opportunity to contribute to the content of the lesson, that was the content which learners would claim to have learned. Slimani (1992) reports that:
about 77.45 percent of the topicalisation was effected by the teacher. This is not particularly surprising in view of the fact that the discourse was unidirectionally controlled by the teacher. . . . What appears to be strikingly interesting though is that a further analysis of the effect of the teachers versus the learners scarce opportunities . . . for topicalisation showed that the latter offered much higher chances for items to be uptaken. Learners benefited much more from their peers rare instance of topicalisation than from the teachers . . . . topics initiated by learners attracted more claims from the learners than the ones initiated by the teacher. (p. 211)
Block (1994, in press) provides further insights into the perceptual gap between teachers and learners. Working in an EFL situation, Block used an oral diary technique, in which a teacher and six of her learners provided daily audiotaped accounts of the lesson. In particular, they were asked for their point of view on the activities that stood out most, the purpose of the activities, what the student learned, what the teacher did to facilitate the learning process, and other events from the class worth mentioning. The oral diary accounts were supplemented by classroom observations and interviews. Block discovered that whereas certain perceptions were shared by all informants, the accounts differed in certain significant ways. Block suggests that each learner comes to class with a hobby horse, that is, a particular pedagogical preoccupation that colors his/her perceptions as to what is going on, why, and what value it has. One learner, for example, was preoccupied with the utilization of class time and evaluated all tasks in terms of whether they utilized class time effecLEARNING AND INSTRUCTION 135
tively. Block also provides evidence of a gap in the perception of the teacher and the students. The major gap appeared to revolve around perceptions (and misperceptions) of the pedagogical purpose of activities. Students often either had misperceptions about the rationale for the task or no idea at all why they were being asked to do particular tasks. The teacher, on the other hand, was able to provide a rationale, although this was not spelled out to the students. I believe that the gap between teaching and learning in the experiential content domain can be narrowed by moving toward the implementation of a learner-centred approach to pedagogy. As I indicated earlier, learner centredness is not an all-or-nothing concept. I would like to suggest that there are degrees of learner centredness, that there is a continuum, from relatively modest to rather radical levels of implementation. In the rest of this section, I discuss some of the practical things we can do in the classroom to realize the concept of learner centredness. How far along the continuum one moves will depend on the pedagogical context in which one is working. Where feasible, I illustrate, with practical examples, some of the ways in which these ideas can be realized in the classroom tasks and materials. I should stress, however, that these examples have been inserted for illustrative purposes only, and it is not possible to provide exhaustive illustrations. In the experiential content domain, I would suggest that the first step along the path to learner centredness would be to make learners aware of the goals and the content of the curriculum, learning program, or pedagogical materials. This may not seem particularly radical. However, in my recent study of classroom interaction, there was only one instance in which the teacher began a lesson by laying out the pedagogical terrain to be covered (see Nunan, in press). Failure to spell out lesson objectives was also noted by Block (1994, in press). Making salient the goals of a lesson or unit of work is relatively easy to achieve regardless of whether one is teaching to a state-mandated curriculum with materials supplied, whether one is teaching to an examination, or whether one is teaching in a foreign language situation in which students may be required to undertake another language whether they want to or not. There is evidence, in fact, that interest and motivation are enhanced when the purpose and rationale of instruction is made explicit to the learners (see, e.g., Brindley, 1984). The following is an extract from an intensive EFL course in Australia for mainly preintermediate-level Japanese adults in which the teacher does set the agenda for the learners. The teacher is using a mandated textbook in which the goals and objectives are implicit, and yet she is able to make the goals of the lesson explicit to the learners. She does so by actively involving them in the process rather than simply informing them.
136 TESOL QUARTERLY
T: Today were going to practice talking about likes and dislikes, and were going to talk about music and movies and stuff. OK? OK Kenji? Now, I want you to open your books at page 22, thats where the unit starts, and [inaudible comment from student] . . . Whats that? . . . . Yeah, thats right. Now, I want you to look quickly through the unit and find one example, one example of someone saying they like something, and one example of someone saying they dont like something? OK? One example of each. And Im going to put them here on the board.
If one is producing ones own materials or adapting those written by others, it is relatively easy to make the goals explicit. Once again, learners can be actively involved, as the following example shows:
Unit Goals In this unit you will: 1. Make comparisons: The city is busier than the countvy. 2. Ask for and give advice: Ive missed the bus. What should I do?
The unit could be completed by asking the learners to carry out a self-checking exercise such as the following. Although this has been extracted from a commercial source, it is the sort of exercise that teachers can readily create.
Review the language skills you practiced in this unit. Check [ 4 ] your answers. CAN YOU? Make comparisons? [ ] yes [ ] a little [ ] not yet Find or give an example: Ask for and give advice? [ ] yes [ ] a little [ ] not yet Find or give an example: (Nunan, 1994, p. 108)
I believe that the idea of making the pedagogical agenda explicit to the learners is something which can be done to various degrees with all but the youngest of learners and those at the very beginning proficiency levels. The different levels of learner involvement in the experiential content domain are summarized in Figure 1. At the second level, learners themselves would be involved in selecting goals and content. Whether it is possible or desirable to implement this level and succeeding levels of the continuum will very much depend on the context and situation in which one is teaching, something upon which I shall have more to say later. There are several welldocumented accounts in which learners have been involved in making choices about what they will learn. Dam and Gabrielsen (1988) found
LEARNING AND INSTRUCTION 137
FIGURE I
Learner Centredness: Levels of Implementation in the Experiential Content Domain
Level 1 2 3 4 5
Learner Action Awareness Involvement Intervention Creation Transcendence
Gloss Learners are made aware of the pedagogical goals and content of the program. Learners are involved in selecting their own goals and objectives from a range of alternatives on offer. Learners are involved in modifying and adapting the goals and content of the learning program. Learners create their own goals and objectives. Learners go beyond the classroom and make links between the content of the classroom with the world beyond the classroom.
that even relatively young learners were capable of making decisions about the content and processes of their own learning. Learners, regardless of their aptitude or ability, were capable of a positive and productive involvement in selecting their own content and learning procedures. Furthermore, learners were also positive in accepting responsibility for their own learning. Further along the learner-centred continuum, we would see learners modifying and adapting goals and content. The next step would see learners creating their own goals and content. An interesting and practical way of involving learners at this level is reported in Parkinson and OSullivan (1990), who were working with high intermediate-level adult learners in an ESL context. They report on the notion of the Action Meeting as a way of involving learners in modifying course content.
A mechanism was needed for course management: as the guiding and motivating force behind the course, it would have to be able to deal with individual concerns and negotiate potential conflicts of interest, need, and temperament. It would also have to satisfy the individual while not threatening the groups raison dtre. As foreshadowed in the orientation phase, the group would now experiment with a mechanism suggested by the teachers, namely a series of Action Meetings . . . . [These] would provide an opportunity for individuals to participate (interpersonally and interculturally) in an English-medium meeting, negotiating meaning and authentic content. They would also be a means of facilitating group cohesion and motivation and would be a primary mechanism for ongoing program evaluation by the participants. (pp. 119120)
The final level is one in which learners transcend the classroom and link content to the world beyond the classroom. Some years ago, I
138 TESOL QUARTERLY
investigated the notion of the good foreign language learner. What I wanted to find out was whether learners who had developed high levels of competence in a foreign language had certain learning experiences in common. Although I found quite a variety of learning experiences at the level of classroom strategies, virtually all learners demonstrated an ability to relate the content of the classroom to the world beyond the classroom (Nunan, 1989, 1991). Furthermore, they all identified this ability as the critical ingredient in their success as language learners. This idea of the importance of consciously developed activation of the language beyond the classroom is also reported in a L2 context by Schmidt and Frota (1985). In practical terms, it is obviously much easier to encourage learners to activate their language outside of the classroom in L2 contexts (for practical examples, see Aiken & Pearce, 1994) and situations where English is widely spoken within the community (such as in Hong Kong). However, even in foreign language contexts, it is possible to find ways of practicing the target language outside the classroom. In the good foreign language learner study mentioned above (Nunan, 1989, 1991), in which virtually all 44 informants said that success was partly due to activating language outside the classroom, learners exploited a range of resources. These included English language newspapers, radio and television, international hotels and airline offices, multinational companies, and international airports. The following classroom extract illustrates the way in which one teacher encouraged students to think about activating their language outside of the classroom. The class was a mixed-proficiency group of adult ESL learners in Australia.
[The students are sitting in small groups of two to four as the teacher addresses them.] T: Well students, as you know, this morning were going to be looking at ways that we can help learners improve their English without a teacher, without, um, a class to come to. Whatve we got all around us that can help us? Well the first thing that were going to be looking at are these things. [She bends down and picks up a plastic shopping bag.] Now in the bagIve got a bag full of mystery objects in heredifferent things, but they all have one thing in common. We can use them to help improve our language. Now this is going to be lucky dip type activity. Have you ever done a lucky dip? Ss: Yes, yes. T: Yes. Where you put your hand in and you take one thing out. Ill do it the first time. Put my hand in and Ill just bring . . . something out. [She pulls out a mirror.] Oh, a mirror. Now how can this help us improve our language you got any ideas? Irene? LEARNING AND INSTRUCTION 139
S: We can help, er, our voc . . . vocabulary. T: Vocabularys one thing, yes. How? S: We can look, er, how we pronounce the words. (Mmm) We can look in the mirror and see how our mouth moves. T: Good. Yes, we can see how our mouth movesby looking at our reflection in the mirror. For example, the sound th. Can you all say th? Ss: No. [Laughter] [The teacher distributes the rest of the objects in the bag and the students, working in groups, spend 10 minutes discussing the ways in which the different objects they have chosen can be used for practicing English outside the class. The teacher then calls the activity to a bait.] (Nunan, 1991, p. 182)
THE LEARNING PROCESS DOMAIN

I suggested earlier that one answer to the question of why learners do not learn what teachers teach is that they come into the classroom with different mind sets, different points of focus, or, as I put it above, different agendas. Turning from the experiential content to the learning process domain, I should like to suggest that a partial answer to the question can be found in a mismatch at the level of learning process. There is, in fact, evidence to support this notion. Some years ago, I carried out a comparative study into the learning preferences of teachers and learners in the Australian Adult Migrant Education Service program (Nunan, 1987). When I compared the preferences of learners and teachers in relation to selected learning tasks and activities, I found some stark contrasts and dramatic mismatches. The results of this study are summarized in Figure 2 below. The figure shows that there are mismatches between teachers and learners on all but one of the items (students and teachers agreed that conversation practice was a very high priority). In all other cases, there were mismatches between the teaching preferences of the teachers and the learning preferences of the students. For example, students gave a low rating for pair work, whereas teachers gave this item a very high rating. The same was the case with student self-discovery of errors. Now, I am not suggesting that student views should be acceded to in all cases. However, I would argue that at the very least, teachers should find out what their students think and feel about what and how they want to learn. Willing (1988) carried out a large scale study into the learning styles and learning strategy preferences of adult (17- to 78-Year-old) immigrant learners of English as a second language in Australia. With 517 learners, Willing had a substantial database. Using a questionnaire and interview, Willing investigated possible learning style differences which
140 TESOL QUARTERLY
FIGURE 2
Teacher/Student Mismatches in the Learning Process Domain
Activity Explanation to class Conversation practice Error correction Vocabulary development Using cassettes Student self-discovery of errors Using pictures, film, video Pair work Language games
Student Very high Very high Very high Very high Low Low Low Low Very low
Teacher High Very high Low High Medium high Very high Low medium Very high Low
could be attributed to a range of biographical variables such as learners ethnic backgrounds, ages, levels of education, time in the target country, and speaking proficiency levels. The study yielded certain surprising findings. In the first instance, learners did have views on the learning process and were capable of articulating these. Perhaps the most surprising finding was that none of the biographical variables correlated significantly with any of the learning preferences:
None of the learning differences as related to personal variables were of a magnitude to permit a blanket generalization about the learning preferences of a particular biographical sub-group. Thus, any statement to the effect that Chinese are X or South Americans prefer Y, or Younger learners like Z, or High-school graduates prefer Q is certain to be inaccurate. The most important single finding of the study was that for any given learning issue, the typical spectrum of opinions on that issue were represented, in virtually the same ratios, within any biographical subgroup. (Willing, 1988, pp. 150-151)
What can be done about the gap between instruction and learning in the learning process domain? A similar process, I believe, to the one suggested in the experiential content domain. A continuum in the learning process domain, similar to that which has been proposed for the experiential content domain, can help lead learners in the direction of autonomy, and equip them with process skills for negotiating the curriculum. (See Figure 3.) I would like to suggest that the first step in the direction of the process is to encourage learners to identify the strategy implications of pedagogical tasks. Underlying this first step is the fact that everything we do in the classroom involves a learning strategy. This is so regardless of whether we are talking about communicative tasks such as role plays, selective listening, or debates, or more mechanical exerLEARNING AND INSTRUCTION 141
FIGURE 3
Learner Centredness: Levels of Implementation in the Learning Process Domain
Level 1 2 3 4 5
Learner Action Awareness Involvement Intervention Creation Transcendence
Gloss Learners identify strategy implications of pedagogical tasks and identify their own preferred learning styles/strategies. Learners make choices among a range of options. Learners modify/adapt tasks. Learners create their own tasks. Learners become teachers and researchers.
cises such as pronunciation drills, vocabulary memorization, or cloze exercises. Again, I would stress, that although, in certain contexts it may not be feasible to travel very far along the continuum, it is possible to take the first step with most learners in most contexts. A reasonable first step, then, would be to raise learner awareness of the strategies underlying the particular task in question. This is something that all teachers can do, regardless of whether they are working with a mandated curriculum and materials, or whether are relatively free to decide what to teach and how to teach it. This is illustrated in the following classroom extract. The students are a high beginning-level group of young adult Japanese EFL learners.
T: One of the things, er, we practice in this course . . . is . . . or some of the things we practice are learning strategies. And one of the learning strategies that will help you learn new words is the learning strategy of classifying. Do you know what classifying means? Ss: No no T: Have you heard this word before? Ss: No T: Classifying means putting things that are similar together in groups. OK? So if I said, er, I want all of the girls to go down to that corner of the room, and all the boys to go into this corner of the room, I would be classifying the class according to their sex or their gender. What Id like you to do now in Task 5 is to classify some of the words from the list in Task 4. OK? (In the preceding task, students had read a postcard and circled the words that describe people. They were then given a three column table with the headings: color, age, and size.)
The next step in the development of a learner-centred classroom would be to train learners to identify their own preferred learning styles and strategies. Detailed guidance on how this might be achieved are beginning to appear in the literature. Excellent starting points for
142 TESOL QUARTERLY
those who are interested are provided by Ellis and Sinclair (1989) and Willing (1989). The following example was adapted from an L2 situation for use in a foreign language context. It has been used successfully with learners in many different pedagogical situations (see Figure 4). FIGURE 4
Learning Style Tasks
Task 1 Close your book. Your partner is going to find out your learning style. Answer your partners questions. Task 2 a. Now find out your partners learning style. If he/she agrees with the statement, put a check mark [ ] in the box. HOW DO YOU LIKE TO LEARN? Type 1: I like to learn by watching and listening to native speakers. I like to learn by talking to friends in English. At home, I like to learn by watching TV / videos in English. I like to learn by using English out of class. I like to learn English words by hearing them. In class, I like to learn by conversations. TOTAL: Type 2: I like the teacher to explain everything to us. I want to write everything in my notebook. I like to have my own textbook. In class, I like to learn by reading. I like to study grammar. I like to learn English words by seeing them. TOTAL: Type 3: In class, I like to learn by games. In class, I like to learn by looking at pictures, films and video. I like to learn English be talking in pairs. At home, I like to learn by using cassettes. In class, I like to listen to and use cassettes. I like to go out with the class and practice English. TOTAL: Type 4: I like to study grammar. At home, 1 like to learn by studying English books. I like to study English by myself (alone). I like the teacher to let me find my mistakes. I like the teacher to give us problems to work on. At home, I like to learn by reading newspapers. TOTAL: b. Now add up the number of check marks for each section, and put the number in the Total box. The highest total shows what kind of learner your partner is. Source: Adapted from Willing, K. (1989). Teaching how to learn. Sydney, Australia: National Centre for English Language Training and Research. LEARNING AND INSTRUCTION 143
At the next level, learners would be involved in making choices among a range of options. The notion that learners are capable of making choices has been questioned by some commentators. It has also been suggested that the notion of choice is a Western one, which is less familiar to the non-Western psyche. This has been contested by several researchers, who have data, rather than opinion, bias, or experience to draw on. Widdows and Voller (1991), for example, investigated the ability of Japanese university students to make choices regarding learning preferences. As a result of their study they found that students were able to make choices and that their preferences were often markedly at odds with the content and methodology that they were exposed to in classes. They report that
Students do not like classes in which they sit passively, reading or translating. They do not like classes where the teacher controls everything. They do not like reading English literature much, even when they are literature majors. Thus it is clear that the great majority of university English classes are failing to satisfy learner needs in any way. Radical changes in the content of courses, and especially in the types of courses that are offered, and the systematic retraining of EFL teachers in learner-centred classroom procedures are steps that must be taken, if teachers and administrators are seriously interested in addressing their students needs. (Widdows & Voller, 1991)
In some foreign language contexts, the notion of student choice may be a relatively unfamiliar or even alien one. In such a case, it is preferable to engage the learners in a relatively modest level of decision making in the first instance. For example, if the data for a lesson include a reading passage and a listening text, learners might be asked to decide which they would rather do first, the reading or the listening. If teachers are uncomfortable with the idea of students doing different things at the same time, then it can be put to a class vote. They could then gradually be involved in making choices such as the following, in which the activity type and task is similar. The point is not that learners in different groups will be doing things that are radically different but that they are being sensitized to the notion of making choices.
You choose: Do A or B. A Group Work. Think about the last time you went grocery shopping. Make a list of all the things you bought. Compare this list with the lists of three or four other students. Whose list is the healthiest? B Group Work. Think about all the healthy things you did last week. Make a list. Compare this list with the lists of three or four other students. Who had the healthiest week?
Once learners are used to the idea, they can be invited to make more elaborate choices, as in the following example.
144 TESOL QUARTERLY
Task 1: You Choose Look quickly at the next three tasks and decide whether these are listening, speaking, reading, or writing tasks. Now decide the order in which you wish to do them. Circle your choices. Task I11 do this task . . . . . . Task 2: A . . . . . . . . . . . . . task 1st 2nd 3rd Task 3: A . . . . . . . . . . . . . task 1st 2nd 3rd Task 4: A . . . . . . . . . . . . . task 1st 2nd 3rd
These examples illustrate the point that even within the various points on the learner-centred continuum, there is a wide range of possibilities. Having encouraged learners to make choices, the next step on the continuum would be to provide them with opportunities to modify and adapt classroom tasks. This could be a preliminary step to teaching students to create their own tasks. This need not involve highly technical materials design skills, which would clearly be unrealistic. I have started learners on the path toward developing their own materials by giving them the text but not the questions in a reading comprehension task and asking them, in small groups, to write their own questions. These are then exchanged with another group to be answered and discussed. At a more challenging level, learners would become teachers. There is nothing like the imminent prospect of having to teach something for stimulating learning. Lest this should be thought utopian, I can point to precedents in the literature. Assinder (1991), for example, gave her students the opportunity of developing video-based materials which they subsequently used for teaching other students in the class. Her class was composed of 9 high intermediate-level EFL students undertaking an intensive English course in Australia. The innovation was a success, the critical factor of which, according to Assinder, was the opportunity for the learner to become the teacher:
I believe that the goal of teaching each other was a factor of paramount importance. Being asked to present something to another group gave a clear reason for the work, called for greater responsibility to ones own group, and led to increased motivation and greatly improved accuracy. The success of each groups presentation was measured by the response and feedback of the other group; thus there was a measure of in-built evaluation and a test of how much had been learned. Being an expert on a topic noticeably increased self-esteem, and getting more confident week by week gave [the learners] a feeling of genuine progress. (Assinder, 1991, p. 228)
The final level on the learning process continuum I would like to propose here is the notion of the learners becoming language researchLEARNING AND INSTRUCTION 145
ers. Once again, for those who think this notion fanciful or utopian, there is a precedent in the literature. Heath (1992), working with black dialect speakers in an Alabama high school and a group of ESL students in south Texas, asked her collaborators to document the language they encountered in the community beyond the classroom.
Students were asked. . . . to work together as a community of ethnographers, collecting, interpreting, and building a data bank of information about language in their worlds. They had access to knowledge I wanted, and the only way I could get that knowledge was for them to write to me. They collected field notes, wrote interpretations of patterns they discovered as they discussed their field notes, and they answered the questions I raised about their data collection and their interpretations. (p. 42)
Despite the struggle involved, students learned through the process of becoming ethnographic researchers that communication is negotiation, and they got to reflect on the important relationships between socialization, language, and thought. In substantive terms, all students moved out of the basic English into regular English classes, and two moved into honors English. As Heath (1992) reports, Accomplishments were real and meaningful for these students (p. 144). There is a rapidly growing literature on learning strategies and learner strategy training which, I believe, supports the thrust of what I have had to say so far. OMalley and Chamot (1990) provide a comprehensive review of the literature, as well as presenting topologies, models, and pedagogical strategies. Their work points to the generally positive effect of strategy training. Wenden and Rubin (1987) present a series of empirical investigations designed to illuminate two central questions: What do learners do to acquire second language competence? and What can be done to facilitate this process? (p. xvii). The studies show the diversity of learning skills and strategies which learners bring to the task of learning another language and also illustrate and illuminate the metalinguistics awareness of learners themselves of the processes underlying their own learning. In terms of classroom practice, the studies reinforce the ideas set out here. For example, Chamot (1987) reports that teachers can profitably encourage students to identify and record their own use of strategies, and then direct students to utilize strategies for a variety of activities. Less proficient students could be encouraged to employ the strategies used by more proficient students. Oxfords (1990) book also draws on current research on learner strategies but focuses more directly on the practical pedagogical implications of this research for incorporating strategy training into language learning. She develops an exhaustive taxonomy of learning strategy types which are illustrated in detail in the book. A great deal of early research focused on the issue of the good
146 TESOL QUARTERLY
language learner (see, e.g., Rubin, 1975, 1981; Rubin & Thompson, 1982). Rubin, who pioneered this work, found that all learners do apply strategies and that certain of these strategies seemed to be consistently utilized by good language learners. The implication here is that once strategies used by good language learners are identified, they can be taught to less effective language learners. The unanswered issue here, however, is whether strategy preferences can be changed or whether they are symptomatic of deeper styles that are fundamental to the cognitive and personality styles of the individual and therefore impervious to change. From an intercultural perspective which sees behavior as being affected by context, conclusions on the effectiveness of learner strategy training reached in (largely) Western educational contexts should be treated with caution. At best they should be seen as interesting working hypotheses to be investigated rather than firm conclusions to be embraced. Learners who have reached a point where they are able to define their own goals and create their own learning opportunities have, by definition, become autonomous. Concepts of self-direction and learner autonomy, which gained a certain degree of prominence during the 1970s (see, e.g., Holec, 1979; Riley, 1982), and then appeared to wane, are recapturing the interest of language educators and researchers. Evidence for this can be found in several recent publications (an excellent example is Gardner & Miller, 1994) as well as conference papers and presentations. In fact, in June 1994, an international conference entitled Autonomy in Language Learning was held in Hong Kong. I would argue that autonomy, like the other central constructs dealt with in this paper, is not an all-or-nothing concept. The ability of individuals to take responsibility for their own affairs (in this case language learning) will be largely determined by the context in which the learning takes place. Contextual factors impinging upon learning will include the age and proficiency level of the students, previous and current educational experiences, the goals of the language program, and the attitude and training of the teacher. The cultural and contextbound nature of autonomy is highlighted in a project described by Roberts, Davies, and Jupp (1992). The aim of the project was to develop a more student-centred approach to language learning within a multiethnic workplace context. In evaluating the project, they point out that because of cultural difference and language difficulties, some students found exercises in autonomous learning bewildering, irrelevant, and unfamiliar, given the strong tradition of learning through other methods which the majority had experienced (p. 318). However, they also point to the benefits of the student-autonomy project. These included a greater appreciation of course objectives on the part of
teachers and learners, greater student awareness of language, and a growing ability on the part of students to carry out needs analysis and self-assessment. If the concept of autonomy is problematic in Western contexts, one might expect it to be even more so in non-Western ones. Several recent studies bear out this observation, although they also give cause for optimism to those who believe that encouraging some degree of learner autonomy is justified. Farmer (1994), for example, describes an independent learning program in Hong Kong. In this context, where learning is highly structured,
learners are expected to, and themselves expect to, adopt a highly passive role. Indeed it might be said that formal education in this context teaches the need to be taught: learners are conditioned to believe that in order to learn one must be taught and that the teacher holds a monopoly over the transmission of knowledge. (p. 14)
The program Farmer described attempted to incorporate elements from the local culture. For example, because the society is a grouporiented one, the learning program was group driven and group negotiated. An evaluation of the program showed that it had achieved a degree of success. Although a large majority of the students still wanted a teacher to be present at all times, a majority also gave positive evaluations of key aspects of the program including analysis of needs and selfselection of materials. One third of the students felt better equipped to work independently as a result of the program. A question sometimes posed by foreign language teachers (and learners) is: Why should I teach/learn this language when the chance that I shall ever use it for genuine communication is an extremely distant prospect indeed? I believe that one of answers to this question can be found in the ideas presented above. By sensitizing learners to the nature of the learning process, by helping them develop skills in cognitive operations such as classifying, brainstorming, inductive and deductive reasoning, by getting them to cooperate with each other, by giving them opportunities to make choices and to develop independent learning skills, we are fostering the cognitive, affective, interpersonal and intercultural knowledge, skills, and sensitivities which provide a rationale for a great many educational systems around the world. In fact, the knowledge, skills, and sensitivities referred to above are not peculiar to language learning, and could just as readily be taught through other subjects on the school curriculum. However, it is difficult to think of another subject more appropriate for developing such skills than a foreign language, particularly when one considers the readymade advantage that a foreign language has for developing intercultural sensitivities and understandings.
148 TESOL QUARTERLY
In this section of the article, I have suggested that one way of narrowing the gap between teaching and learning is to incorporate into language programs opportunities for learners to reflect on and, where feasible, take charge of their own learning processes. I have argued that this involvement is not an all-or-nothing issue but that it can happen at any number of levels. The extent to which it is feasible for learners to become involved will depend on critical contextual variables such as the objectives of the language program, the age, stage, and previous learning experiences of the students, the attitude and training of the teachers, and the philosophy of the institution within which the learning takes place. I have also illustrated through practical examples ways in which the ideas set out in the section can be operationalized in both second and foreign language contexts.
THE LANGUAGE CONTENT DOMAIN

The final domain which can affect the gap between teaching and learning is the language content domain. Although the positive effects of instruction are no longer in question, there is still a great deal of controversy and doubt over the relationship between instruction and learning (see, e.g., Ellis, 1991; Long, 1988; Pienemann, 1989). To deal adequately with the debate would necessitate an article in its own right. Given the centrality of the issues to the question under discussion, it needs to be dealt with however superficially. Although the following review is highly selective, it is representative of the kinds of research carried out in this area. These studies investigated the acquisition of grammar by children and adults in both second and foreign language contexts. The name most closely associated with the debate on the relationship between learning and acquisition is Krashen (see, e.g., Krashen, 1982), who argued that there are two distinct mental processes operating in the development of a L2 language: conscious learning and subconscious acquisition. Krashen argued that instruction leads to conscious learning, not to subconscious acquisition and that it is the latter that underlies the ability to communicate in a foreign language. A study by Ellis (1984) involving ESL children in Britain appeared to support Krashens position. Ellis found that formal instruction appeared to have little effect on the acquisition of question forms. However, Swain (1985), in reviewing work carried out in Canada, argued that comprehensible input is a necessary but not sufficient condition for acquisition, that learners need opportunities to interact as well. Montgomery and Eisensteins (1985) study appears to support this view, finding that learners who were given opportunities to interact in addition to receivLEARNING AND INSTRUCTION 149
ing grammatical instruction developed greater fluency than those who received instruction only. This is not surprising. What is surprising is their finding that the grammar plus opportunities to interact group also outperformed the grammar only group on tests of grammatical knowledge. Similarly, Schmidt and Frota (1986), in a case study of the acquisition of Portuguese as a second language, found that both instruction and opportunities to interact out of class were both necessary. They also argue the case for conscious learning, claiming that improvements occurred when the subject of the study noticed the gap (pp. 310311) between the language he was using and that of the native speakers with whom he was interacting. Pienemann (1989), working with adult learners of German as a foreign language in Australia, proposed a series of development stages and argued that a target form would only appear in the productive repertoire of learners when they were developmentally ready to acquire that particular form. Pienemanns answer to the question of why learners do not learn what teachers teach is that teachers are often trying to teach the unteachable. In her review of a great deal of Canadian research into the relative merits of traditional and communicative classrooms, Spada (1990) concluded that classrooms that were basically communicative in orientation, but that contained opportunities for explicit grammar instruction, were superior to traditional classrooms that focused heavily on grammar, as well as immersion programs that eschewed explicit grammatical instruction. A similar outcome was obtained by Doughty (1991), who investigated the acquisition of relativization by adult learners in Australia. She found that learners receiving instruction outperformed learners who received only exposure In an EFL context in Singapore, Lim (1992) found that the frequency and quality of learner participation related significantly to qualitative aspects of learner participation such as the range of speech acts and control of conversational management techniques. Furthermore, learner participation in class related significantly to improvements in language proficiency. Fotos (1993), working with adult EFL students in Japan found that small-group, problem-solving tasks are as effective as formal teacherfronted instruction for grammatical consciousness raising. Her study places squarely on the agenda the issue of just what is meant by instruction. In an adult EFL context in China, Wudong (1994) found that declarative knowledge (i.e., the ability to identify errors and state rule violations) did not automatically lead to procedural knowledge (i.e., the ability to put known forms to communicative effect). What did facilitate
150
TESOL QUARTERLY
this transference was the addition of opportunities to activate knowledge through output activities. Zhou (1991), also working in China, although with children rather than adults, found that formal instruction resulted in acquisition of some structures (passives) but not others (tense and aspect). Zhou also claims on the basis of the research that explicit (declarative) knowledge can be converted to implicit (procedural) knowledge through practice. Finally, Mllering & Nunan (1994) investigated the acquisition of modal particles by adult learners of German as a foreign language in Australia. They found that instruction made a difference but, once again, only in certain areas. The studies reviewed in this section are summarized in Figure 5. What do we make of this rich array of outcomes? It seems to me that four critical variables appear and reappear in the research. These can be divided into factors external to the learner (e. g., the provision of instruction; the provision of opportunities for learners to use the target language), and factors which are learner-internal (e.g., the ability to describe the rules of the target language; the ability to put these rules to use in communication). The challenge for research is to tease out the complex interplay between these variables. The one thing that does seem to be beyond dispute in this selective review is that the instruments which have been thus far employed are relatively blunt. In relation to instruction, one would want to know what kind of instruction. Regarding interaction, one would want to know about the type of interaction, how it was arranged, and how the data were collected. Similarly with the learner-internal factors, we need to know which aspects of the lexicogrammar are being investigated and reported, how the data were collected, in what contexts, and under what conditions. In practical terms, the studies summarized in this section suggest that there is value in teaching tasks which encourage learners to come to an inductive understanding of grammatical rules and principles through communicative small-group tasks and discussion. There is no room here to provide detailed examples of what these might look like, although the following examples should illustrate the procedure. Rutherford (1987) provides the following example of a grammar consciousness-raising task.
A. Which, if any, of these sentences contains an error? Find the errors and correct them. 1. In Lake Maracaibo was discovered the oil. 2. After a few minutes the guests arrived. 3. In my country does not appear to exist and constraints on womens rights. (Rutherford, 1987, p. 161)
LEARNING AND INSTRUCTION
151
FIGURE 5
A Selective Review of Investigations into the Effects of Instruction on Acquisition
Researcher Krashen Ellis Swain Montgomery/ Eisenstein Schmidt/ Frota Pienemann Spada
Date Location 1982 Child ESL United States 1984 Child ESL Britain 1985 Child FFL Canada 1985 Adult ESL United States 1986 Adult PSL Brazil 1989 Adult GFL Australia 1990 EFL/FFL Canada 1991 Adult ESL Australia 1992 EFL Singapore
Results Instruction does not lead to acquisition. Comprehensible input is necessary and sufficient for acquisition. Formal instruction on question forms has little effect on the acquisition of question forms. Comprehensible input alone does not lead to acquisition. Grammar plus opportunities to communicate lead to greater improvements in fluency and grammatical accuracy than grammar only. Instruction and opportunities to communicate out of class are both necessary. Improvement occurred when subject consciously noticed the gap. Grammatical forms will only be acquired when instruction matches the learners developmental stage. (Communicative classrooms with instruction plus opportunlties,for interaction are superior to traditional instruction and also to immersion programs. Learners receiving instruction (both meaning and form focused) outperformed exposure only learners on knowledge of relativization. Frequency/quality of learner participation related significantly to qualitative aspects of learner participation, for example, range of speech acts and control of conversational management techniques. Learner participation in class related significantly to improvements in language proficiency. Small-group, problem-solving tasks are as effective as formal teacher-fronted instruction for grammatical consciousness-raising. Declarative knowledge does not lead to procedural knowledge without opportunities to activate knowledge through output activities. Formal instruction resulted in acquisition of some structures (passives) but not others (tense and aspect). Explicit (declarative) knowledge can be converted to implicit (procedural) knowledge through practice. Instruction made a difference in the acquisition of German modal particles, although acquisition is relativistic, complex, and organic.
Doughty Lim
Fotos Wudong Zhou
1993 Adult EFL Japan 1994 Adult EFL China 1991 Child EFL China
Mllering/ Nunan
1994 GFL Australia
152
TESOL QUARTERLY
B. Complete the sentences in normal English. 1. Many French Canadians find [They learn English] important. 2. Quebec makes [Quebec preserves its French-speaking identity] a rule. 3. Quebec takes [French is to be given priority over English] for granted. 4. The government left [Will French be the official language of Quebec?] up to the people to decide. (Rutherford, 1987, pp. 165-166)
Most grammar practice activities can be modified so that they provide learners with interactive opportunities for this sort of inductive grammar work, as the following example (one of four exercises on personal pronouns) illustrates:
Group Work Class survey. Ask your classmates questions like this: Do you go to the movies often? Write one name in each box in the chart. See how many boxes you can fill. FIND SOMEONE WHO . . . . YES NO likes hamburgers plays tennis speaks three languages likes classical music goes to the movies often wants to be an actor lives alone likes modern art drives a car works at night Do you know the rule? Fill in the blanks with the correct pronouns from this list: I, you, he, she, it, we, they. Use do with . . . . . . . . . . . . . . . . . . . . . . . . Usedoes with . . . . . . . . . . . . . (Nunan, 1994, p. 20)
As Fotos (1994) points out, the advantage of these tasks is that they can be employed both in communicative classrooms as substitutes for grammar lessons and in traditional, teacher-fronted classrooms as a method of studying grammar while providing essential opportunities for communicative use of the target language (p. 327). They would seem to be particularly suited to those foreign language contexts in which learners have limited opportunities to interact in, and form hypotheses about, the target language.
DISCUSSION
I should like to summarize the ground covered in this paper by taking the relatively unusual step of proposing an hypothesis. It is a
rather large hypothesis, with numerous subsections. I propose it, not as an hypothesis to be tested through a formal experiment, but as a series of statements which individual teachers can contest through the realities of their own classroom. I should like to suggest that, all other things being equal, the gap between teaching and learning will be narrowed when learners are given a more active role in the three key domains of content, process, and language. In the experiential content domain, when: instructional goals are made explicit to learners learners are involved in selecting, modifying, or adapting goals and content learners create their own goals and generate their own content active links are created between the content of the classroom and the world beyond the classroom. In the learning process domain, when:
q
learners are trained to identify the strategies underlying pedagogical tasks learners are encouraged to identify their own preferred learning styles and to experiment with alternative styles learners are given space to make choices and select alternative learning pathways learners are given opportunities to modify, adapt, create, and evaluate pedagogical tasks and learning processes learners are encouraged to become their own teachers and researchers. In the language content domain, when:
learners are given opportunities to explore the organic, nonlinear relationships between language forms and communicative functions or in Hallidays (1985) terms, to explore the relationships between what language is and what it does classroom learning opportunities are created which enable learners to draw on the external factors of instruction and interfactional opportunities in order to articulate their understanding of how language works as well as putting language to communicative use in real or simulated contexts.
In such use, the three domains discussed in this article begin to converge.
154
TESOL QUARTERLY
CONCLUSION
One of my favorite pieces of learner data was contributed by an informant in Johnstons (1987) seminal study of language acquisition among immigrants in Australia. In reflecting on his own knowledge of English, Genghis, a Turkish immigrant, mused:
Before I came here I was knowing all the English language tenses(s) . . . present tense . . . past tense . . . present perfect tense . . . perfect tense . . . future tense . . . future in the past . . . everything . . . I was knowing . . . I am knowing now . . . I just asked, er, one day the boss, I said to him How you knowing this tense? for example go . . . How can you use this word? . . . past tense? present tense? the other tense? He just looked at me like that . . . he told me I dont know Genghis. This is Australian people. I am Turkish people. I am knowing, he doesnt know. Can you explain this? (Genghis, cited in Johnston, 1987, p. 5)
It is the delightful dissonance between Genghiss ability to articulate his understanding of how English works and his ability to put that understanding to work in using language for communication that partly explains the gap between teaching and learning. Although we have made progress in our understanding of language learning and teaching, I do not believe the point will ever be reached when we can say with the arrogance of the normative scientist, Ah, yes, now I know! As long as our conceptions of language, learners, and the learning process continue to evolve, and the teaching/learning process is transformed through practice and research, so too will the problems and challenges confronting the profession. As the problems change, so too will the solutions. It is up to each of us, as professionals, neither to accept proposals uncritically, nor to reject them out of hand but to reflect upon them and to contest them against the reality of our own context and situation. In the final analysis, however, it is the learner who must remain at the centre of the process, for no matter how much energy and effort we expend, it is the learner who has to do the learning. All of this notwithstanding, I believe that there are substantive steps we can take to narrow the gap between teachers and learners, between instruction and outcome, between the curriculum documents (which all too often sit on staffroom shelves gathering dust), and the language classroom where teachers and learners collaboratively engage in the co-construction of the learning process. Looking back, it may well be that following Allwright (personal communication, April, 1994) the question ought to be not Why dont learners learn what teachers teach?, but Why dont teachers teach what learners learn?
155
ACKNOWLEDGMENTS
I should like to thank David Block and two anonymous reviewers for their detailed and insightful comments on earlier drafts of this paper.
THE AUTHOR
David Nunan is Professor of Applied Linguistics and Director of the English Centre at the University of Hong Kong. He has published extensively on the language classroom and learner-centred instruction. His latest publication (edited with Kathleen Bailey, Cambridge University Press) is entitled Voices and Viewpoints: Qualitative Research in Second Language Education.
REFERENCES
Aiken, G., & Pearce, M. (1994). Learning outside the classroom. Sydney, Australia: National Centre for English Language Teaching and Research. Allwright, R.L. (1984). Why dont learners learn what teachers teach?The interaction hypothesis. In D. M. Singleton & D. G. Little (Eds.), Language Learning in Formal and Informal Contexts (pp. 318). Dublin, Ireland: IRAAL. Allwright, D. ( 1987). Observation in the language classroom. London: Longman. Allwright D., & Bailey, K. (1991) Focus on the language classroom: An introduction to classroom research for language teachers. Cambridge: Cambridge University Press. Assinder, W. (1991). Peer teaching, peer learning; one model. ELT Journal, 45, 218229. Block, D. (1994). A day in the life of an English class: Teacher and learner perceptions of task purpose in conflict. System, 22, 153175. Block, D. (in press). A window on the classroom: Classroom events viewed from different angles. In K. Bailey & D. Nunan (Eds.), Voices and viewpoints: Qualitative research in second language education. New York: Cambridge University Press. Brindley, G. (1984). Needs analysis and objective setting in adult migrant education. Sydney, Australia: Adult Migrant Education Service. Chamot, A. U. (1987). The learning strategies of ESL students. In A. Wenden & J. Rubin (Eds.), Learner strategies in language learning. Oxford: Pergamon Press. Dam, L., & Gabrielsen, G. (1988). Developing learner autonomy in a school context: A sixyear experiment beginning in the learners first year of English. In H. Holec (Ed.), Autonomy and self-directed learning: Present fields of application. Strasbourg, France: Council of Europe. Doughty, C. (1991). Second language instruction does make a difference: Evidence from an empirical study of SL relativization. Studies in Second Language Acquisition, 14, 431-469. Ellis, G., & Sinclair, B. (1989). Learning to learn English: A course in learner training. Cambridge: Cambridge University Press. Ellis, R. (1984). Can syntax be taught? A study of the effects of formal instruction on the acquisition of WH questions by children. Applied Linguistics, 5, 138152. Ellis, R. (1991). Grammaticality Judgments and second language acquisition. Studies in Second Language Acquisition, 13, 161184.
156
TESOL QUARTERLY
Farmer, R. (1994). The limits of learner independence in Hong Kong. In D. Gardner & L. Miller (Eds.), Directions in self-access language learning (pp. 1327). Hong Kong: Hong Kong University Press. Fotos, S. (1993). Consciousness raising and noticing through focus on form: Grammar task performance versus formal instruction. Applied Linguistics, 14, 385407. Fotos, S. (1994). Integrating grammar instruction and communicative language use through grammar consciousness-raising tasks. TESOL Quarterly, 28, 323351. Gardner, D., & L. Miller (Eds.). (1994). Directions in self-access language learning. Hong Kong: Hong Kong University Press. Halliday, M. A. K. (1985). Spoken and written language. Victoria, Canada: Deakin University Press. Heath, S. B. (1992). Literacy skills or literate skills? Considerations for ESL/EFL learners. In D. Nunan (Ed.), Collaborative language learning and teaching (pp. 40 55). Cambridge: Cambridge University Press. Holec, H. (1979). Autonomy and foreign language learning. Oxford: Pergamon Press. Johnston, M. (1987). Understanding learner language. In D. Nunan (Ed.) Applying second language acquisition research (pp. 544). Sydney, Australia: National Centre for English Language Teaching and Research. Krashen, S. (1982). Principles and practice in second language acquisition. Oxford: Pergamon. Lim, S. (1992). Investigating learner participation in teacher-led classroom discussions in junior colleges in Singapore from a second language acquisition perspective. Unpublished doctoral dissertation, National University of Singapore. Long, M. H. (1988). Instructed interlanguage development. In L. Beebe (Ed.) Issues in second language acquisition: Multiple perspectives (pp. 115141). New York: Newbury House. Montgomery, C., & Eisenstein, M. (1985). Reality revisited: An experimental communicative course in ESL. TESOL Quarterly, 19, 317334. Mllering, M., & Nunan, D. (1994). Pragmatic in interlanguage: German modal particles. Unpublished manuscript. Nunan, D. (1987, April). Hidden agendas in the language classroom. Paper presented at the RELC Regional Seminar, Singapore. Nunan, D. (1988). The learner-centred classroom. Cambridge: Cambridge University Press. Nunan, D. (1989). Understanding language classrooms. London: Prentice Hall. Nunan, D. (1990). Action research in the language classroom. In J. C. Richards & D. Nunan (Eds.), Second language teacher education (pp. 6281). Cambridge: Cambridge University Press. Nunan, D. (1991). Language teaching methodology. London: Prentice Hall. Nunan, D. (1994). ATLAS. Learning-centred communication. Boston: Heinle & Heinle. Nunan, D. (in press). Hidden voices: Insiders perspectives on classroom interaction. In K. Bailey & D. Nunan (Eds.), Voices and viewpoints: Qualitative research in second language education. New York: Cambridge University Press. Nunan, D., & Lamb, C. (in press). The self-directed teacher: Managing the learning process. New York: Cambridge University Press. OMalley, J. M., & Chamot, A. U. (1990). Learning strategies in second language acquisition. Cambridge: Cambridge University Press. Oxford, R. (1990) Language learning strategies. New York: Newbury House. Parkinson, L., & OSullivan, K. (1990). Negotiating the learner-centred curriculum. In G. Brindley (Ed.), The second language curriculum in action (pp. 112-127). Sydney, Australia: National Centre for English Language Training and Research.
157
Pienemann, M. (1989). Is language teachable? Psycholinguistic experiments and hypotheses. Applied Linguistics, 10, 4072. Riley, P. (1982). Learners lib: Experimental autonomous learning scheme. In M. Geddes & G. Sturbridge (Eds.), Individualisation (pp. 6163). London: Modern English Publications. Roberts, C., Davies, E., & Jupp, T. (1992). Language and discrimination: A study of communication in multi-ethnic workplaces. London: Longman. Rubin, J. (1975). What the good language learner can teach us. TESOL Quarterly, 9, 4151. Rubin, J. ( 1981). The study of cognitive processes in second language learning. Applied Linguistics, 2, 117131. Rubin, J., & Thompson, I. (1982). How to be a more successful language learner. Boston: Heinle & Heinle. Rutherford, W. (1987). Second language grammar: Learning and teaching. London: Longman. Schmidt, R., & Frota, S. (1986). Developing basic conversational ability in a second language: A case study of an adult learner of Portuguese. In R. Day (Ed.) Talking to learn: Conversation in second language acquisition (pp. 237326). Rowley, MA: Newbury House. Slimani, A. (1992) Evaluation of classroom interaction. In J. C. Alderson & A. Beretta. (Eds.) Evaluating second language education (pp. 197211). Cambridge: Cambridge University Press. Spada, N. (1990). Observing classroom behaviors and learning outcomes in different second language programs. In J. C. Richards & D. Nunan (Eds.), Second language teacher education (pp. 293310). Cambridge: Cambridge University Press. Swain, M. (1985). Communicative competence: Some roles of comprehensible input and comprehensible output in its development. In S. Gass & C. Madden (Eds.), Input in second language acquisition (pp. 235-253). Rowley, MA: Newbury House. Widdows, S., & Voller, P. (1991). PANSI: A survey of ELT needs of Japanese university students. Cross Currents, 8, 2. Wenden, A., & J. Rubin (Eds.). (1987). Learner strategies in language learning. Oxford: Pergamon Press. Willing, K. (1988). Learning styles in adult migrant education. Sydney, Australia: National Centre for English Language Teaching and Research. Willing (1989). Teaching how to learn. Sydney, Australia: National Centre for English Language Teaching and Research. Wudong, W. (1994). English language development in China. Unpublished doctoral dissertation. University of Tasmania. Zhou, Y. P. (1991). The effect of explicit instruction on the acquisition of English grammatical structures by Chinese learners. In C. James & P. Garrett (Eds.), Language awareness in the classroom (pp. 254277). London: Longman.
158
TESOL QUARTERLY
THE FORUM
The TESOL Quarterly invites commentary on current trends or practices in the TESOL profession. It also welcomes responses or rebuttals to articles or reviews published in the Quarterly. Unfortunately, we are not able to publish responses to previous Forum exchanges.
Comments on Joy Reids Responding to ESL Students Texts: The Myths of Appropriation
There Are Myths and Then There are Myths
CHRIS HALL
Wright State University For the thing about a myth is not whether it is true or not, nor whether it should be true, but that it is somehow truer than truth itself Thomas Keneally (1982)
s Joy Reids Responding to ESL Students Texts: The Myths of Appropriation (Vol. 28, No. 2) offers some constructive advice for responding to students writings that experienced and inexperienced ESL teachers alike can add to their growing stockpile of teaching tips. I would add to her list of questions on intervention strategies and goals a simple reminderwhen in doubt about the writing, feel free to let the writer explain. Although I find much of her advice reasonable and am pleasantly surprised that we share much in common about providing honest and constructive feedback to our students, her article invites some further comment. How to respond to student texts and avoid the abuses of appropriation is a good place to begin. Unlike Reids experience, some of us came to terms with the phenomenon of appropriation much earlier and were able to work out reasonably sound systems for providing responses to our ESL students about their writing. During the past few decades, the majority of research does not seem to me to obscure the distinction between giving constructive feedback and coopting a students intellectual property. Take Sommers (1982) work, for example, which Reid uses to introduce the term appropriating. The article has been widely reproduced and is familiar to ESL researchers. In fact, Robb, Ross, and Shortreed (1986) in their empirical study of how feedback can affect editing tasks in a L2 context mention it favorably. After her unassuming introduction of the term, Sommers (1982) offers
159
examples and discussion on the state of text commenting clearly identifying the excesses associated with text appropriation. In the article, her advice for responding more effectively to student writers complements Reids quite nicely:
We need to develop an appropriate level of response for commenting on a first draft, and to differentiate that from the level suitable to a second or third draft. Our comments need to be suited to the draft we are reading. In a first or second draft, we need to respond as any reader would, registering questions, reflecting befuddlement, and noting places where we are puzzled about the meaning of the text. Comments should point to breaks in logic, disruptions in meaning, or missing information. Our goal in commenting on early drafts should be to engage students with the issues they are considering and help them clarify their purposes and reasons in writing their specific texts. (Sommers, 1982, p. 155)
I do not believe that research has intimidated a significant number of ESL teachers from limiting their responses to student texts. Leki (1990), for example, mentions the problem of appropriation without noting an overreaction one way or the other. Her advice for dealing with a case of appropriation coincides with Greenhalghs (1992) and does not suggest that we should be so irresponsible as to completely disengage ourselves from students texts to allow for some sort of vacuous writing experience:
If we do not want to appropriate a students paper to ourselves by marking it in accord with our own mental image of the Ideal Text to which a paper seeks to conform, then we need to compare the paper to some other text one that corresponds to the students intentions in a given piece of writing. How can we know what those intentions might be, especially with inexperienced L2 writers? The easiest way is to engage students in dialogue on those intentions. (Leki, 1990, p. 64)
With that said, I would not dispute Reids earnest account of coming to terms with the troubling issues associated with appropriation and finding a satisfactory solution for her teaching. We have all had these private doubts and struggles to define our classroom practices. But I cannot accept as accurate her analysis of the type of influence process and writer issues have exerted on our discipline. There are myths and then there are myths. Reid seems to be suggesting in her analysis that process research is too heavily influenced by researchers and teachers compulsively critical of certain traditional practices in academic writing and narrowly supportive of expressive writing. I do not deny that criticism coming from
160 TESOL QUARTERLY
some process supporters has been too sweeping. Like others, I have found myself flinching at some of the reprimands. As for the expressive bias, however, I see a truer than truth myth that could stand another look. For instance, Johns (1990), does not see expressive views as dominant. Using Faigleys (1986) distinction between expressivist and cognitivist groups in the process movement, she maintains it is the cognitivist, or writing as problem-solving, group that has had more effect upon ESL research and teaching, however (pp. 2526). In an early work, Raimes (1983) seems to maintain a utilitarian view when applying process research to the classroom, and she certainly is not blind to the importance of rhetorical issues such as purpose and audience. Zamel (1984) also acknowledges a general concern for a product and the reader. I detect a more eclectic influence in L2 literature as well as in the practices of teachers who find process and writer concerns worthwhile; this eclecticism is evident in Reids article. Although committed to the English for academic purposes (EAP) movement (which often seems to devalue expressivists attitudes about writing), Reid unreservedly cites Peter Elbow and Donald Murray, radical expressivists, to bolster her suggestions for avoiding the evils of appropriation. Even more ironic is the use of journals in her classroomwhich is inextricably linked to expressivism and, according to Johns (1990), to expressivists. It would seem therefore that there are some salutary functions for expressive writing even in the EAP curriculum and that expressive writing need not result in appropriation. Appropriation certainly can exist in an expressivist, cognitivist, and eclecticism classroom, and Reid has every right to alert us to the dangers as she sees them. But she misses the full extent of where and when appropriation can take hold, for it is even possible in her model. In the hands of a sincere but naive or overly enthusiastic proponent of contrastive rhetoric, for example, the same sorts of extremes and abuses can happen as with processand writeroriented research. Like any other supporter of an evolving but controversial approach in our profession, an advocate of contrastive rhetoric must avoid its extremes. As Raimes (1991) points out:
A broad use of contrastive rhetoric as a classroom consciousness-raising tool can point to linguistic variety and rhetorical choices; a narrow use would emphasize only prescriptions aimed at countering L1 interference. (p. 418)
Intentionally or unintentionally, an advocate who veers too far toward rhetorical relativism can convey to students the message that a
THE FORUM 161
single hypothesis about writing and the writer can encompass each students culture, writing behavior, and needs. When that happens, the teacher runs the risk of reinforcing the classroom myth that one teacher (or a select group of similarly minded teachers) represents the categorical and irrefutable arbiter for what is good or bad academic writing as well as good and bad writing conduct. The student then learns a crippling lesson about academic writing in particular and written communication in generalno decision about organization and audience is valid until the teacher says it is valid. This has got to be egregious appropriation. Of course, not all students succumb to such thinking just as not all students embrace contrastive rhetorics impressions of culture and writing as Liebman-Kleines (1987) investigation demonstrated. However, for those that would, there is the distinct prospect that the expertnovice schism is reinforced, the goal of helping ESL writers to become independent academic writers fades, and the students image of a discourse community remains one of a mysterious and inflexible oligarchy that must be pleased. Is this as likely to happen as other scenarios for appropriation? There is some reason for concern if the current trends in teacher education continue. In their surveys of how writing theory and pedagogy is being presented in TESOL doctoral and masters programs across the U.S., MacDonald and Hall (1990, 1991) found that little attention is given to theoretical and pedagogical issues in either contrastive rhetoric or process, and those programs that do expand on contrastive rhetoric themes have tended to concentrate on what Liebman (1992) has described as old contrastive rhetoric, which attended to a restricted area of rhetoric: the organization of finished expository texts (p. 142). Reid sees the threat of abusive appropriation bearing down on our profession from a group of vocal and narrowly focused researchers and misguided teachers. I see the problem lurking in any individual who misapplies the research and disregards classroom experience. I am sure someone will discover that both researchers and teachers have their truer than truth myths. Until then, I wish we would bury this futile division between process and product and get on with teaching composition.
REFERENCES
Faigley, L. (1986). Competing theories of process: A critique and a proposal. College English, 48, 527542. Greenhalgh, A. M. (1992). Voices in response: A postmodern reading of teacher response. College Composition and Communication, 43, 401410. 162 TESOL QUARTERLY
Johns, A. M. (1990). L1 composing theories: Implications for developing theories of L2 composition. In B. Kroll (Ed.) Second language writing: Research insights for the classroom. (pp. 24-36). New York: Cambridge University Press. Keneally, T. (1982). Schindlers list. New York: Simon & Schuster. Leki, I. (1990). Coaching from the margins: Issues in written response. In B. Kroll (Ed.) Second language writing: Research insights for the classroom (pp. 5768). New York: Cambridge University Press. Liebman, J. D. (1992). Toward a new contrastive rhetoric: Differences between Arabic and Japanese rhetorical instruction. Journal of Second Language Writing, 1, 141165. Liebman-Kleine, J. D. (1987, March). The student as researcher: An ethnographic study of contrastive rhetoric. Paper presented at the 38th Annual Meeting of the Conference on College Composition and Communication, Atlanta, Georgia. (ERIC Document Reproduction Service No. ED 281 194) MacDonald, M., & Hall, C. (1990, March). Training the next generation of ESL writing teachers. Paper presented at the 24th Annual TESOL Convention, San Francisco, California. MacDonald, M., & Hall, C. (1991, March). How are ESL writing teachers trained? Paper presented at the 25th Annual TESOL Convention, New York, New York. Raimes, A. (1983). Techniques in teaching writing. New York: Oxford University Press. Raimes, A. (1991). Out of the woods: Emerging traditions in the teaching of writing. TESOL Quarterly, 25, 407430. Robb, T., Ross, S. & Shortreed, I. (1986). Salience of feedback on error and its effect on EFL writing quality. TESOL Quarterly, 20, 8393. Sommers, N. (1982). Responding to student writers. College Composition and Communication, 33, 148156. Zamel, V. (1984). In search of the key: Research and practice in composition. In J. Handscombe, R. A. Orem, & B. P. Taylor (Eds.), On TESOL 83: The question of control (196207). Washington, DC: TESOL.
The Author Responds . . .

JOY REID University of Wyoming
sI
am grateful to those Quarterly readers who contacted me about my article. I especially appreciate the time and energy Chris Hall spent responding to my article. The satisfaction of writing an article is expanded by discussion with others; I am therefore happy to have this opportunity to continue the dialogue. Chris Halls response focuses on three ideas (personalized here for the sake of discussion):
that I overstate the ability of research to intimidate ESL teachers that I believe the expressivists are wholly responsible for all appropriation of student text, and q that I am bent upon resurrecting and continuing the process/product split.
q q
THE FORUM
163
To the first, I confess that am a worrier; my now grown children remind me regularly of that vice. Indeed, my tendency is to take what I read too seriously, always to be concerned with improving my teaching (and my students learning), and at least initially to be intimidated by the proclamations of others. I tend to worry an idea as a dog worries a bone, working it over and over until I come to some resolution. Thus, it took me a decade to decide to write this article on myths and nearly 2 years to complete it. My secondary objective in writing the article was to reconstruct the impact of the appropriation issue on my teaching, reflecting on what I had learned. The primary objective was to share this reflection with other teachers and teachersin-training who might as a consequence learn, more quickly and efficiently than I did, that we teachers need to have more self-confidence in our choices regarding intervention. However, I believe Hall may understate the importance of the appropriation issue in the writing classroom. Despite the fact that the discussion of the issue has all but disappeared from the research literature, I still see the question of response to and possible appropriation of student text as central to the whole enterprise of teaching writing. The tension between appropriation and intervention appears whenever a teacher responds to a student paper: how to achieve that precarious balance between teaching and simply turning students back into the writing process, between guidance that helps students avoid serious problems and allowing students to learn from their own experiences, between intervening directly and providing opportunities for students to revise independently? Whereas Hall may well have come to terms with the issues and feels content with the way he deals with response to student writing, I continue to struggle with this issue, and I think that many teachers also struggle to balance teacherly instruction with student learning. Thus, I believe that although the research discussion on appropriation of student text is no longer visible, the issue is at the heart of teaching writing and continues to be viable, even essential, for teachers to consider. Second, it is true that I used what I knew might be a controversial example of appropriation when I discussed the single, end-of-semester portfolio grade. But I do not consider myself an antiexpressivist. I agree with Hall that any method (cognitivist, expressivist, contrastive rhetoric, and even eclectic), carried to extremes, can become deleterious to students. However, most researchers and teachers who have previously discussed appropriation have used product-based approaches and examplesthe Ideal Text is, almost certainly, a final product. So I decided to use a process-based approach to make my point: that appropriation was not the sole purview of nonexpressivist methods.
164
TESOL QUARTERLY
And, to be honest, experience has demonstrated that the more time I spend with student drafts, the less time students spend on them; that is, once students see how interested I am in making suggestions to improve their drafts, they do their best to transfer all responsibility to me. Note that the possibility for appropriation is great in this situation. As a result, I have rearranged the processes of intervention in my writing classes. I depend more on multiple audiences (e. g., peers, writing center specialists, and writing tutorials) to assist students on rough drafts. I then spend more time on what students know is a final draft that will receive a grade, and I require that they revise these essays in quick ways that respond mostly to my comments, in the same way that I respond to reviewers of grant proposals and articles sent for review. Then (as the students know from the first day of class) I use a modified portfolio approach during the last 3 weeks of the semester; I ask students to completely revise and rewrite three of their six essays to turn in for a final additional (not substituted) grade. By the end of the semester, I hope, students have enough distance from their papers, more resources and coping strategies, and, most important, more information about being writers, that they can use resources and their own writing experiences to make meaningful and substantial changes on those essays. Hall also suggests that I have misused expressivist theorists in my attempt to explain the differences between intervention and appropriation. In truth, I quoted Donald Murray and Peter Elbow without irony; in addition to being expressivists, they are both canny writing teachers who have addressed appropriation issues in ways that I respect. In much the same way, I use metacognitive (not personal narrative) journals in my writing classes; they work well, and if these journals were originally tied to strictly expressivist teaching, they are now wisely used in write-to-learn content courses across the university curriculum. Third, I am committed, in this article as well as in other writing I have published, to demonstrating that process and product exist along a continuum and that a concentration on either can make learning to write more difficult for students. The false dichotomy between the two that I define in my article has not only been written to death but has had a negative effect on ESL writing research. My point, in my article, is rather to probe the differences between appropriation and intervention in light of (a) teacher and student responsibilities and (b), writing as a contextualized, social process. Finally, I continue to worry: How can I balance my determination to make my scientific and technical writing class student-responsible in light of the students clear need to depend heavily on me? What is the best way for TESOL to approach accreditation of intensive English
THE FORUM
165
language programs? And what about revisionlots has been written about it, but we teachers do not really teach revision strategies in the same quantity and quality as we teach composing strategies . . . .
Comments on Virginia LoCastros Learning Strategies and Learning Environments

Making Sense of Learning Strategy Assessment: Toward a Higher Standard of Research Accuracy
REBECCA L. OXFORD
University of Alabama
JOHN M. GREEN
Salem State College and University of Puerto Rico at Mayaguez
researchers who frequently work with the Strategy Inventory for Language Learning (SILL) (Oxford, 1990), we were at first delighted to see Virginia LoCastros article in the TESOL Quarterly (Volume 28, No. 2). We felt that we would be receiving new information that could be compared to the SILL data we have collected over the years. We also thought we might find some interesting results that might aid our theory building process. Unfortunately, we found both conceptual and methodological problems that rendered such possibilities moot. We also discovered some statements that could be misleading to future researchers and to consumers of research. CONCEPTUAL PROBLEMS LoCastro gathered quantitative date (using the SILL) and qualitative data (using group interviews), then contrasted the two kinds of data. The conclusions she drew from comparing these two kinds of data are unjustified. Normally people who use the SILL understand the restrictions placed on use and interpretation of this (and any other) summative rating scale. Apparently LoCastro, in comparing her SILL results with findings from group interviews, was quite surprised that the outcomes of these two modes were not the same and on that basis, questioned the validity of the SILL. If one were so inclined, one could just as easily have used the results to question the validity of the group interviews. In fact, with groups of students from junior high age through adult-
s AS strategy
166
TESOL QUARTERLY
hood in many countries around the world, the SILL has proven highly valid. Its predictive and concurrent validity are demonstrated by strong relationships between the SILL on the one hand and language performance on the other (Bedell, 1993; Boraie, Kassabgy, & Oxford, 1994; Dreyer, 1992; Ehrman & Oxford, 1989; Green, 1992; Mullins, 1992; Oh, 1992; Oxford, 1986; Oxford & Burry, 1993; Oxford & BurryStock, in press; Oxford & Nyikos, 1989; Phillips, 1991; Watanabe, 1990; Wildner-Bassett, 1992). Moreover, it has very high reliability, with internal consistency coefficients in the .90s (see summaries of reliability studies in Oxford, 1990; Oxford & Burry, 1993; Oxford & Burry-Stock, in press). The real issue underlying LoCastros difficulty relates to the fact that the two different measurement modes have different purposes: Interviews are designed to produce highly spontaneous, rich information that is idiosyncratic to individuals, and summative rating scales are designed to produce standardized, general information about individuals that can be summarized over the whole group of students. It has been repeatedly said that written, closed-ended self-report instruments (summative rating scales) are not meant to produce the same thing as interviews, think alouds, note-taking, diaries, or other open-ended formats. For example, in Oxford (1990), from which LoCastro obtained the SILL, it is stated:
Because more structured surveys use standardized categories for all respondents, such surveys [more accurately speaking, summative rating scales] make it easier to summarize results for a group and objectively diagnose problems of individual students. However, these surveys might miss the richness and spontaneity of less-structured formats. (p. 199)
Similarly, in a review article in Annual Review of Applied Linguistics, Oxford (1993) states: Structured strategy surveys [i.e., rating scales like the SILL] do not leave much room for the individuals creative responses. However, they are very useful for statistical treatment and group summaries. (p. 176) The SILL and other similar measurement devices provide a good general picture of strategy use. However, these scales do not serve all purposes. For example, they do not purport to cover every single strategy a given student might use in all instances (or even regarding just one main language learning task). Thus, although offering a clear overall picture, these scales do not pretend to be exhaustive in their strategy coverage. Moreover, such scales do not offer latitude for stu-
THE FORUM
167
dents to write in their own strategies, a purpose better served by interviews, diaries, or open-ended surveys. Much of LoCastros critique of the SILL seems to be based on the fact that her qualitative (group interview) data produced richer data for her very small (N = 28) group of subjects than did her SILL data. As she states, The inventory may not be sensitive to the concerns of the respondents and thus may not generate a clear picture of their learning strategies (p. 413). What she seems to really mean is that students may be using strategies that are not included on the SILL. However, as mentioned above, the SILL is not a complete list of all possible strategies, and its purpose is not to produce a picture of all the strategies each individual in the groups might be using or might have used. If one wants to catalogue in much greater detail the strategies used by a small group of students, qualitative methods such as interviews or diaries would make more sense. LoCastro comments that much of the published work on learning strategies is based on research carried out ESL programs in North American universities . . . . [It] seems questionable that the list of learning strategies generated on the basis of such studies can apply to L2 learners with different educational and social backgrounds (p. 409). LoCastro is apparently not aware of the vast body of SILL research across a wide range of cultural settings. This research has been done in Japan, China, Taiwan, Korea, Egypt, Indonesia, Thailand, South Africa, Spain, the U.S., and many other countries. We dispense many technical reports, handouts, and articles on the SILL every week to anyone who asks (see partial list above and the references). A major review article (Oxford & Burry-Stock, in press) on the SILL is coming out in System, 23 (2), and its precursors have been disseminated for years. LoCastro also observes that the SILL includes no strategies specifically addressing listening as a means to learn (p. 412). We find this not only puzzling but also misleading and untrue. There are a number of strategies, such as watching television in the target language or paying attention to what people say, in which listening is an essential component. All the conversation-related strategies in the SILL involve listening. Researchers and teachers have repeatedly used the SILL to obtain information on strategies for listening. Furthermore, Oxford (1990) provides a complete appendix of every strategy listed in the book (many of these strategies were included in the SILL) cross-listed by the four skills of listening, reading, speaking, and writingand listening is just as strongly represented as any other skill. LoCastro also indicates in a footnote that she has still other theoretical criticisms but that this brief report is not the place (p. 413) to air them. It would be far more professional to take one of these three
168 TESOL QUARTERLY
steps: (a) drop such an ambiguous but negative footnote, (b) list the problems in the article, or (c) best of all, exchange information with those who work with the SILL or similar inventories. She indicates that there is some confusion in her mind about the statement that most learning strategies can be applied equally well to both ESL and EFL situations, which she says comes from page 6 of Oxford (1990) and about the fact that the version of the SILL she used was Version 7.0 for ESL/EFL. She says, No explanation is given of the meaning of the label (pp. 411-412). This is patently wrong. Oxford (1990, p. 6) spends three full paragraphs (with their own major headings) discussing the differences between second and foreign language settings. The appendix to the same book contains SILL Version 5.1 (for native English speakers learning another language) and Version 7.0 (for nonnative speakers learning English). The differences, along with when to use each version, are fully explained on page 199 in the assessment chapter of the book. Apparently, LoCastro chose not to refer to these easily locatable explanations.
METHODOLOGICAL PROBLEMS
There are methodological problems in LoCastros study. The sample was very small, compared to the usual hundreds of most studies involving the SILL. She does not address the issue of where the SILL results might be different with a larger sampling or with a different or more heterogeneous group. (As any experienced teacher knows, a class of 28 students will as often as not have an atypical distribution of ability levels or just about any other variable one might care to observe.) LoCastros SILL results indicated medium or average overall strategy use among her subjects, and medium use of the six categories of strategies as well. She reports this finding, but because she fails to report any other SILL date to compare it tono students at other levels, and no statement of what she expected to find-it is hard to say what, if anything, the finding means. (The numbers she reports are actually consistent with what we found in our Puerto Rico study, where all the means were within or very close to the medium range and yet there were significant differences across proficiency levels in out study.) It is hard to tell exactly what LoCastro used as her research procedures, so no one can replicate her study. In order to identify the most frequently used strategies, she writes, I further analyzed eight strategies which has been selected by more than five respondents (p. 411). The methodology here is very unclear. Just how did she analyze these strategies? What does she mean by selected by more than five
THE FORUM 169
respondents: response options of 5 (almost always used) or something else? Her remarks about specific strategies are hard to interpret because she does not specify what strategies she is discussing. What were the eight strategies which had been selected by more than five respondents, for instance? The same problem exists in the third paragraph under Findings, in which she says that Other low-use strategies [according to SILL results] involved vocabulary and imagery, although her interview data revealed that, of the list of the most frequently used strategies, three were vocabulary strategies (p. 411). What specific vocabulary strategies is she referring to? Were the vocabulary strategies identified as frequently used by the interview technique included in the SILL or were they not?
CONCLUSIONS
We have four suggestions, which overlap somewhat. First, it would have been helpful if LoCastro had done her basic homework about the purposes of the SILL before using the instrument (see especially Oxford, 1990, p. 199 and appendices). We cannot expect any datagathering techniqueinterview, summative rating scale, diary, think aloudto serve all conceivable purposes. Researchers need to be concerned with why various techniques were designed and to choose the one technique (or several) suitable to specific needs. Second, if conducting a study in a rather new and rapidly evolving ESL/EFL area such as language learning strategies, the researchers very first step should be to contact earlier investigators and try to obtain up-to-date information that has been published around the world (or that has been completed and is in press). Even a minor attempt at cooperative interchange would have provided LoCastro with abundant information. Third, incomplete and imprecise research reporting, in the confusion about vocabulary strategies and the untruth regarding a supposed absence of strategies involving listening on the SILL should not be tolerated, particularly when investigators worldwide are trying to create a solid foundation for language learning strategy research. Fourth, in our opinion, even rather brief reports of research in progress should be published with adequate citation of existing research. Lack of such citation results in inaccuracies, as we have seen. We hope these recommendations will enable others to avoid the conceptual and methodological pitfalls exhibited in the LoCastro piece. The SILL helps teachers obtain a rapid, broad-brush picture of the strategies students are using and enables teachers and researchers to plan strategy instruction more effectively. Experience has shown that
170 TESOL QUARTERLY
when used in this fashion, the SILL has to date aided hundreds of teachers and approximately 8,000 students worldwide.
REFERENCES
Bedell, D. (1993). Crosscultural variation in the choice of language learning strategies: A Mainland Chinese investigation with comparison to previous studies. Unpublished masters thesis, University of Alabama, Tuscaloosa, AL. Boraie, D., Kassabgy, O., & Oxford, R. (1994, March). Empowering teachers and learners: Style and strategy awareness. Paper presented at the 28th Annual TESOL Convention, Baltimore, MD. Dreyer, C. (1992). Learner variables as predictors of ESL proficiency among Afrikaansspeakers in South Africa. Unpublished doctoral dissertation, Potchefstroom University, South Africa. Ehrman, M. E., & Oxford, R. L. (1989). Effects of sex differences, career choice, and psychological type on adult language learning strategies. Modern Language Journal. 74, 311327. Green, J. (1992, October). Language learning strategies of Puerto Rican university students. Paper presented at the Annual Meeting of Puerto Rico Teachers of English to Speakers of Other Languages. Mullins, P. (1992). Successful English language learning strategies of students enrolled in the Faculty of Arts, Chulalongkorn University, Bangkok, Thailand. Unpublished doctoral dissertation, United States International University, San Diego, CA. Oh, J. (1992). Learning strategies used by university EFL students in Korea. Language Teaching, 1, 353. Oxford, R. L. (1986). Development and psychometric testing of the Strategy Inventory for Language Learning (SILL). ARI Technical Report 728. Alexandria, VA: Army Research Institute. (Appendices published as ARI Research Note 8692) Oxford, R. L. (1990). Language learning strategies: What every teacher should know. Boston: Heinle & Heinle. Oxford, R. L. (1993). Research on second language learning strategies. Annual Review of Applied Lingutitics, 13, 175187. Oxford, R. L., & Burry, J. A. ( 1993, April). Evolution, norming, and psychometric testing of Oxfords Strategy Inventory for Language Learning (SILL). Paper presented at the Annual Meeting of the National Council on Measurement in Education, Atlanta, GA. Oxford, R. L., & Burry-Stock, J. (in press). Assessing the use of language learning strategies worldwide using the Strategy Inventory for Language Learning (SILL). System, 73 (2). Oxford, R. L., & Nyikos, M. ( 1989). Variables affecting choice of language learning strategies by university students. Modern Language Journal, 73, 291300. Phillips, V. (1991). A look at learner strategy use and ESL proficiency. CATESOL Journal, 57-62. Watanabe, Y. (1990). External variables affecting language learning strategies of Japanese EFL learners: Effects of entrance examination, years spent at college/university, and staying overseas. Unpublished masters thesis, Lancaster University, Lancaster, England. Wildner-Bassett, M. (1992, November). Relationships among learning strategies, personality types, and learning styles. Paper presented at the Annual Meeting of the American Council on the Teaching of Foreign Languages, Chicago, IL. THE FORUM 171

VIRGINIA LoCASTRO
International Christian University
Rebecca Oxfords and John Greens reaction to my research report is to be commended as discussion and debate about research is a vital part of academic discourse. However, because of the limitations of space, I will focus on three main points with reference to the content of my report. First of all, I would like to underline the fact that the work I reported on represents only a small part of a larger project (in progress) concerning class size and the role of different classroom-related variables including such learner factors as learning strategiesin successful second language acquisition (SLA). This research is primarily ethnographic in nature and includes quantitative survey data. The Strategy Inventory for Language Learning (SILL) was administered in 1990 on the recommendation of a well-known ESL researcher, and I used it to try to begin to understand what learners in large classes do to achieve considerable success in SLA despite the less than facilitative learning environment. The report in the TESOL Quarterly (LoCastro, 1994), which is a short, seven-page text, was written to say something about the inconsistencies I found, not to criticize the SILL. Some of Oxfords and Greens comments reflect different assumptions about the merits of different research styles. From an ethnographic perspective, I have immersed myself in the sociocultural context under study in that I have lived and worked in Japan for over 10 years and have taught Japanese learners of English in a variety of situations during that time. I have read their writings and listened to their discussions about learning and studying languages. I have generally conformed to the tenants of the ethnographic research tradition, and it is in respect to this tradition that my work should be judged. Ethnographic research data collection and analysis techniques document patterns of values, attitudes, and/or behavior across a particular sociocultural community. Use of these techniques ensures both the validity (credibility) and reliability (dependability) of the study. Rather than assume generalizability across social and cultural settings, ethnographic research reports allow for readers to determine possible transfer of findings to other situations. The reader must test out in further research whether or not findings are transferable to the particular sociocultural context of interest (Davis, 1992). Thus, what I have found through ethnographic means provides equally valid evidence which, together with the SILL results, may help us better understand
s
172
TESOL QUARTERLY
what learners in a different learning environment from that of North America do. The fact that there are inconsistencies in the results leads me to further research, as I assume it would do for readers of my report. Within the qualitative tradition, a classification of learning strategiesor of anything else for that matterdeveloped for a particular learning environment cannot be brought without question and testing into a different learning context. Watanabe (1991) states it might be too early to prescribe a strategy classification which applies to any learner (p. 94). With regard to the second issue, Oxford and Green claim the SILL Version 7.0 is valid for both ESL and EFL settings. Reference is made to Oxford (1990), which has a section on differences between second and foreign languages. I have no quarrel with the distinctions which are offered; however, I do not agree with statements in Oxford (1990) such as the following: these differences occasionally (my emphasis) have implications for language learning strategies and most learning strategies can be applied equally (my emphasis) well to both situations (p. 6). Furthermore, Oxford (1990, p. 199) is referred to for information on the differences between Version 5.10 (for native English speakers learning another language) and Version 7.0 (for nonnative speakers learning English). I find the information presented on page 199 insufficient to address my concerns about how the 7.0 Version was developed. The fact that the versions have been used with learners of many different languages does not necessarily mean the classification of the learning strategies nor the different kinds of learning strategies on the inventory are the most appropriate for assessing what learners do to learn languages with the L1 and sociocultural backgrounds cited by Oxford. Given the ongoing debate about the relationship of language, thought, and culture, unlike Oxford, I am reluctant to claim that what learners do in one learning environment to learn is the same as or very similar to what learners do in other environments. There may be cognitive processing universals; however, as sociocultural practices pervade so many other areas of life, it is difficult to argue that they are absent with regard to learning strategies and processes. The lack of attention to differences in the learning environments can be seen in Oxfords and Greens comment that the SILL does include listening strategies. They suggest that any strategy involving speaking also requires listening. Clearly this is so; however, there have to be situations where the learners have frequent opportunities to speak the language. Moreover, Oxford claims that television viewing requires listening. Again, this is true, but there have to be Englishlanguage television programs. Relatively speaking, there are few opportunities to speak English and to watch English-language television
THE FORUM 173
programs in Japan. There are English-language movies; however, they are all subtitled in Japanese. Thirdly, I would like to react to the comment about my not having read any of the recent literature on the subject of learning strategies, specifically related to the use of Oxfords strategy inventory. Oxford attaches to her five-page response a reference list with 18 entries, 12 of which are unpublished masters theses or doctoral dissertations, papers read at conferences, published in a journal unavailable in Japan or marked in press. It is simply not possible for overseas researchers to have access to such materials. I would happily participate in collaborative research. However, although the call for collaboration is useful, researchers in Asia often find a patronizing attitude toward their work and their ideas. I am glad Oxford and Green are working to overcome this by actively working toward collaborative research. To conclude, my position is that there are different kinds of research which produce different results which may be of interest. Research dealing with human beings is notoriously fuzzy and shows a great deal of variation. I welcome that and see such variation as the norm. It is only through supportive collaborative research as well as supportive feedback to others that we can ever hope to begin to fill in the gaps on our journey to better understanding of how human beings learn and communicate in the world.
REFERENCES
Davis, K. A. (1992). Validity and reliability in qualitative studies. Another researcher comments. Research issues. TESOL Quarterly, 26, 605608. LoCastro, V. (1994). Learning strategies and learning environments. TESOL Quarterly, 28, 409414. Oxford, R. L. (1990). Language learning strategies: What every teacher should know. New York: Newbury House. Watanabe, Y. (1991). Classification of language learning strategies. International Christian University Language Research Bulletin, 6, 75102.
Comments on B. Kumaravadivelus The Postmethod Condition: (E)merging strategies for Second/Foreign Language Teaching
Alternative to or Addition to Method?
DILIN LIU
Oklahoma City University
s I read B. Kumaravadivelus article (Vol. 28, No. 1) with great interest and appreciation. The article is enlightening as well as thought pro174 TESOL QUARTERLY
voking. Although I applaud the authors fresh look at the conventional notion of method, and value his informative discussion of the macrostrategies for TESOL, I have strong reservations concerning his assumption about the futility/demise of the search for better methods and his proposal of a strategic framework as an alternative to method (p. 29). Kumaravadivelus call for a strategic framework is timely and well founded. A great deal of literature has focused on finding good teaching methods, which, no matter how useful they are, may never work for all situations nor for all students. As a result, we have overlooked the need for and the importance of an overall strategic framework that can guide us in our teaching practice and enable us to succeed with any class. Such a framework consists primarily of a series of universal ESOL teaching principles or what he terms macrostrategies (however, principle may be a more accurate term because Maximize learning opportunities, Minimize perceptual mismatches, and so forth are closer to principles than strategies). But no matter how important and useful they are, principles/strategies should not and cannot replace methods. Strategies defined in Kumaravadivelus terms and methods as we understand them, though they overlap to some extent, fall into two distinctive and complimentary domains. Celce-Murcia and McIntosh (1979), for example, clearly differentiate the two when they write that ESOL teachers should have the knowledge of teaching methods, background on and strategies for teaching the language skills (p. ix). That is, teachers will need both methods and strategies to successfully accomplish their instructional work in class. Methods, according to many experts (e.g., Krashen & Terrell, 1983; Richards & Rodgers, 1986), involve theories about language and teaching, procedures, and techniques. Krashen and Terrell (1983), for example, even equate methods with techniques: The series method advocated by Franois Gouin was perhaps the best known technique [emphasis added] used by the psychological methodologists (p. 10). Kumaravadivelus definition of method is similar because a method consists of not only a single set of theoretical principles derived from feeder disciplines but also a single set of classroom procedures (p. 29). These classroom procedures and teaching techniques are often unique because they are developed from the language/ learning theory(ies) that the method embraces. Methods so defined with their procedures and techniques are thus not the same as the universal macrostrategies that Kumaravadivelu introduces because the latter are principles teachers use to guide their classroom teaching procedures. As overall principles, strategies cannot take the place of specific methods/procedures/techniques because only through the latter can
THE FORUM 175
teachers accomplish their classroom objective. Kumaravadivelus argument for replacing methods/procedures seems to be that none of these methods can be realized in their purist form (p. 29), and none of the methods work for all students at all times. Although this may be correct, Kumaravadivelus reasoning for substituting methods is, I believe, flawed. Few things occur in their purist form. The fact that the methods cannot be realized in their purist form does not diminish their value. The fact that none of the methods work in all situations for all students does not mean that they are useless. On the contrary, these methods are invaluable because good teachers can select the best method for the specific situation. Total Physical Response (TPR), for example, has proven to be a very effective method for beginning students, children and adults alike. Grammar-Translation often works very well for a homogeneous class of educated adults who need only reading knowledge in the target language. The problems we have are not with the methods but with those who use the methods in the wrong place at the wrong time. A teaching method entails specific teaching activities. For example, TPR involves a series of activities centering around giving and following commands and modeling. The Silent Way uses colored rods and charts in activities. Suggestopedia involves the seance or concert session. For us to replace these activities with Kumaravadivelus strategies is like asking basketball players to replace their various techniques with a set of principles concerning how to use the techniques in a game. Thus, our job should not be to replace or to find an alternative to method. Instead, we should search for both effective strategic frameworks and good teaching methods. We should, on the one hand, empower ourselves with principles, theories, or macrostrategies, whatever they are called, that will enable us to make judicious decisions concerning the use of methods. On the other hand, we should continue our search for innovative methods and equip ourselves with them. An empowered teacher, as Richards and Rodgers (1986) point out, may constantly revise, vary, and modify teaching/learning procedures on the basis of the performance of the learners and their reactions to instructional practice (p. 19). For these empowered teachers to do so, however, they also need to know the various existing methods with their unique procedures and techniques. If Kumaravadivelus strategic framework does encompass these procedures/techniques/skills (included perhaps in his microstrategies, though we do not know that because he gives no definition nor real illustration of his microstrategies), he then should revise both his claim of a postmethod condition and his proposal of a strategic framework as an alternative to method. Rather than a postmethod condition, it should be a method redefining condition, and instead of being an
176 TESOL QUARTERLY
alternative to method, the strategic framework should be an addition to method, or to be more accurate, it should be a framework subsuming method.
REFERENCES
Celce-Murcia, M., & McIntosh, L. (Eds.) (1979). Teaching English as a second or foreign language. Rowley, MA: Newbury House. Krashen, S. D., & Terrell, T. D. (1983). The natural approach. Englewood Cliffs, NJ: Prentice Hall. Richards, J. C., & Rodgers, T. S. (1986). Approaches and methods in language teaching: A description and analysis. Cambridge: Cambridge University Press,

B. KUMARAVADIVELU
San JOS State University
I thank Dilin Liu for a critical reading of my article. His concerns are that (a) we should not stop searching for better and newer methods and (b) my strategic framework is not an alternative to method. Before I address these two concerns, a note on terminology is in order. It seems to me that some of the comments Liu has made can be attributed to the ambiguity surrounding terms such as approach, method, technique, design, and procedure, which are being used in the TESOL literature rather indiscriminately. In my article, I have tried to make a distinction between method as conceptualized by theorists and method as actualized by practitioners. In the postmethod condition, the term method refers to method as conceptualized by theorists. Such a method, in its idealized version, consists of a single set of theoretical principles derived from feeder disciplines and a single set of classroom procedures directed at classroom teachers (p. 29). Additionally, a method, so defined, should be able to guide and sustain various aspects of a language learning/teaching operation, particularly in terms of content (e.g., grammar, vocabulary), skills (listening, speaking, reading and writing), and levels (beginning, intermediate, and advanced). I have accordingly classified language teaching methods into three broad categories: language centered, learner centered, and learning centered methods. From such a perspective, it is doubtful whether one could even consider Total Physical Response (TPR), Suggestopedia, and The Silent Way as methods; to me, they are innovative classroom procedures that are consistent with the theoretical underpinnings of a learner-centered method.
s
THE FORUM
177
THE FUTILITY OF SEARCHING FOR METHODS

Liu strongly believes that my claim about the demise of method is premature. Luckily, as the bibliography cited in my article attests, I am not alone in making that claim. Nor is TESOL the only academic discipline that is fast realizing the limitations of method. Scholars in education have been drawing our attention to the methods fetish for quite some time (see Bartolom, 1994 for a recent argument). In fact, they have called for an antimethods pedagogy that refuses to be enslaved by the rigidity of models and methodological paradigms (Macedo, 1994, p. 8). Any perceptive reader of recent scholarly literature in TESOL would have noticed a new and fast emerging perspective that aims to propel L2 pedagogy beyond the limited and limiting boundaries of the traditional concept of method. The new perspective seeks to provide teachers and teacher educators with an alternative pedagogic tool. The push to go beyond the boundaries of method is evident in several recent publications meant for graduate students, practicing teachers, and teacher educators (e.g., Brown, 1994; Freeman & Cornwell, 1993; Richards & Lockhart, 1994). In addition to the terminological ambiguities, Lius commentary has some conceptual ambiguities as well. He concedes that methods no matter how useful they are, may never work for all situations nor for all students. He also argues that there is nothing wrong with any of the methods currently in use. He asserts that the problems we have are not with the methods but with people, with those who use the methods in the wrong place at the wrong time. If that is the case, why should we search for more innovative methods? Should we not search for more effective teacher education models? It seems to me that, for reasons discussed in my article, the search for method has reached a dead end. I do not anticipate any new method with an entirely new set of principles and procedures emerging any time soon. What we can anticipate is that for each of the currently used language-, learner-, and learning-centered methods, there will emerge newer and more innovative procedures, just as TPR, Suggestopedia, and The Silent Way provided new and creative procedures within the broad spectrum of a learner-centered method. In other words, for a long time to come, any new method can only be a variation of the familiar theme.
MACROSTRATEGIC FRAMEWORK AS AN ALTERNATIVE TO METHOD

The second concern of Lius is that my strategic framework is only an addition, not an alternative, to method. He has interpreted my
178 TESOL QUARTERLY
work exclusively in classical methodological terms whereas the very purpose of my article is to show that there are alternative ways of approaching L2 teaching. Unlike traditional methods, the strategic framework is theory neutral, that is, it is not constrained by any one specific theory of language, learning, and teaching. Unlike traditional methods, the framework is method neutral, that is, it is not confined to a particular set of theoretical principles and a particular set of classroom techniques associated with a particular method. Unlike traditional methods, the open-ended framework is not a theorists construct (something theorists invent and transplant into the classroom), but a practitioners construct (something classroom practitioners themselves should be able to generate from their classrooms). My article presents a case for restructuring teacher education so that prospective/practicing teachers can be provided with the knowledge, skill, and autonomy necessary to generate their own location-specific, classroom-oriented innovative macro/microstrategies. A sizable portion of Lius comments is based on his interpretation that my strategic framework is only about principles and not about procedures. In fact, the framework is conceived and constructed in terms of macrostrategies (principles) and microstrategies (procedures). Macrostrategies are broad guidelines derived from current theoretical, empirical, and pedagogical insights, and they are made operational in the classroom through microstrategies or classroom techniques (p. 32). Although the primary focus of my article is macrostrategies, I have nevertheless given examples of microstrategies. For instance, page 34 of my article contains a discussion of three microstrategies for one macrostrategy: Facilitating negotiated interaction.
CONCLUSION
What I have presented is no more than a basic frameworkone possible alternative to method. There can be others. Clearly, more exploration is needed. What is crucial is that teacher educators establish and maintain a good productive relationship with classroom teachers. Teachers already know a lot about classroom language learning, but have typically seen their knowledge marginalized by researchers who, if they collaborate with teachers at all, tend to want to do so on their own terms, pursuing their own research agendas, and following their own procedures, however irrelevant the former may appear to the classroom teacher, and however disruptive or time-consuming the latter (Allwright, 1994, p. 17). In short, we need to restructure teacher education so that teachers gain adequate knowledge, skill, and autonomy to continue to reflect, recreate, and reinvent.
THE FORUM 179
REFERENCES
A1lwright, R. L. (1994, October/November). Teachers as researchers: Un-mixing the metaphor? TESOL Matters, p. 17. Bartolom, L. I. (1994). Beyond the methods fetish: Toward a humanizing pedagogy. Harvard Educational Review, 64, 173194. Brown, D. H. (1994). Teaching by principles: An interactive approach to language pedagogy. Englewood Cliffs, NJ: Prentice Hall. Freeman, D., & Cornwell, S. (Eds.). (1993). New Ways in Teacher Education. Alexandria, VA: TESOL. Macedo, D. (1994). Preface. In P. McLaren & C. Lankshear (Eds.), Conscientization and Resistance (pp. 18). New York: Routledge. Richards, J. C., & Lockhart, C. ( 1994). Reflective teaching in second language classrooms. Cambridge: Cambridge University Press.
Teaching Issues
The TESOL Quarterly publishes brief commentaries on aspects of English language teaching. For this issue, we asked two testing specialists to address the following question: What would you like ESL teachers to know about test development?
Edited by BONNY NORTON PEIRCE Ontario Institute for Studies in Education
Language Tests and ESL Teaching

Examining Standardized Test Content: Some Advice for Teachers
FELICIA DeVINCENZI
Educational Testing Service
s Discussions
on test washback have often included the teacher as a cog in the washback mechanism. Although there has been some research on washback (see Hughes, 1988; Wall & Alderson, 1993), it is still unclear how the presence of an externally developed proficiency test influences teaching approaches and course content. Data collection in this area is complicated by the fact that variables such as pedagogical methods, levels of teaching expertise, and attitudes toward teaching and testing
180 TESOL QUARTERLY
are as diverse as the political, social, and economic environments in which they are embedded. For large-scale standardized tests like the Test of English as a Foreign Language (TOEFL)1 or the International English Language Testing System (IELTS)2 tests, EFL teachers participate in question writing, test assembly, review, and scoring, thus contributing substantively to the test development process. However, as individuals with direct knowledge of how such tests are developed, these teachers are relatively few in number. How the great majority of language teachers learn about standardized tests and draw conclusions about their content and purpose is therefore a highly inferential and potentially flawed process. This situation may stem from the testing industrys obscurity. Test development and statistical analysis are rather esoteric career paths chosen by a small number of subject specialists, and developing expertise in this area takes years of concentrated, specialized work experience. Therefore, these disciplines are not well understood by most teacher practitioners. Secondly, the maintenance of test security, with its restrictions on distribution and storage of test materials, contributes a secretive aura to testing.
Teachers as Informed Consumers

If standardized tests are to have a positive impact on learning, teachers need to evaluate these products as informed consumers. Teachers must be able to make accurate assumptions about test content in order to influence administrative decisions about test use and decide how to help students to perform at their best. Responsible test companies prepare descriptive materials about their tests and respond in a timely fashion to inquiries. Sample questions, a statement of test purpose and valid score uses, and a general content outline are usually provided upon request, free of charge. Often such information is sent automatically to program administrators, who tend to be the primary users of test scores for decision-making purposes; the information does not trickle down to classroom teachers, students, or their parents. Nonetheless, the classroom teacher can directly access
TOEFL is a multiple-choice test of listening, grammar, and reading and was designed to measure the English proficiency on nonnative-English-speaking students wishing to study at colleges and universities where English is the language of instruction. Educational Testing Service (ETS) of Princeton, New Jersey, USA, produces this test. 2 The IELTS is jointly managed by the British Council, the University of Cambridge Local Examination syndicate (UCLES), and the International Development Program of Australian Universities and Colleges (IDP) and administered by UCLES. IELTS assesses the English language proficiency of nonnative speakers who intend to study or train in the medium of English.
1
THE FORUM
181
this material by writing to the test producer. In many cases, as a service to the field, test developers will be happy to discuss the test development process at in-service workshops and professional meetings. It is important to have realistic expectations about the disclosure of test information, however. Although a standardized test is often the product of a nonprofit organization, the material is still proprietary. Also, especially in the United States, the testing industry is subject to public scrutiny. On the one hand, this accountability ensures that test products and procedures are accessible to candidates. On the other hand, it can engender a legal prudence that sometimes constrains sharing information with the general public.
Considering the Content of a Test Form

With descriptive information about the test in hand, the teacher should examine a form of the test if possible. If the teacher has the opportunity to study a test form, the following strategies might help shed light on the test design. Judge the relevance of the test content according to the purpose of the test. Many teachers arrive at false conclusions about a test by critiquing test content against their own pedagogical assumptions and materials rather than by considering the appropriateness of the tests content in light of the intended measurement purpose. Study a complete form of the test and formulate what the test specifications may look like. To help ensure that every form of a standardized test is parallel and comparable, each one is assembled according to the same blueprint, or test specifications. Test specifications constrain test content and list details such as categories for types of questions, the frequency at which particular question types should occur in each test form, and the level of difficulty and reliability of the items in the test. Consider the test as a whole. Although there may be subparts with companion subscores, a test is designed as one instrument, and decisions about a students abilities are most likely to be made based on overall score. For example, if a course focuses on improving reading skills, the teachers analysis should not be limited to the reading section of the test. Other subsections, such as grammar or listening, may be tapping skills that can be practiced in the context of the reading class. Try to infer the superordinate categories in the test specifications under which a particular question type might fall. A particular test question is often one of several possible ways to test a particular point, and one
182 TESOL QUARTERLY
test question can be used to satisfy more than one content category in the test specifications. For example, in a reading passage, asking the student to infer the authors purpose may be just one of many ways of testing comprehension of the gist of a passage and inferencing. When a student actually sits for the test, he or she may never have to answer an inference question about the authors purpose. Instead, there might be a different type of question testing the gist, and the students inferencing abilities might be tested by another question. Study the relative emphasis of skills and the types of texts and communicative functions represented in the stimulus material. Once one can see the broad skill categories under which different questions fall, the relative emphasis given to those skills can be estimated by examining how frequently they are tested. If a teacher wishes to choose particular materials for practice activities, it is helpful to consider the category of discourse under which particular test passages could be grouped. For example, in a business English test, a customer letter inquiring about the warrantee for a 35-mm camera might be satisfying two test specification categories pertaining to the discourse in the stimulus: Business Letter and Understanding Requests for Information. Value reliability as well as content validity. With the current emphasis on authenticity in testing, it is tempting for teachers as content specialists to embrace tests that look attractive and appear to test real skills with the right level of emphasis. For a comprehensive discussion of performance assessment and the concept of authenticity, see Messick (1994). However, unless reliability is built in, fairness cannot be ensured. For example, in the case of an essay test, your students essay will be judged against the essays of other students, most likely writing on other topics. In this type of test, steps need to be taken to ensure that (a) the writing tasks across forms are comparable in terms of skill tested, (b) writing tasks are equal in difficulty, and (c) essays are scored by trained raters with an acceptable level of interrater reliability. In writing and speaking tests, the consistency of ratings needs to be maintained across raters and across tests forms. As indication of this consistency, inter-rater reliability is a method of estimating the reliability of independent ratings of examinee performances. It consists of the correlation between different raters ratings of the same performances, adjusted by the Spearman-Brown Prophecy Formula. Especially for speaking and writing tests, the teacher should acquire as much information as possible with respect to test reliability and fairness. Satisfying reliability standards3 is a
In the United States, the American Psychological Association, the American Educational Research Association, and the National Council on measurement in Education share common standards, known as the Standards for Educational and Psychological Testing. The 1985 version is currently under review by the Joint Technical Committee to Revise the Standards.
3
THE FORUM
183
task responsible test producers keep well documented and reliability information is usually available for public inspection.
THE AUTHOR
Felicia DeVincenzi is an Examiner at Educational Testing Service and specializes in the assessment of aural/oral skills. She currently participates in planning, research, and development activities for the TOEFL 2000 project, a development effort to design a successor to the current TOEFL. In addition to computerbased testing issues, she is interested in teacher education and uses of multimedia technology for education.
REFERENCES
Wall, D., & Alderson, J. C. (1993). Examining washback: The Sri Lankan impact study. Language Testing, 10, 4169. Hughes, A. (1988). Introducing a needs-based test of English language proficiency into an English-medium university in Turkey. In A. Hughes (Ed.), Testing English for university study, ELT Documents 127 (pp. 134153). London: Modern English Publications. Messick, S. (1994, March). The interplay of evidence and consequences in the validation of performance assessment. Educational Researcher, pp. 1323.
Assessing Student Performance in the ESL Classroom

J. CHARLES ALDERSON and CAROLINE CLAPHAM
Lancaster University
s Many if not all language testers are, or have been at some stage, language teachers. However, although almost all teachers find themselves constructing classroom tests, the majority do not produce national or international standardized tests and would not think of themselves as language testers. It is this majority that we are referring to in this article. Many teachers view testing with either alarm or distrust because it is all too often associated with an arcane terminology, a heavy emphasis on numbers and statistics, and an aura of objectivity and rigor which makes people feel that testing is too difficult and that it needs to be left to experts. This attitude is often encouraged by testers, many of whom are more interested in high-stakes proficiency tests than in the low-stakes achievement tests that most teachers are concerned with. All testers focus on reliability (test consistency) and validity (test accuracy) and ways in which these concepts can be established and measured. However, both testers and teachers need to understand that concepts, procedures, standards and criteria that are relevant to na184 TESOL QUARTERLY
tional and international tests on which students futures depend are not necessarily appropriate for achievement tests and quizzes in the classroom. It is true that teachers will benefit from some understanding of what professional testing is all about if they want to understand their own tests more fully and to learn how to improve them. For example, insights into how item analysis can be carried out by busy teachers and how this can show which items worked well and which did not perform as expected are interesting and useful for teachers, provided the information is presented in as simple a way as possible. (See, e.g., Cohen, 1994, and Alderson, Clapham, & Wall, 1995). However, although the construction of major, standardized tests demands expertise in test construction and analysis, the construction of class-based tests requires less specialist knowledge and is related far more closely to the devising of class exercises. In principle there is no difference in design between an item for a classroom test and an item for an exercise. Both are intended to elicit student performance. In a test, however, the student works alone, without help from teacher or peers. This means that the test items themselves have to provide all necessary support. For example, the instructions have to be clear so that students know exactly how they should respond. (This is also true, of course, of exercises, but the results of any inherent ambiguity can be more easily remedied.) Designing test items, therefore, is more difficult than designing exercises, but it is possible to use exercises similar to those in textbooks or course materials as the basis for such items. The value of a test relates to what it is being used or misused for. Because an end-of-unit or end-of-week test is usually much less important for students than is an admissions test, its consequences being less dire, test reliability is less important. However, since students are likely to see such tests as reflecting the teachers view of what is important, the tests should obviously relate closely to the teaching. It is probably inappropriate, therefore, to use off-the-shelf tests for such purposes. Such tests might be more reliable than tests designed by the teacher, but would probably be less valid. For course-related, shortterm achievement tests, therefore, as with class exercises, teachers need to write items which correspond with their teaching methods, reflect their teaching objectives, and mirror in some way their teaching materials. They also need to make sure that such items are not ambiguous and that they produce the expected type of response. It is surprising how often students will interpret an item in quite a different way from that intended by the item writer. The best way of checking this is to try the draft test out on colleagues or a few students. Similarly, if teachers have to make subjective judgments about students writing or oral abilities, the marking criteria they use should
THE FORUM 185
reflect the views about language and language learning which they hold and which, possibly subconsciously, they refer to when talking to students about their work. Teachers, and indeed their students, may not be aware of exactly what is being taken into account when students work is assessed, and so developing explicit criteria for marking a test may not only improve the accuracy of the test scores, but may also have a useful effect both on the marking of exercises and the teaching itself. It is also helpful for teachers to have an understanding of what the reliability of ratings means. Although most teachers do not need to know about complex reliability formulas, an appreciation of the amount that different markers ratings of scripts or spoken performances can vary, and how much a single markers system of marking may change within the space of a few hours is invaluable. It is sobering for teachers to mark a set of scripts one day, and to re-mark them a couple of days later. In many cases the two sets of scores will differ widely, and the discrepancies in the marks will alert teachers to the impact that such variation will have on their students. One of the purposes of this article has been to remove some of the mystique surrounding language testing. After discussing the differences between classroom testing and standardized testing and the differences between what professional testers need to know and what teachers writing classroom tests need to do, we made three main points. First, test items and marking criteria should reflect teachers beliefs about language and language learning. Second, both exercises and tests should be carefully written so that students understand what is expected of them and the items elicit the required language behavior. Third, the criteria for marking written and oral tasks should be explicitly stated so that students know what is expected, and the marker can produce consistent results. These three requirements do not demand detailed knowledge about language testing but will nevertheless contribute to the production of valid and reliable information about student performance in the ESL classroom.
THE AUTHORS
J. Charles Alderson is Professor of Linguistics and English Language at Lancaster University and Immediate Past President of the International Language Testing Association. He is author, coauthor, and editor of numerous articles and books on language testing and reading in a foreign language and is currently directing research into the relationship between language proficiency, metalinguistics knowledge, and language aptitude. 186 TESOL QUARTERLY
Caroline Clapham, Lancaster University, recently completed a research project into the effect of background knowledge on English for academic purposes reading test performance. She is joint editor of Language Testing Update and chair of the International Language Testing Association Awards Committee.
REFERENCES
Alderson, J. C., Clapham, C. M., & Wall, D. (1995). Language test construction and evaluation. Cambridge: Cambridge University Press. Cohen, A. D. (1994). Assessing language ability in the classroom. Boston: Heinle & Heinle.
THE FORUM
187
BRIEF REPORTS AND SUMMARIES

The TESOL Quarterly invites readers to submit short reports and updates on their work. These summaries may address any areas of interest to Quarterly readers. Authors addresses are printed with these reports to enable interested readers to contact the authors for more details.
Edited by GRAHAM CROOKES and KATHRYN A. DAVIS

University of Hawaii at Manoa
Mainstream Classroom Teachers and ESL Students

NANCY CLAIR
University of Massachusetts
s The
number of ESL students in the U.S. is on the rise. Between 1985 and 1991, the ESL student population in K12 classrooms increased by 51.3 percent (Olsen, 1993) to approximately 2.3 million students (U.S. General Accounting Office, 1994). By the year 2000, the ESL student population will be increasing at two and a half times the rate of the general student population (U.S. Congressional Record, 1989). Presently, there are many programs for ESL students, but because of social, political, and economic factors, many ESL students spend only a portion of their day in the ESL or bilingual classroom. The rest of their day is spent in the mainstream classroom, yet classroom teachers are generally not prepared to integrate these students (Clair, 1993; Penfield, 1987; Wong Fillmore & Meyer, 1992). As the ESL student population increases, questions of mainstream classroom teachers ability to effectively instruct these students remain. The inclusion of multicultural education policies in teacher preparation programs (which include but are not limited to an understanding of linguistic and cultural diversity) are found in accreditation standards of the National Council for Accreditation of Teacher Education (NCATE) and guidelines of numerous professional education organizations. Even with multicultural standards and guidelines, Gollnick (1992) found that preservice teacher education programs may not be adequately preparing teachers to work with linguistically and culturally diverse students. In a study that focused on how institutions were doing in respect to NCATES accreditation standards, only 1 institution out of 59 had a preservice program in bilingual education, ESL, or multicultural education. Moreover, only 8 institutions were in compliance with NCATES multicultural education
189
standards. These findings are distressing because more than 50 percent of all public school teachers interact with ESL students (Penfield, 1987). Given demographic trends and the limits of preservice teacher preparation programs regarding ESL students, mainstream classroom teachers will continue to share in the education of ESL students without adequate preparation. In-service professional development, therefore, must provide an opportunity for mainstream classroom teachers to explore beliefs, pose questions, and gain new knowledge, skills, and attitudes with regard to ESL students. Underlying the need for ongoing professional development is the complexity of educating ESL students in the mainstream classroom and the notion that even expert classroom teachers are not necessarily effective with ESL students (Enright, 1986; Lucas, Henze, & Donato, 1990).
METHOD
To gain an understanding of the mainstream classroom teachers perspectives regarding ESL students, I conducted a year long qualitative study exploring the beliefs, self-reported practices, and professional development needs of three mainstream classroom teachers with ESL students (Clair, 1993). I focus hereon one aspect of the larger study: mainstream classroom teachers professional development needs concerning ESL students. I chose this focus because of an increased emphasis in many states on in-service professional development for mainstream classroom teachers with ESL students. For example, Florida and California have state mandates that require mainstream classroom teachers to be trained in ESL issues; Massachusetts recertification includes training for mainstream classroom teachers regarding ESL. I undertook case studies of three mainstream classroom teachers (Grades 4, 5, and 10) based on interview transcripts, notes from classroom observations, and our journals. Anita, (all names are pseudonyms) a 40year-old classroom veteran with 20 years teaching experience, teaches fourth grade in a suburban elementary school. She has two ESL students in her class. Laura, 27, teaches fifth grade in an urban elementary school. This is her first year in the classroom, and she has four ESL students. Joshua, 27, teaches ninth and tenth grade global studies and history in an urban high school. He has been teaching for 6 years. With 160 students per day (typical numbers for an urban high school teacher), Joshua estimates that 40 to 50 percent are ESL students. Interview questions elicited previous professional development opportunities that focused on essential skills and knowledge necessary to instruct ESL students and suggestions for future in-service professional development regarding ESL.
RESULTS
In discussing with these teachers their previous professional development opportunities regarding ESL, I found that Anita was the only teacher
190 TESOL QUARTERLY
who attended in-service workshops; her district provided 12 in a row. However, Anita expresses dissatisfaction with these workshops:
It wasnt helpful to me because of time. Give me stuff. Give me a goody bag. You can use this with your fourth grade students who dont speak English. I will use it.
Joshua never attended in-service ESL. Nevertheless he comments:

You get these know-it-all professors telling you whats effective and whats not. Weve spent $80,000.00 on this study. But I ask what did you get in the kids hands. It doesnt take a rocket scientist to figure out what these kids need. They need stuff they can read that corresponds to what they are learning.
Finally, Laura did not attended in-service staff development ESL workshops because the particular session did not fit her needs:
With the teaching of ESL it was through puppetry. And I really didnt see that it was useful for me now. I didnt see that I was going to bring puppets in the classroom. So if it were different, if it were ESL with other kinds of materials. Something that I thought that I would use this year.
Both Anita and Joshua commented that they would rather have materials and support as opposed to in-service training if given the choice. Joshua wants bilingual vocabulary lists and truly bilingual materials. For Joshua that means text in the native language with corresponding translation in English. Anita also wants bilingual vocabulary lists that she feels would lessen the confusion for students.
The thing with Luz, she had the confusion with those vocabulary terms. I think that three or four worksheets that were designed by an ESL person who knows the approach to the stages of language development. I was not aware that in Spanish milos [sic] is thousand. Now I can see where that would cause confusion.
Anita believes that overall she has the skills and knowledge necessary to instruct ESL students:
For myself personally, I think I have enough experience as a classroom teacher that I know already. I buildup their culture, boost their self-esteem. I think that all the things that apply to any good teacher apply to ESL.
Joshua concurs:
I mean as far as teaching goes, teaching is the same no matter what kinds of kids you have. Its really true. It doesnt matter what Im teaching. If youre a teacher, youre a teacher . . . . So I mean for myself, I do OK. BRIEF REPORTS AND SUMMARIES 191
When asked about suggestions for future professional ESL development opportunities, Anita and Joshua re-emphasized their desire for appropriate materials. Responding to the fact that teachers do not have time to communicate, Laura suggests lunchtime workshops. She imagines
hearing about other problems that other teachers are having and ways that they have dealt with it. Strategies that they use. You could even hear about other projects that are being done in other classrooms.
Anita, Joshua, and Lauras cases illuminate two essential problems. The first is teachers desire for quick fixes in terms of materials and curricular ideas. How have teachers come to believe that quick fixes will solve complex educational problems? Part of the answer lies in teacher education and socialization. Most U.S. preservice teacher education programs stress technical conceptions of teacher competence as opposed to a more critical approach to teaching and learning (Howey & Zimpher, 1989). Technical conceptions, such as Competency-Based Teacher Education (CBTE) programs, promote the notion that curriculum content is given as opposed to constructed, and they stress skill development and demonstration over critical questioning and problem solving. The hidden message in technical conceptions of teacher education is that teachers are merely implementers of instructional systems; there is no room for teacher questions, decisions, or the generation of knowledge. Technical conceptions serve to de-skill many teachers because they were never taught to make instructional decisions or directed toward viewing decision making as an integral part of their role (Ginsburg, 1988). In addition to the inadequacies of many preservice teacher education programs, the accountability movement contributes to the de-skilling of teachers (Benveniste, 1985) and perpetuates teachers desire for quick fixes. Accountability, the need to hold schools responsible for the perceived crisis in education, will continue as long as schools receive state and federal funding. The main tool for accountability is the standardized test. And although standardized tests are not necessarily linked to what is actually taught, test scores are used to rank, reward, or most often punish students, teachers, and schools. The power of such high stakes tests is that they begin to drive the curriculum; that is, if teachers are held accountable for their students and if accountability is linked to funding and measured by standardized achievement tests, then teachers will be pressured to teach to the test. The result is, once again, a given curriculum (in this case driven by standardized tests); there is no need for teachers to co-construct knowledge with their students, question, or make decisions about what is to be taught. Teaching is routinized and teacher discretion is reduced (Benveniste, 1985). Add one-shot professional development workshops (like the ones that Anita, Joshua, and Laura experienced) that may or may not address issues pertinent to mainstream teachers and the deskilling and de-professionalization is perpetuated. The second problem illustrated by Anita, Joshua, and Lauras case histories is captured in the teachers statement that good teaching is good
192 TESOL QUARTERLY
teaching. This statement demonstrates the need not only for mainstream teachers to understand second language acquisition but also for mainstream teachers to change their beliefs, values, and attitudes toward ESL students. Specifically, this statement denies the usefulness of specialized knowledge concerning L2 development and ESL students (Clair, 1993). Second, it negates the importance of individual differences in second language acquisition (Ellis, 1986; Larsen-Freeman & Long, 1991) and learning in general. Third, it fails to acknowledge the complexities of the social and academic integration of ESL students in mainstream classroom settings (Penfield, 1987). The difficulties inherent in changing teachers attitudes have been documented in a number of studies. Sleeter ( 1992) studied public school teachers who attended a multicultural education staff development program over a 2-year period. Sleeter found that although many of the participating teachers perceived that they had learned much, there was little change in their attitudes and practice. Ahlquist (1992) noted that teacher attitudes and beliefs remained unchanged for the most part during a multicultural foundations course. If relatively long-term professional development opportunities and semester courses fail to change teacher beliefs and attitudes toward ESL students, then it is no surprise that teacher workshops are unsuccessful. McDiarmid (1990) studied teachers attitudes toward ESL students both before and after a 3-day workshop designed to influence these attitudes and found that the multicultural presentations had little influence on the teachers beliefs about ESL students.
AN ALTERNATIVE APPROACH
To avoid teachers inclination for seeking quick fix solutions to complex educational problems, alternatives to short-term workshops are clearly needed. Mainstream teachers need ongoing opportunities to reflect on nonmainstream student issues because educating ESL students is complex; it challenges social, political, and pedagogical assumptions; it is context specific and dilemma ridden. Furthermore, practicing teachers are not empty slates; they have intuitive knowledge, varied experiences, and professional needs. Finally, considering the forces that de-professionalize teachers, ongoing professional development that is constructed by the teachers themselves has the capacity to individually empower and socially transform (Cochran-Smith & Lytle, 1993; Ginsburg, 1988; Lather, 1991; Willet & Jeannot, 1993). Teacher study groups provide a format for teachers to shape their own professional development opportunities through problem posing and critical reflection. Critical reflection and problem posing convey an image about the relationship that teachers have to knowledge (Cochran-Smith, 1991). Through reflection and problem posing, teachers intimately discover the complex dynamics of teaching and learning and see themselves as critical players in the process. The purpose of critical reflection is for teachers to understand the complex dynamics of their situation and acBRIEF REPORTS AND SUMMARIES 193
quire a significant voice in the transformation of their community (Willet & Jeannot, 1993). For example, mainstream teachers with concerns regarding ESL students could form a study group. They could generate specific questions individually or within the group; read relevant materials; explore their classrooms, school, or community; and consult other teachers (ESL, bilingual) or an outside specialist. The point is that there is no set format for teacher study groups as they are driven by the teachers themselves. The notion that teachers take responsibility for their professional development within teacher study groups is problematic given the ways that teachers are socialized. Resistance is inevitable (Ahlquist, 1992; Lather, 1991; Willet & Jeannot, 1993) because teacher study groups are a radical change from one-shot after school workshops. Resistance may manifest itself in indifference, protest, or subversion (Willet & Jeannot, 1993). Regardless of how it is manifested, resistance needs to be named, understood, and critiqued as it may ultimately serve as a productive tool in challenging the de-skilling structures of the profession. The potential for teacher reflection and problem posing, key characteristics of teacher study groups, emerged during the study. All three teachers, without explicitly being asked, commented on the interview process, began to pose questions and engage in dialogue with me (the researcher) about ESL students and the greater societal influences. Specifically, Anita called me twice to discuss an incident that occurred with an ESL student; Laura asked for specific materials focusing on second language acquisition. Joshua mailed mea student composition that he felt was especially interesting. In addition, he commented on the process of talking about ESL students during the interviews.
Anytime I speak to someone, it forces me to re-evaluate and make assessments and adjustments . . . the more that I hear myself say things, the more that I will be conscious to do them.
These examples demonstrate that teachers respond to opportunities to discuss and pose questions around issues that impact their classrooms. Teacher study groups provide this opportunity.
CONCLUSION
This qualitative investigation of three classroom teachers beliefs regarding the teaching of ESL students suggests that with inadequate teacher preparation and nonexistent or inappropriate professional development, these mainstream classroom teachers are learning to educate ESL students on the job. The fact that teachers indicated their desire for specific ESL materials over professional development suggests two problems: Teachers tend to desire easy answers to complex educational problems, and teachers not only lack understanding of second language acquisition but also the attitudes which would facilitate ESL student achievement. In order for mainstream teachers to gain the knowledge, skills, and attitudes necessary
194 TESOL QUARTERLY
to socially and academically integrate ESL students in the mainstream classroom, I suggest ongoing teacher study groups as opposed to traditional one-shot staff development workshops. Teacher study groups, comprising critical reflection and problem posing, provide an in-depth opportunity to explore complex issues and may serve as a catalyst for individual empowerment and social transformation.
REFERENCES
Ahlquist, R. (1992). Manifestations of inequality: Overcoming resistance in a multicultural foundations course. In C. A. Grant (Ed.), Research and multicultural education (pp. 89105). Washington, DC: Falmer Press. Benveniste, G. (1985). The design of school accountability systems. Educational Evaluation and Policy Analysis, 7, 261279. Clair, N. (1993). Beliefs, self-reported practices and professional development needs of three classroom teachers with language-minority students. Unpublished doctoral dissertation. Teachers College, Columbia University. (ERIC Document Reproduction Service No. ED 365 166) Cochran-Smith, M. (1991). Learning to teach against the grain. Harvard Educational Review. 63, 279310. Cochran-Smith, M., & Lytle, S. L. ( 1993). Inside outside: Teacher research and knowledge. New York: Teachers College Press. Ellis, R. (1986). Understanding second language acquisition. New York: Oxford University Press. Enright, D. S. (1986). Using everything you have to teach English: Providing input to young language learners. In P. Rigg & D. S. Enright (Eds.) Children and ESL: Integrating perspectives (pp. 113162). Washington, DC: TESOL. Ginsburg, M. (1988). Contradictions in teacher education: A critical analysis. Washington, DC: Falmer Press. Gollnick, D. M. (1992). Multicultural education: Policies and practices in teacher education. In C. A. Grant (Ed.), Research and Multicultural Education (pp. 218 239). Washington, DC: Falmer Press. Howey, K. R, & Zimpher, N. (1989). Profiles of preservice teacher education. Albany, NY: SUNY Press. Larsen-Freeman, D., & Long, M. (1991). An introduction to second language acquisition research. New York: Long man. Lather, P. (1991). Getting smart: Feminist research and pedagogy within the postmodern. New York: Routledge. Lucas, T., Henze, R., & Donato, R. (1990). Promoting the success of latino language-minority students: An exploratory study of six high schools. Harvard Educational Review, 60, 315340. McDiarmid, G. W. (1990). Prospective teachers views of diverse learners: A study of the participants in the ABCD project. East Lansing, MI: National Center for Research on Teacher Education. Olsen, R. E. W-B. (1993, March). A survey of LEP and adult ESL enrollments in US public schools. In Language minority student enrollment data. Symposium conducted at the 27th Annual TESOL Convention, Atlanta, GA. Penfield, J. (1987). ESL: The regular classroom teachers perspective. TESOL Quarterly, 21, 21-39. Sleeter, C. E. (1992). Keepers of the American dream: A study of staff development and multicultural education. Washington, DC: Falmer Press. BRIEF REPORTS AND SUMMARIES 195
United States Congressional Record. (1989). Proceedings and debates from the 10lst congress. First Session, 135(25) Washington, DC: Author. United States General Accounting Office. (1994). Limited English proficiency: A growing and costly challenge facing many school districts. (GAO/HEHS 9438). Washington, DC: Author. Willet, J., & Jeannot, M. (1993). Resistance to taking a critical stance. TESOL Quarterly. 27, 477-495. Wong Fillmore, L., & Meyer, L. M. (1992). The curriculum and linguistic minorities. In P. Jackson (Ed.), Handbook of research on curriculum (pp. 626658). New York: MacMillan. Authors address: Nancy Clair, 95 Jackson Street, Cambridge, MA 02140.
Erratum On page 809 of Janet Anderson-Hsieh and Horabail Venkatagiris article, Syllable Duration and Pausing in the Speech of Chinese ESL Speakers (Vol. 28, No. 4), the first line of Paragraph 3 under Results should read: shows only clause boundary pauses rather than shows only word boundary pauses. Also, in the reference list on page 811, Anderson-Hsieh, J. (1992) should read Anderson-Hsieh, J., Johnson, R., & Koehler, K. (1992). We regret these errors.
196
TESOL QUARTERLY
REVIEWS
The TESOL Quarterly welcomes evaluative reviews of publications relevant to TESOL professionals. In addition to textbooks and reference materials, these include computer and video software, testing instruments, and other forms of nonprint materials. Edited by H. DOUGLAS BROWN San Francisco State University
How Languages Are Learned.

Patsy M. Lightbown and Nina Spada. Oxford: Oxford University Press, 1993. Pp. xv + 135. and Spadas How Languages Are Learned (hereafter HLL), part of the Oxford Handbooks for Language Teachers series, provides an introduction to (mostly second) language acquisition research for second and foreign language teachers and is the result of the authors 15 years of experience addressing issues of classroom research for such an audience. HLL covers theories of both first and second language acquisition, individual difference in second language acquisition (SLA), L1 and L2 development, and instructed SLA. Because the text was developed as the result of a series of teacher-education workshops, the presentation is concise and focused, and manages to communicate the most central aspects of a complex field in such a way that research on language acquisition is made both relevant and accessible. The introduction of HLL includes a brief questionnaire designed to assess various (misconceptions readers might have about language acquisition. Inclusion of the questionnaire and other activity sections in Chapters 25 reflects the books workshop origins, which is useful in teacher-education contexts. However, it is worth pointing out that whereas only 1 of the 12 questionnaire items deals exclusively with first language acquisition (FLA), 7 deal with instructed SLA, which is a good indication of the emphasis of the text as a whole. Chapter 1 deals with FLA theories and covers the well-trodden ground of behaviorism, innatism, the critical period hypothesis, and child-directed speech. HLL does an excellent job of introducing most of the important issues in clear and accessible terms (as is the case throughout the text), but noticeably absent are several topics which generally receive coverage in introductory-level FLA texts (e.g., Berko197
s Lightbrown
Gleason 1989; Menyuk, 1988), such as Piagetian cognitivism, learnability, and information-processing approaches. Although HLL does refer to the evolution of generative approaches to FLA, one does not come away with a clear idea of the sweeping nature of such changes or of the extensive literature on Universal Grammar (UG) and FLA that has been produced (see, e.g., Goodluck, 1991; Hyams, 1986; Roeper & Williams, 1987). Chapter 2 covers SLA theory. The most space is allotted to a critical assessment of Krashens (1982) contributions to SLA theory, with behaviorism (via the contrastive analysis hypothesis) taking a close second. Interactionism and cognitivism are given rather short shrift through cursory mention of Longs (1985) modified interaction hypothesis and restructuring, respectively. No mention is made of other theories which usually come up in discussions of this kind (e.g., pidginization, acculturation), but an even more significant omission is that HLL fails to cite any of the work on SLA being done in a UG framework (see, e.g., Eubank, 1991; Gass & Schacter, 1989; White, 1989). Although perhaps not directly relevant to classroom practice, this growing body of research merits at least a passing reference. In fact, enough research on UG in SLA has been carried out to warrant the publication of at least one introductory SLA text with UG as its main focus (see, e.g., Cook, 1993). Chapter 3 successfully covers the subject of individual differences (IDs) in SLA. Again, there is a brief questionnaire at the outset of the chapter in which readers are to assess the importance of various characteristics/activities to being a good language learner, squarely placing the work on IDs in this tradition (where much, but not all, of it belongs). Topics include intelligence, aptitude, personality, motivation/ attitudes, learning styles, and a lucid discussion of age, which takes up more than half of the chapter. Although there are some good reasons for devoting so much space to age (one being that it is perhaps the sole ID which does not engender controversies of definition and measurement), the amount of space it receives may not be representative of the amount of research conducted on age compared with other IDs. Chapter 4 addresses the issue of learner language, and (in addition to the first chapter) represents the texts only other treatment of FLA. Again the standard topics are covered, with a useful juxtaposition of L1 and (mostly naturalistic) L2 development of grammatical morphemes, negation, interrogatives, and relative clauses. Although this approach underscores the similarities between FLA and SLA, HLL avoids falling into the L1 = L2 trap by briefly noting that the role of the L1 in SLA cannot be dismissed. Helpful for teachers in this chapter is a taste of
198
TESOL QUARTERLY
error analysis in the form of one paragraph each from a Frenchspeaking child and Chinese-speaking adult learning English. This activity could have been made even more useful if particular examples of L1 transfer had been discussed for each text, as were developmental errors. Instructed SLA is the topic of Chapter 5, and this is where the strengths of HLL are most evident. Given the important contributions that Lightbown and Spada have made (individually and jointly) to this area, this should come as no surprise. Treatment is thorough indeed the chapter takes up over a third of the entire text. Of particular value is the manner in which research is presented, making findings more accessible to language teachers than other introductory SLA texts. The bulk of the discussion is organized to present the various theories of/ approaches to instructed SLA as proposals for classroom teaching (e.g., behaviorism is characterized as Get it right from the beginning, whereas Krashens Natural Approach is referred to as Just listen). The result is that HLL succeeds in making the relevance of SLA research to the classroom as clear as it could be. Also useful is this chapters activity, which consists of classroom transcripts comparing audiolingual and communicative approaches to illustrate differences in areas such as error correction and negotiation of meaning. And there are at least twice as many suggestions for further reading at the end of this chapter than the other chapters. Chapter 6 frames the text nicely by addressing each of the 12 items which appeared on the questionnaire in the introduction. With a brief statement based on research relevant to each item, this chapter makes a final attempt to push readers past popular misconceptions concerning language learning (e.g., that languages are learned mainly through imitation). Although HLL is clearly well suited for teacher education, it is rather unfortunate that the books back cover claims it provides a comprehensive and readable introduction to first and second language acquisition. Readable, yes, but comprehensive, surely not. At a slim 135 pages, it is hard to imagine how it could be comprehensive for both FLA and SLA, especially compared with other texts which restrict themselves to one or the other, such as Ingram (1989), a hefty 572 pages, or Ellis (1994), a massive 824 pages. For those who need a comprehensive FLA/SLA text, HLL is not it. But despite this unfortunate claim, for the right audience, HLL has much to offer. That is, for those who need a teacher-education text which presents thumbnail sketches of issues in language acquisition research not directly relevant to the classroom, and a good dose of what is most relevant to language teachingresearch on instructed SLA HLL is an outstanding text.
REVIEWS
199
REFERENCES
Berko-Gleason, J. (Ed.). (1989). The development of language (2nd ed.). Columbus, OH: Merrill. Cook, V. (1993). Linguistics and second language acquisition. New York: St. Martins Press. Ellis, R. (1994). The study of second language acquisition. Oxford: Oxford University Press. Eubank, L. (Ed.). (1991). Point counterpoint: Universal grammar in second language acquisition. Amsterdam: John Benjamins. Gass, S., & Schacter, J. (Eds.). (1989). Linguistic perspectives on second language acquisition. Cambridge: Cambridge University Press. Goodluck, H. (1991). Language acquisition: A linguistic introduction. Oxford: Blackwell. Hyams, N. (1986). Language acquisition and the theory of parameters. Dordrecht, Netherlands: Reidel. Ingram, D. (1989). First language acquisition. Cambridge: Cambridge University Press. Krashen, S. (1982). Principle and practice in second language acquisition. Oxford: Pergamon. Long, M. H. (1985). Input and second language acquisition theory. In S. Gass & C. Madden (Eds.), Input in second language acquisition (pp. 377393). Rowley, MA: Newbury House. Menyuk, P. (1988). Language development. Glenview, IL: Scott, Foresman. Roeper, T., & Williams, E. (Eds.). (1987). Parameter setting. Dordrecht, Netherlands: Reidel. White, L. (1989). Universal grammar and second language acquisition. Amsterdam: John Benjamins.
KENNETH R. ROSE
Hong Kong Baptist University
The Discovey of Competence: Teaching and Learning with Diverse Student Writers.
Eleanor Kutz, Suzy Q. Groden, and Vivian Zamel. Portsmouth, NH: Heinemann Books, 1993. pp. xii + 204.
s Although The Discovery of Competence is not written specifically for
an ESL audience, it is a valuable resource for ESL writing teachers. Informed by the research and pedagogical developments in language acquisition and in the writing processes of ESL and native speaker (NS) students, the authors begin with the assumption that all students come to the classroom as competent speakers of a first language, and perhaps of others (p. 17). Building on the concept of student competence rather than student deficit, the authors present the results of their 10-year collaboration
200 TESOL QUARTERLY
in developing a successful writing program at the University of Massachusetts at Boston. The authors wide range of experiences working with adult ESL and NS writers adds to the integrity of their text. They include examples of their ESL and NS students writing and conversations to illustrate how writers write and think. Kutz, Groden, and Zamel present a language-based curriculum framed by the theory that language is socially constructed. They have also developed a theory of thinking that is meant to enable adult students to become more active learners. The first chapter sets out the authors underlying assumptions that language competencies can be built within an academic context and that academic discourse can and should be learned. The next two chapters focus on describing two aspects of competencelanguage acquisition and development of thought. Regarding language acquisition, they make the point that language is embedded in and learned through participation in a community. The authors are concerned that traditional writing instruction too often reduces the whole of writing to an assemblage of structures and removes it from a socially constructed context that gives it value, purpose, and meaning (p. 30). They feel that writing teachers should move from the discourse of error and deficit to the discourse of learning and possibility (p. 61). In response, the authors advocate a language acquisition-oriented classroom. They define this as one that provides a medium for active learning and for the trial-and-error process in which students test out hypotheses that help them develop a concept of global discourse features as well as an awareness of surface features in their writing. One of the values of the book is the clear description of how to do this in a college writing class. According to the authors, an acquisitionoriented classroom relies on the power of actively engaging students in community projects. In their view, language needs to be used to make meaning and to fulfill genuine purposes. The second aspect of competence they describe is the development of thinking. The authors succinctly present the developmental theories of Piaget, Perry, Kohlberg, Vygotsky, Bakhtin, Scollons, Gilligan, and Bruner and explain how they can be applied to the writing classroom to promote intellectual growth. The authors also position their students stories and ideas and their own in the context of these developmental theories. Then by examining carefully the kinds of talking and writing their ESL and NS students do through the stories they tell, the authors analyze the adult thinking process. They categorize three types of thinking and suggest methods that will enable students to shift from one to another. The authors propose that adults engage in analytical, dialectical, and metaphorical thinking. Analytical thinking is especially useful for
REVIEWS 201
categorizing, for science courses, and for developing political arguments. Dialectical thinking refers to the ability to see one position in relation to another, to see different perspectives, and to tolerate ambiguities, contradictions, and conflict. Metaphorical thinking is interpretive and reflective. It is problem finding rather than problem solving and helps thinkers generate new questions outside the context of their own immediate experiences. These three types of thinking are important for ESL teachers and researchers because they may provide a focus for assessing ESL writing that is outside of the traditional examination of structure and form. The authors also suggest ways that these types of thinking can be taught. In the fourth chapter, the authors describe some of the types of evidence that indicate students increased abilities in acquiring and shifting through these types of thinking. They also briefly mention features to look for in assessing whether students are beginning systematically to approximate academic discourse, such as increased numbers of nominalizations, infinitive phrases, and the use of particular words and phrases to create logical relationships. One weakness of the book is the shortness and lack of depth in this chapter. The authors do not examine their assumption that academic discourse is a stable entity that can and should be emulated. They do not discuss the problems that can emerge from their own acknowledgment that students may actually begin to make more surface errors as they move toward academic discourse. In Chapter 5, Kutz, Groden, and Zamel provide detailed descriptions of assignments and curricula that have been effective in their ESL and NS writing classes. They believe that a curriculum should be participatory and encourage students to ask questions that may not be easily answered; it should require authentic inquiry. In Chapter 6, the authors present a project that fulfills their own requirement for authentic inquiry. They describe a language-focused project in which students do ethnographic studies of language use in their families, home communities, and/or the academic community. For this project, students make observations, tape and transcribe material, group and sort data, construct general understandings based on their data, support their general conclusions with evidence, and report on their findings. Students learn to reflect on their experiences and to analyze their reflections in their groups. Through working on these projects, participants learn how to do meaningful research and develop shared knowledge. Kutz, Groden, and Zamel believe in the power of stories to transform and to understand experiences. In Chapter 7, the authors examine some of their students stories and illustrate methods which enable students to analyze their own work. Students learn to create a literature
202 TESOL QUARTERLY
based on their own families and experiences and then to analyze what writing and discussing this literature has taught them. Students work together to discover commonalities and diversity in the ways that they tell stories and in the values expressed in their stories. Chapter 8 deals with the difficult problem of assessment. The authors begin this chapter by describing how teachers typically respond to students texts. Zamels earlier research (1985) into teachers responses to ESL students texts showed that teachers often responded primarily to surface-level features and were inconsistent in their reactions and many times offered vague prescriptive. Teachers rarely offered specific strategies to help students re-examine their writing. As a response to this, the authors describe their method of developing writing conversations between students and teachers about specific pieces of writing. The conversations focus on describing strategies and in making concrete suggestions to help individual students improve their writing. These conversations are also meant to help students see how they are progressing as learners and how they can become effective participants (p. 145). The conversations model what teachers expect from their studentsthey are explicit, defining terms precisely and presenting adequate supporting data. The authors readily admit, however, that although there have been problems in teachers responses to writing, most of the conflicts in assessment occur between teachers and the institutions in which they teach. The authors acknowledge the conflict between the model of learning they propose and their own institutions model of assessment that relies on a singlesample essay. They describe the criteria for proficiency in writing at University of Massachusetts/Boston and explain that although they do not always agree with it, they have made an attempt to translate the colleges criteria into forms of engagement with the question, the texts, and the reader. The authors have entered a conversation about writing assessment at their school and are attempting to move the assessment procedure from a single-sample essay format to one including various types of writing and student portfolios. Chapter 9, the final chapter in the book, is made up of personal stories written by each of the writers. These stories support their commitment to recognizing diversity and to building a multicultural academic community. This book brings together theories of language acquisition, the social construction of knowledge, and developmental theories of thinking in a new and productive way that will be useful for ESL teachers and teacher educators. Although not specifically an ESL book, it includes many examples of ESL writing and issues. In addition, The Discourse of Competence has an important message to impart: ESL students bring a lifetime of experiences, competencies, and strengths to the classroom and the institution needs to offer them a
REVIEWS 203
chance to express them. By writing this book, Kutz, Groden, and Zamel have concretely illustrated the power of collaboration to create a text that allows for both the expression of individual diversity and unity of purpose. The writers reveal their own process in action as they worked together for 10 years to create a writing program that honors the pluralism of ESL and NS writers and provides genuine learning experiences for all students. The clarity of explanations and the presentation of specific models and methods in a theoretical framework make this book useful for novice as well as experienced ESL writing teachers and teacher trainers.
REFERENCE
Zamel, V. (1985). Responding to student writing. TESOL Quarterly, 19, 79101.
TRUDY SMOKE
Hunter College, City University of New York
The Other Tongue: English Across Cultures. (2nd ed.).

Braj B. Kachru (Ed.). Champaign, IL: University of Illinois Press, 1992. pp. XXV + 384. s In 1981, in the preface to the first edition of The Other Tongue, Charles Ferguson noted that the volume had begun outlining aspects of a theory of English language variation and spread. He suggested that the development of such a theory could make a tremendous contribution toward human self-understanding and thoughts about possible futures for the inhabitants of the planet (p. xv). In the decade and a half since this was written, the field of World Englishes has indeed expanded, and to a large degree, Fergusons predictions have come to pass. It is thus appropriate to revisit issues addressed in this second edition of The Other Tongue. The text is dedicated to the memory of Peter Strevens, a consummate professional language educator and scholar known to many for his commitment to helping teachers explore the primary causes for success and failure in language classrooms around the world. The foreword to the second edition was written by Strevens, a close friend and professional colleague of Braj and Yamuna Kachru, just prior to his untimely death. This volume is marked by a spirit of cooperation and collegiality that permeates the contributions. It is clearly a text which ought to be in all libraries serving TESOL students and scholars and on the personal reference shelves of those interested in the field of World Englishes.
204 TESOL QUARTERLY
For this second edition, Kachru has chosen carefully, retaining critically important articles from the first edition, including relevant new articles, and performing his editorial work with the same care demonstrated in the rest of his scholarship. This volume contains 19 articles, 8 of which are new. Nine articles from the original edition have been deleted, although in four cases, their authors have contributed new articles, allowing readers to see where their theory building has gone. The text is divided into six parts: (1) English in the Global Context: Directions and Issues; (2) Nativization: Formal and Functional; (3) Contact and Change: Question of a Standard; (4) Literary Creativity in the Other Tongue; (5) Discoursal Strategies: Text in Context; and a new section, (6) World Englishes in the Classroom: Rationale and Resources. There are a number of ways to read this volume. The first is as an anthology which provides a theory base for the current state of World Englishes research. The second is as a source for models for research. For example, papers by Larry Smith, Spread of English and Issues of Intelligibility, and Yamuna Kachru, Culture, Style, and Discourse: Expanding Noetics of English, provide sound examples of how to conduct empirical speaker-based research and text-type research with computational analysis. Third is to discover what authors of the various papers in the volume feel are challenges to the World Englishes paradigm and where future research should go. Papers like Braj Kachrus Meaning in Deviation: Toward Understanding Non-Native English Texts, for example, highlight the lack of research on a multilingual persons linguistic behavior. A fourth way to view this book is as a collection of hard-to-find classic articles such as Kamal K. Sridhars and S. N. Sridhars Bridging the Paradigm Gap: Second-Language Acquisition Theory and Indigenized Varieties of English, and Rodney Moags The Life Cycle of Non-Native Englishes: A Case Study. Yet another way to use this volume is as a textbook for a World Englishes class or to augment information traditionally available to students in TESOL preparatory programs. Conceptual arguments such as those outlined by Peter Lowenberg in his Testing English as a World Language: Issues in Assessing Non-Native Proficiency, and Cecil Nelson in his My Language, Your Culture: Whose Communicative Competence? are critically important for both scholars and language educators to confront. A sixth reason to use this volume is to access information about particular varieties of English in both the spoken and literary context in what Kachru terms the Outer and Expanding circles. The varieties dealt with include African English, Nigerian English, Chinese English, Japanese English, and Indian English. The volume provides a historical perspective on language standardization in Henry Kahanes American English: From a Colonial SubREVIEWS 205
standard to a Prestige Language, and a policy planning perspective in Shirley Brice Heaths American English: Quest for a Model. In spite of the tight editing, updating of citations in most of the articles, and solid range of topics covered, there are some weaknesses to the text. Several of the articles are much longer than the others and perhaps could have been edited more as, for example, Edwin Thumboos The Literary Dimension of the Spread of English. Although the text contains examples of two Expanding Circle varieties of English (Chinese and Japanese), none of the articles deals with emerging research questions which may affect these varieties differently from Outer Circle varieties. Nevertheless, just as the first edition gently nudged readers toward expanding their familiarity with an emerging paradigm in applied linguistics, this second edition continues to articulately push readers into an exploration of dimensions of the disciplines of linguistics and language education which are rarely covered elsewhere. KIMBERLEY BROWN
Portland State University
206
TESOL QUARTERLY
BOOK NOTICES
The TESOL Quarterly welcomes short evaluative reviews of print and nonprint publications relevant to TESOL professionals. Book notices may not exceed 500 words and must contain some discussion of the significance of the work in the context of current theory and practice in TESOL.
Classroom Observation Tasks. Ruth Wajnryb. Cambridge:

Cambridge University Press, 1992. Pp. vii + 145.
s Informed
teaching is a prerequisite for teaching that engages students while meeting their needs. Informed teaching is the product of planning before teaching and reflecting after teaching on the process of what occurs in the classroom. An excellent aid to informed teaching is observation of teaching as it occurs. Ruth Wajnryb has written a very useful resource which carefully structures a series of specific tasks for such observation. These tasks encourage focused reflection on what occurs in the classroom, leading to informed and thoughtful teaching, and extrication from ritualized practice on the part of the observer and observed. The book is divided into two parts. Part 1 discusses the usefulness of observation, the intended audience for the book, and how the tasks in Part 2 can be used. The book is primarily designed for practicing teachers to use with other peers. However, the observation tasks can also be used by teachers in training, teacher educators, and school support personnel. Part 2 provides a series of specific tasks grouped under the following seven chapters: The Learner, Language, Learning, The Lesson, Teaching Skills and Strategies, Classroom Management, and Materials and Resources. Each chapter has five tasks. For example, The Lesson contains tasks on lesson planning, openings and closures, lesson phases and transitions, grammar as content, and lesson breakdowns. All of the tasks cover essential basics. However, every task in the book deals with important issues that an informed teacher needs to constantly consider. An index provides quick access to all of the tasks. Each task is divided into four parts: background, objective, procedure, and reflection. Wajnryb often connects theory to practice in the background section. Objectives are clearly stated and always relevant. The section on procedure is divided into three parts: Before the Lesson, During the Lesson, and After the Lesson. Clear instructions are given for each part. The observer collects data during the observation, and Wajnryb provides necessary charts and worksheets to use. She also provides extensive, thoughtful questions for the observer and observed to consider together after the lesson. These allow for teachers to bring in their experiences in seeking answers. The short reflection section always brings the focus back to the classroom of the observer. This format is effective in
207
encouraging reflection on the part of the observer on his/her own teaching in regard to the issues of the task. Observation of and by peers, followed by supportive reflection and analysis, can enhance professional development and lead to a more collegial work environment. Both teachers and students can benefit from this process. Classroom Observation Tasks provides a very concrete focus to facilitate observation and reflection on what is done in the classroom. A wise teacher would benefit from reading the book even without using it in actual observations; it stands as a reminder of the basic issues that are sometimes forgotten when a teacher falls into ritualized patterns of teaching. An even wiser teacher would become involved in supportive, nonjudgmental peer observation. This book would be an essential aid for that process. JAMES RIEDEL
Kwansei Gakuin University
Literature and Language Teaching: A Guide for Teachers and Teacher Trainers. Gillian Lazar. Cambridge: Cambridge
University Press, 1993. Pp. xii + 254. and Language Teaching helps language educators find meaningful, practical ways of using literature in the classroom, enabling learners to achieve their main purpose for being in the classroom, that is to improve their English (p. xxiii). The book presents some issues on this ever-evolving subject, then asks its readers to think about them by participating in activities and tasks outlined in each chapter. This makes Literature and Language Teaching an ideal text, not only for the language teacher but also for those involved in the training and development of teachers. The text is primarily for intermediate- to advanced-level learners, although the use of literature with beginning students is briefly touched upon. The text has nine chapters that move progressively from theory to practice. This structure begins with Chapters 1 and 2, which focus specifically on the theory and approaches for using literature in the language classroom. Chapters 3 and 4 are concerned with the selection and evaluation of materials, emphasizing the cultural aspects of literary texts. The most practical chapters of the text, Chapters 5,6, and 7, concentrate on designing materials and planning lessons which involve a variety of literary tools and genres: novels, short stories, poetry, and plays. The author discusses the distinctive features of each genre, preparing the language teacher for any difficulties students may have. For example, Chapter 6 contains 27 tasks designed to help teachers explore the theory, approach, and practice of using poetry in the classroom. Each task involves learning by doing. Thus, Task 1 asks the learner to write any associations brought about by the word sea before being introduced to the e.e. cummings poem, maggie and milly and molly and may, a poem centered around the sea. In Task 2, the learner is asked to reorder jumbled lines
208 TESOL QUARTERLY
s Literature
of the original poem. The four subsequent tasks involve analyzing the reasons for ordering the jumbled poem as the learner did, examining the linguistically distinctive aspects of cummingss original poem, and then establishing ways to adapt the activities for differing classroom needs. Chapter 8 exemplifies the completeness of the text by focusing on how to evaluate the success of a lesson. Finally, Chapter 9 explains how to set up a literature self-access center which is described by the author as a library or small collection of texts for students to read on their own with minimal supervision (p. 179), an added bonus for any educator. I found that Literature and Language Teaching provided me with practical classroom applications and new insights. It is an important text for language teachers and teacher educators seeking creative and engaging ways for using literature in the classroom. GINA KEEFER
Kwansei Gakuin University
Longman Language Activator: The Worlds First Production Dictionay. Essex, England: Longrnan. 1993. Pp. xxxiv + 1587.
ESL learner dictionaries can be useful tools for helping learners understand what they read and, to a lesser degree, what they hear, but they are less able to help learners diversify and control the words they produce. This inadequacy motivated the creation of the Longman Language Activator, a self-described production dictionary that attempts to provide precise meanings so learners can know which words to use in specific contexts and which words occur together. The Activators layout is different from that of traditional dictionaries. Its alphabetized items typically are followed by a key word learners should turn to instead of a definition. The 1,052 key words were determined to be familiar to most intermediate- and advanced-level learners on the basis of Longmans examination of learners writing from 70 different countries. Each key word has an introduction box that contains a meaning menu. In the case of throw, nine different types of throwing are listed. If the idea learners want is the second, they go down the columns to the section numbered two and find four lexical entries ordered according to their frequency in the approximately 40-million-word English computer corpora Longman has examined. Each entry includes (a) a phonemic transcription, (b) a brief definition written in the simple language of Longmans 2,000-word defining vocabulary, (c) labels for the parts of speech and grammatical characteristics of the word, for example, v for verb and T for transitive, (d) in bold print, the ways the word typically fits into phrases, and (e) two to three sentences based on or taken directly from the corpora that illustrate how the word is used. This contextualization of items is among the Activators greatest strengths. It is example rich, and the examples demonstrate meanings clearly. The Activator is also strong in its indication of the registers and
BOOK NOTICES 209
s Traditional
appropriacy of words and phrases. The example sentences show situations for which the entry is best suited, and the definition section comments on when the term is used, frequently labeling an entrys formality but sometimes being much more specific. Words used primarily in speech are marked with a line drawing of an ear and the word in their definitions. Finally, the Activators entries are exceptionally thorough. For instance, it devotes more than five pages to find and find out, including 9 different types of finding and 10 different types of finding out and 76 individual entries11 featuring the key words and 65 presenting synonyms. Yet, despite its strengths, the Activator is probably not all it claims to be, namely a production dictionary. To produce language, learners need to know not only when words are appropriate but also their syntactic properties, properties clearly stated in many traditional learner dictionaries but often only implied in the Activator. For instance, the Activator does not state the key word throw is an irregular verb. True, 7 of the 13 sample sentences are in the past tense and three more use the past participle because they are written in the passive voice, but an explicit statement of the irregular forms would take little space and would highlight their use in the examples. Another of the Activators shortcomings is it does not always provide entries which intuitively seem to be among those most favored by proficient English speakers. For instance, under the key word manager, there is no mention of the acronym CEO. Perhaps these omissions are the result of the corpora relying more on British publications and speakers than U.S. The Activators British leanings are also noticeable in its representation of pronunciations. For example, the vowel in throw is transcribed as rather than the /o/ or /ow/ typical in learner dictionaries which model U.S. pronunciation. A further shortcoming is the limited number of nouns among the Activators entries. For example, of the 117 key words that begin with a and b, only 27 are nouns or pronouns. This conscious omission was based on the notion that concrete nouns and content words in general are less difficult for learners to use than abstract nouns and function words. But the Activator is not able to give learners all they need to know, for instance, to keep them from saying they live in the third floor. Instead, the Activator provides idiomatic uses like be in on something and synonyms like within and interior. Because it excels in presenting different registers of use and fitting synonyms for particular situations, it would appear nouns are an area of the language where the Activator could be very effective, especially when learners remember having heard or read a special term, but they cannot recall exactly what it was when they want to use it. The descriptor, production dictionary, is probably too ambitious for the Activator, yet if it is instead thought of as a learners thesaurus, it is exceptional. With its thorough contextualizations and. its avoidance of infrequently used terms, the Activator prevents two problems associated with learners use of native-speaker thesauruses: the use of synonyms which are inappropriate or inaccurate in a given context and the use of words which are so obscure they are unfamiliar to many native speakers.
210 TESOL QUARTERLY
Also, calling the Activator a thesaurus would rightfully indicate that it is not a replacement for the traditional learner dictionaries which learners will still need for the greater number of words and the more extensive syntactic information they contain. CHERYL EASON and ROBERT YATES
Central Missouri State University
Process Your ThoughtsWriting with Computers. Marianne

Phinney. Boston, MA: Heinle & Heinle, 1994. Pp. xiv + 236 and software for MS-DOS (IBM Compatible) or Macintosh. Your Thoughts (hereafter PYT) is an integrated textbook/software package for high intermediate- to advanced-level ESL students in computer assisted writing classes. Even though PYT does not focus on academic writing, it will lead the students to acquire basic writing and word processing skills that will prepare them for an academic course. The complete package includes a student textbook, accompanying software, and an instructors manual. The student text is organized into two sections. The first section contains three chapters that introduce the students to the use of computers for writing purposes. Each chapter contains worksheets for the students to become acquainted, through hands-on experience, with the type of computer and word processing software they will be using during the course. The second section is composed of six chapters dealing with six topics. The process approach is the organizing principle for this section. The chapters deal with prewriting, drafting, reviewing, revising, and editing. Also, they include questions for journal writing and portfolio self-evaluations, as well as computer and writing worksheets. The latter can be completed either in the textbook or on the computer. One very valuable characteristic of PYT is its spiral design. Each chapter develops the writing stages one step further by presenting a different prewriting technique, an additional component of the drafting, revising, or editing stage, and a new word processing feature. This systematic process promotes the students development of writing and computer skills in a slow but steady fashion. Furthermore, it allows them to identify the writing strategies that best suit their personal needs and preferences. The Macintosh version of the demonstration software for PYT is a user-friendly program that can be easily and independently operated either through the icons within the interactive program or through the ruler bar menu at the top of the screen. In addition, the interactive program offers two different help menus. One menu provides the users with a description of the icons displayed on the various screens and the other with instructions on how to complete the worksheets. There is not, however, any help menu that the users can consult when choosing the chapter they want to work with. Another positive feature is its good readability. The screens are easy to read thanks to careful screen design and appealing graphics. Each screen
BOOK NOTICES 211
s Process
has a moderate quantity of information and graphics related to the theme of the chapter that enhances comprehensibility. One aspect that could be improved in this software is the users control of the review sequence. In fact, the program does not allow for the review of questions within the worksheets. Once the Next Question button has been clicked and a new question appears on the screen, it is not possible, at least in the demonstration software, to return to the previous question. Adding this feature to the program would certainly enhance its interactive capabilities, thus making it simpler to use and more attractive to the user. PYT is a welcome addition to the shelf of ESL materials. It is a wellconceived and stimulating textbook/software package that should prove to be a valuable asset for the development of basic writing and word processing skills. GLADYS VEGA SCOTT
Purdue University
212
TESOL QUARTERLY
INFORMATION FOR CONTRIBUTORS

EDITORIAL POLICY
The TESOL Quarterly, a professional, refereed journal, encourages submission of previously unpublished articles on topics of significance to individuals concerned with the teaching of English as a second or foreign language and of standard English as a second dialect. As a publication that represents a variety of cross-disciplinary interests, both theoretical and practical, the Quarterly invites manuscripts on a wide range of topics, especially in the following areas: 1. psychology and sociology of language learning and teaching; issues in research and research methodology 2. curriculum design and development; instructional methods, materials, and techniques 3. testing and evaluation 4. professional preparation 5. language planning 6. professional standards
Because the Quarterly is committed to publishing manuscripts that contribute to bridging theory and practice in our profession, it particularly welcomes submissions drawing on relevant research (e.g., in anthropology, applied and theoretical linguistics, communication, education, English education [including reading and writing theory], psycholinguistics, psychology, first and second language acquisition, sociolinguistics, and sociology) and that address implications and applications of this research to issues in our profession. The Quarterly prefers that all submissions be written so that their content is accessible to a broad readership, including those individuals who may not have familiarity with the subject matter addressed.
GENERAL INFORMATION FOR AUTHORS

Submission Categories
The TESOL Quarterly invites submissions in five categories: Full-length articles. Manuscripts should generally be no longer than 20 double-spaced pages. Submit three copies plus three copies of an informative abstract of not more than 200 words. To facilitate the blind review process, authors names should appear only on a cover sheet, not on the title page; do not use running heads. Manuscripts should be submitted to the Editor of the TESOL Quarterly: Sandra McKay Department of English San Francisco State University 1600 Holloway Avenue San Francisco, CA 94132
213
Reviews. The TESOL Quarterly invites succinct, evaluative reviews of professional books, classroom texts, and other instructional resources (such as computer software, video- or audiotaped material, and tests). Reviews should provide a descriptive and evaluative summary and a brief discussion of the significance of the work in the context of current theory and practice. Submissions should generally be no longer than 500 words. Submit two copies of the Review to the Review Editor: H. Douglas Brown American Language Institute San Francisco State University 1600 Holloway Avenue San Francisco, CA 94132 U.S.A. Review Articles. The TESOL Quarterly also welcomes occasional review articles, that is, comparative discussions of several publications that fall into a topical category (e.g., pronunciation, literacy training, teaching methodology). Review articles should provide a description and evaluative comparison of the materials and discuss the relative significance of the works in the context of current theory and practice. Submissions should generally be no longer than 1,500 words. Submit two copies of the review article to the Review Editor at the address given above. Brief Reports and Summaries. The TESOL Quarterly also invites short reports on any aspect of theory and practice in our profession. We encourage manuscripts which either present preliminary findings or focus on some aspect of a larger study. In all cases, the discussion of issues should be supported by empirical evidence, collected through qualitative or quantitative investigations. Reports or summaries should present key concepts and results in a manner that will make the research accessible to our diverse readership. Submissions to this section should be three to seven double-spaced pages (including references and notes). Longer articles do not appear in this section and should be submitted to the Editor of the TESOL Quarterly for review. Send two copies of the manuscript to the Editors of the Brief Reports and Summaries section: Graham Crookes and Kathryn A. Davis Department of English as a Second Language University of Hawaii at Manoa 1890 East-West Road Honolulu, HI 96822 U.S.A. The Forum. The TESOL Quarterly welcomes comments and reactions from readers regarding specific aspects or practices of our profession. Responses to published articles and reviews are also welcome; unfortunately, we are not able to publish responses to previous exchanges. Contributions to The Forum should generally be no longer than five double-spaced pages. Submit two copies to the Editor of the TESOL Quarterly at the address given above.
214 TESOL QUARTERLY
Brief discussions of qualitative and quantitative Research Issues and of Teaching Issues are also published in The Forum. Although these contributions are typically solicited, readers may send topic suggestions and/or make known their availability as contributors by writing directly to the Editors of these subsections. Research Issues: Donna M. Johnson English Department ML 455 University of Arizona Tucson, AZ 85721 Teaching Issues: Bonny Norton Peirce Modern Language Centre Ontario Institute for Studies in Education 252 Bloor St. W. Toronto, Ontario M5S 1V6 Canada
Special-Topic Issues. Typically, one issue per volume will be devoted to a special topic. Topics are approved by the Editorial Advisory Board of the Quarterly. Those wishing to suggest topics and/or make known their availability as guest editors should contact the Editor of the TESOL Quarterly. Issues will generally contain both invited articles designed to survey and illuminate central themes as well as articles solicited through a call for papers.
General Submission Guidelines

1. All submissions to the Quarterly should conform to the requirements of the Publication Manual of the American Psychological Association (4th ed.), which can be obtained from the Order Department, American Psychological Association, P.O. Box 2710, Hyattsville, MD 20784-0710. The Publication Manual is also available in many libraries and bookstores. Authors are responsible for the accuracy of references and reference citations, which must be in APA format. 2. All submissions to the TESOL Quarterly should be accompanied by a cover letter which includes a full mailing address and both a daytime and an evening telephone number. Where available, include an electronic mail address and fax number. 3. Authors of full-length articles should include two copies of a very brief biographical statement (in sentence form, maximum 50 words), plus any special notations or acknowledgments that they would like to have included. Double spacing should be used throughout. 4. The TESOL Quarterly provides 25 free reprints of published full-length articles and 10 reprints of material published in the Reviews, Brief Reports and Summaries, and The Forum sections. 5. Manuscripts submitted to the TESOL Quarterly cannot be returned to authors. Authors should be sure to keep a copy for themselves. 6. It is understood that manuscripts submitted to the TESOL Quarterly
INFORMATION FOR CONTRIBUTORS 215
have not been previously published and are not under consideration for publication elsewhere. 7. It is the responsibility of the author(s) of a manuscript submitted to the TESOL Quarterly to indicate to the Editor the existence of any work already published (or under consideration for publication elsewhere) by the author(s) that is similar in content to that of the manuscript. 8. The Editor of the TESOL Quarterly reserves the right to make editorial changes in any manuscript accepted for publication to enhance clarity or style. The author will be consulted only if the editing has been substantial. 9. The views expressed by contributors to the TESOL Quarterly do not necessarily reflect those of the Editor, The Editorial Advisory Board, or TESOL. Material published in the Quarterly should not be construed to have the endorsement of TESOL.
Statistical Guidelines
Because of the educational role the Quarterly plays modeling research in the field, it is of particular concern that published research articles meet high statistical standards. In order to support this goal, the following guidelines are provided. Reporting the study. Studies submitted to the Quarterly should be explained clearly and in enough detail that it would be possible to replicate the design of the study on the basis of the information provided in the article. Likewise, the study should include sufficient information to allow readers to evaluate the claims made by the author. In order to accommodate both of these requirements, authors of statistical studies should present the following. 1. A clear statement of the research questions and the hypotheses which are being examined 2. Descriptive statistics, including the means, standard deviations, and sample sizes, necessary for the reader to correctly interpret and evaluate any inferential statistics 3. Appropriate types of reliability and validity of any tests, ratings, questionnaires, etc. 4. Graphs and charts which help explain the results 5. Clear and careful descriptions of the instruments used and the types of intervention employed in the study 6. Explicit identifications of dependent, independent, moderator, intervening, and control variables 7. Complete source tables for statistical tests 8. Discussions of how the assumptions underlying the research design were met, assumptions such as random selection and assignment of
216 TESOL QUARTERLY
subjects, sufficiently large sample sizes so that the results are stable, etc. 9. Tests of the assumptions of any statistical tests, when appropriate 10. Realistic interpretations of the statistical significance of the results, keeping in mind that the meaningfulness of the results is a separate and important issue, especially for correlation Conducting the analyses. Quantitative studies submitted to the TESOL Quarterly should reflect a concern for controlling Type I and Type II error. Thus, studies should avoid multiple t tests, multiple ANOVAs, etc. However, in the very few instances in which multiple tests might be employed, the author should explain the effects of such use on the probability values in the results. In reporting the statistical analyses, authors should choose one significance level (usually .05) and report all results in terms of that level. Likewise, studies should report effect size through such strength of association measures as omega-squared or etasquared along with beta (the possibility of Type II error) whenever this may be important to interpreting the significance of the results. Interpreting the results. The results should be explained clearly and the implications discussed such that readers without extensive training in the use of statistics can understand them. Care should be taken in making causal inferences from statistical results, and these should be avoided with correlational studies. Results of the study should not be overinterpreted or overgeneralized. Finally, alternative explanations of the results should be discussed.
Qualitative Research Guidelines

To ensure that Quarterly articles model rigorous qualitative research, the following guidelines are provided. Conducting the study. Studies submitted to the Quarterly should exhibit an in-depth understanding of the philosophical perspectives and research methodologies inherent in conducting qualitative research. Utilizing these perspectives and methods in the course of conducting research helps to ensure that studies are credible, valid, and dependable rather than impressionistic and superficial. Reports of qualitative research should meet the following criteria. 1. Data collection (as well as analyses and reporting) is aimed at uncovering an emit perspective. In other words, the study focuses on research participants perspectives and interpretations of behavior, events, and situations rather than etic (outsider-imposed) categories,
models, and viewpoints. 2. Data collection strategies include prolonged engagement, persistent observation, and triangulation. Researchers should conduct ongoing observations over a sufficient period of time so as to build trust with INFORMATION FOR CONTRIBUTORS 217
respondents, learn the culture (e.g., classroom, school, or community), and check for misinformation introduced by both the researcher and the researched. Triangulation involves the use of multiple methods and sources such as participant-observation, informal and formal interviewing, and collection of relevant or available documents. Analyzing the data. Data analysis is also guided by the philosophy and methods underlying qualitative research studies. The researcher should engage in comprehensive data treatment in which data from all relevant sources are analyzed. In addition, many qualitative studies demand an analytic inductive approach involving a cyclical process of data collection, analysis (taking an emit perspective and utilizing the descriptive language the respondents themselves use), creation of hypotheses, and testing of hypotheses in further data collection. Reporting the data. The researcher should generally provide thick description with sufficient detail to allow the reader to determine whether transfer to other situations can be considered. Reports also should include the following. 1. A description of the theoretical or conceptual framework that guides research questions and interpretations. 2. A clear statement of the research questions. 3. A description of the research site, participants, procedures for ensuring participant anonymity, and data collection strategies. A description of the roles of the researcher(s). 4. A description of a clear and salient organization of patterns found through data analysis. Reports of patterns should include representative examples not anecdotal information. 5. Interpretations that exhibit a holistic perspective in which the author traces the meaning of patterns across all the theoretically salient or descriptively relevant micro- and macrocontexts in which they are embedded. 6. Interpretations and conclusions that provide evidence of grounded theory and discussion of how this theory relates to current research/ theory in the field, including relevant citations. In other words, the article should focus on the issues or behavior that are salient to participants and that not only reveal an in-depth understanding of the situation studied but also suggest how it connects to current related theories.
218
TESOL QUARTERLY

TESOL Quarterly (Spring 1995)

Uploaded by

Copyright:

Available Formats

TESOL Quarterly (Spring 1995)

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

TESOL Quarterly (Spring 1995)

Uploaded by

Copyright:

Available Formats

TABLE OF CONTENTS

SANDRA McKAY, San Francisco State University

BONNY NORTON PEIRCE, Ontario Institute for Studies in Education

Teaching Issues Editor

Brief Reports and Summaries Editors

at Manoa Review Editor

Editorial Advisory Board

California State University, Fresno

Northern Arizona University

California State Polytechnic University, Pomona

Bonny Norton Peirce, Patricia Porter, Kamal Sridhar,

Nancy Hornberger, Sarah Hudelson, Thom Hudson,

San Francisco State University State University of New York

Volume 29, Number 1 u Spring 1995

BRIEF REPORTS AND SUMMARIES

OFFICERS AND EXECUTIVE BOARD 1994-95

q Dana Ferris reports on ESL students reactions to teacher response

q In his article, David Nunan explores ways in which the relationship

Alderson and Caroline Clapham on language tests and ESL teaching.

NEW WAYS SERIES II

Guidelines for contributors

1600 Cameron St., Suite 300l

TESOL QUARTERLY Vol. 29, No. 1, Spring 1995

Social Identity, Investment, and Language Learning*

THE STUDY: IMMIGRANT WOMEN AS LANGUAGE LEARNERS

SOCIAL IDENTITY, INVESTMENT, AND LEARNING

From Motivation to Investment

SOCIAL IDENTITY, INVESTMENT, AND LEARNING

Communicative Competence and the Right to Speak

THE ANALYSIS: IDENTITY, INVESTMENT, AND LANGUAGE LEARNING

Investment and Social Identity

SOCIAL IDENTITY, INVESTMENT, AND LEARNING

Martina: Social identity as Multiple and a Site of Struggle

Eva: Social Identity as Changing Over Time

THE IMPLICATIONS: CLASSROOM-BASED SOCIAL RESEARCH

SOCIAL IDENTITY, INVESTMENT, AND LEARNING

CBSR might include the following objectives and methodologies.

Objective 1: Investigative Opportunities to Interact with Target Language Speakers

Objective 2: Reflect Critically on Engagement with Target Language Speakers

Objective 3: Reflect on Observations in Diaries or Journals

Objective: Pay Attention to and Record Unusual Events

Objective 5: Compare Data with Fellow Students and Researchers

SOCIAL IDENTITY, INVESTMENT, AND LEARNING

Third Summer Institute in English and Applied Linguistics

Context in Language Learning and Language Understanding

Contributors are expected to include:

TESOL QUARTERLY Vol. 29, No. 1, Spring 1995

Student Reactions to Teacher Response in Multiple-Draft Composition Classrooms*

DOES TEACHER FEEDBACK HELP?

WHAT DO STUDENTS THINK ABOUT TEACHER FEEDBACK?

STUDENT REACTIONS TO TEACHER RESPONSE

How Did Students Perceive Themselves as Learners and Writers?

What Did Students Report Doing in Response to Teacher Feedback?

Did Students Report Difficulties in Understanding Teachers Feedback?

STUDENT REACTIONS TO TEACHER RESPONSE

Did Students Say They Received Positive Feedback?

Did Students Feel That Their Teachers Feedback Was Helpful?

CONCLUSIONS Summary of Research Findings

Implications for Pedagogy

The Role of Grammar Feedback

* p < .05. p < . 01. * p < .001. COMMUNICATION STRATEGIES 77