VOLUMES MENU

CONTENTS
ARTICLES
535 Tradition and Revolution in ESL Teaching Ann Raimes 553 Schema Theory and ESL Reading Pedagogy Patricia L. Carrell and Joan C. Eisterhold 575 Student-Teacher Working Journals in ESL Freshman Composition Ruth Spack and Catherine Sadow 595 Increasing Learner Involvement in Course Management Andrew Peter Littlejohn The Least You Should Know About Arabic: Implications for the ESL Writing Instructor 609 Karyn Thompson-Panos and Maria Thomas-Ruzic Recent Language Research and Some Language Teaching Principles 625 Karl J. Krahnke and Mary Ann Christison On the Use of Composition Scoring Techniques, Objective Measures, and Objective Tests to Evaluate ESL Writing Ability 651 Kyle Perkins

To print, select PDF page nos. in parentheses

(8-25) (26-46) (48-66) (68-81) (82-96) (97-121)

(122-142)

REVIEWS
Language and Culture in Conflict: Problem-Posing in the ESL Classroom 673

Nina Wallerstein Reviewed by Linda M. Crawford-Lange

Essential Idioms in EnglishA New Revised Edition

Robert J. Dixon Reviewed by Helen K. Fragiadakis BRIEF REPORTS AND SUMMARIES 679 THE FORUM 685 INFORMATION FOR CONTRIBUTORS 693 Editorial Policy General Information for Authors Guidelines for the Preparation of Manuscripts Publications Received 701 Publications Available from the TESOL Central Office TESOL Membership Application 728

676

703

OFFICERS AND EXECUTIVE BOARD 1983-84

President
Penelope M. Alatis Francis C. Hammond JOHN F. HASKELL Northeastern Illinois University Junior High School Alexandria, Virginia Chicago, Illinois First Vice President Andrew D. Cohen Hebrew University CHARLES H. BLATCHFORD Jerusalem, Israel Fair Oaks, California Second Vice President Holly L. Jacobs Atlanta, Georgia Darlene Larson New York University New York, New York Lin Lougheed United States Information Agency Washington, D.C. Marsha Robbins Santelli Chicago Public Schools Chicago, Illinois

PENNY LARSON Alemany Community College Center San Francisco, California

Executive

JoAnn Crandall Center for Applied Linguistics Washington, D.C. John F. Fanselow Teachers College Columbia University New York, New York

JAMES E. ALATIS Georgetown University Washington, D.C. 530

Director

Executive Assistant CAROL M. LeCLAIR Jean Handscombe North York Board of Education Georgetown University Washington, D.C. Toronto, Ontario

TESOL QUARTERLY

TESOL QUARTERLY
A Journal for Teachers of English to Speakers of Other Languages and of Standard English as a Second Dialect

Editor BARRY P. TAYLOR, University of Pennsylvania Review Editor CHARLENE J. SATO, University of Hawaii at Manoa Brief Reports and Summaries Editor ANN FATHMAN, Caada College Assistants to the Editor SUSAN ARLIN, CATHERINE DOUGHTY, LINDA KEEHN, University of Pennsylvania Editorial Advisory Board
Paul J. Angelis Southern Illinois University Christopher Brumfit University of London Michael Canale Ontario Institute for Studies in Education Marianne Celce-Murcia University of California, Los Angeles Richard Day University of Hawaii at Manoa David E. Eskey University of Southern California Stephen J. Gaies University of Northern Iowa Grace S. Holt University of Illinois at Chicago Ann M. Johns San Diego State University Michael H. Long University of Hawaii at Manoa Joan Morley University of Michigan Ann Raimes Hunter College, City University of New York Arnulfo G. Ramirez State University of New York at Albany Jack C. Richards University of Hawaii at Manoa Pat Rigg Center for the Expansion of Language and Thinking Muriel Saville-Troike University of Illinois Thomas Scovel San Francisco State University G. Richard Tucker Center for Applied Linguistics Rita Wong San Francisco State University Carlos A. Yorio University of Toronto

Additional Readers
Lyle Bachman, Patricia Carrell, JoAnn Crandall, Allene Grognet, Holly Jacobs, Carol Kreidler, John Oller, Fay Pallen, Teresa Pica, Charlene Sato, Jacquelyn Schachter, Linda Schinke-Llano, Richard Schreck, Charles Stansfield, Thurston Womack, Vivian Zamel

Credits
Advertising arranged by Aaron Berman, TESOL Development and Promotions, San Francisco, California Typesetting, printing, and binding by Pantagraph Printing, Bloomington, Illinois Design by Chuck Thayer Advertising, San Francisco, California

TESOL QUARTERLY

531

TESOL QUARTERLY

Editor's Note
Charlene Sato, recently appointed Review Editor of the TESOL Quarterly, has informed me that a very heavy workload makes it necessary for her to give up her Quarterly position. Although Charlie had warned me at the time of her appointment that this might turn out to be the case, it is nevertheless with disappointment that I have accepted her resignation. Despite her short tenure, Charlie has been a strong contributor to the Quarterly, and we thank her for editing the December 1983 and March 1984 Review sections. Starting with the June 1984 issue, the Review section will be edited by Vivian Zamel, a faculty member in the English Department at the University of Massachusetts/Boston. Over the years Vivian has contributed to the Quarterly both as the author of a number of articles and as an evaluator of manuscripts submitted for consideration for publication. We are all fortunate to have Vivian on the Quarterly staff, and we welcome her. Effective January 1, 1984, reviewers of publishers materials are invited to sent their submissions directly to Vivian Zamel at the address listed in the Information for Contributors section of this issue.

In This Issue
This issue of the TESOL Quarterly has a decidedly pedagogical focus. With few exceptions, the articles published here address issues that should be particularly relevant to those ESL professionals whose work brings them in contact with students in classroom settings. Ann Raimes looks at the field of ESL in terms of Thomas Kuhns theory of the structure of scientific revolutions. But unlike others who have used Kuhns model to posit one or more revolutions in ESL teaching during this century, Raimes argues that a new set of shared assumptions about our discipline (a paradigm) is not emerging since our practices are still firmly entrenched in a positivist tradition. 532 TESOL QUARTERLY

Although Raimes points out the signs to indicate that our orientation is in flux, she maintains that the field of language teaching is not experiencing a revolution, but rather a paradigm shift. Patricia Carrell and Joan Eisterhold examine the role of background knowledge (what the reader already knows) within the context of schema theory, which maintains that reading comprehension is characterized by an interaction between a text and the readers prior background knowledge. They argue for the relevance of schematheoretic views in the teaching of reading to non-native speakers of English and provide guidelines to help classroom teachers manipulate the variables of the text and/or the reader in order to devise classroom activities which foster reader-centered reading. Ruth Spack and Catherine Sadow, taking a process-oriented approach to the teaching of writing, discuss the use of student-teacher working journals in ESL composition classes. Unlike many writing instructors who have their students keep diary-like journals, however, Spack and Sadow propose an interactive process in which teachers also write journals and in which topics are motivated not by personal, but rather by academic and class-related, concerns. Andrew Littlejohn addresses the currently popular question of how to increase learners involvement in the management of their language courses. After outlining some of the arguments which have been offered to support a learner-centered approach to teaching, he discusses some activities and techniques which were found to be successful with a group of ESL students who had been categorized as unmotivated and demoralized. Littlejohn argues that the implementation of a learner-centered approach needs to be gradual, especially when classes are composed of students who have specific expectations based on their prior language learning experiences.

salient features of the Arabic language which they feel can help ESL instructors (particularly those who teach writing) understand the problems which their Arabic-speaking students experience. They argue that although contrastive analysis is no longer considered an appropriate foundation for contemporary language instruction, an overview of some of the contrastive linguistic and rhetorical features of Arabic and English can help ESL professionals deal with some of the weaknesses in writing which, based on the authors experiences, Arab students typically exhibit. Karl Krahnke and Mary Ann Christison synthesize a large body of current research in linguistics and second language acquisition, focusing primarily on findings from research on learning/acquisition, conversational interaction, pragmatic, repair, error, and affective variables. From this research and their discussion of it, they extract four basic principles which they recommend to classroom instructors. IN THIS ISSUE 533

 Kyle Perkins summarizes and evaluates the major direct methods of assessing writing ability (holistic scoring, analytical scoring, and primary trait scoring) and the most popular indirect methods of predicting students ability to write (T-unit analysis and a variety of published standardized tests). Examining each in turn, Perkins highlights their principle uses as well as their strengths and limitations. He concludes that all have their drawbacks. As he writes, no test or composition scoring procedure is perfect. Also in this issue: Reviews Brief Reports and Summaries The Forum, in which Mohsen Ghadessy reacts to Patricia Carrells article, Cohesion Is Not Coherence; Carrells response follows. Barry P. Taylor

534

TESOL QUARTERLY

TESOL QUARTERLY, VOI. 17, No.4, December 1983

Tradition and Revolution in ESL Teaching

ANN RAIMES
Hunter College, City University of New York

Revolutions in science, Thomas Kuhn argues, occur when the shared set of intellectual assumptions that form the tradition of a discipline (a paradigm) breaks down and gives way to another, causing controversy and insecurity in the process. This article begins by exploring the development of language teaching in the light of Kuhns theory. It summarizes the theory, points to analogies in the field of the teaching of writing, briefly defines our positivist tradition, and describes the signs of confusion and controversy that mark a paradigm shift. Then it proposes that the current emphasis on communication, far from marking the emergence of a new paradigm, still operates very much within the positivist tradition; the fact that scholars outside and within our discipline are asking more questions about the conceptual and creative function of language indicates that we are still firmly in the middle of a paradigm shift, engaged in philosophical debate. The article concludes by identifying a cluster of mentors who are raising challenging issues and by discussing possible directions for the future in research and method.

You are sitting in a room with an experimenter who shows you

playing cards quickly, one after the other. You identify each card. Among them is a six of spades, but it is a red six of spades. Without hesitation, you identify it as a six of spadesor perhaps as a six of hearts. Then the experimenter shows you the same cards again, but you see each one for a little longer time. Now when you see the red six of spades, you hesitate and become confused. With more exposure time, you finally identify the red six of spades as a red six of spades. Or, you might just give up and retreat, saying something like Im not even sure now what a spade looks like. Such an experiment (Bruner and Postman 1949) paints a picture in miniature of a theory of scientific discovery, proposed by Thomas Kuhn in The Structure of Scientific Revolutions (1970). This influential book has generated discussion about the development of intellectual disciplines, humanistic as well as scientific. In a period of normal 535

science, Kuhn says, the scientists, just like you in that card experiment, work according to a clear set of conceptual categories, an established paradigm. A paradigm can be seen to exist when most of the practitioners in the discipline hold a common body of beliefs and assumptions; they agree on the problems that need to be solved, the rules that govern research, and on the standards by which performance is to be measured (Hairston 1982:76). When an anomaly comes along it is not always identified as such, and scientists try to fit it into their established categories. But only initially. Then, as anomalies recur, the scientists, increasingly beset by indecision, confusion, or even distress, respond by resisting and ignoring the inconsistencies, by improvising ways of coping with the crisis, or by addressing the problems directly. This period between the recognition of the anomaly and the establishment of a new paradigm which accommodates the anomaly Kuhn calls the time of the paradigm shift. After an uneasy time of transition, a new model finally emerges; scientists either lapse into disgruntled incomprehension and cease to be active participants in the discipline, or they adjust their categories so that they can anticipate what was initially anomalous. That is, they subscribe to a new paradigm which breaks from the old. Kuhn thus does not see science as a steady acquisition of knowledge, with a gradual progression toward an ultimate truth. Instead, peaceful periods of normal science are punctuated by intellectual revolutions which introduce a new conceptual view. Such intellectual revolutions are not necessarily swift. Between 1690 and 1781, for example, when a new celestial body was observed, scientists first of all categorized it, in accordance with the prevailing conceptual scheme, as a star. More observations over a long period of time led to dissatisfaction with this category. Perhaps, after all, it was a comet? But it did not appear to fit the usual pattern of cometary orbit. Finally, the new celestial body was identified as a planetUranus. Kuhn sees this as one paradigm superseding another: A celestial body that had been observed off and on for almost a century was seen differently after 1781, because, like an anomalous playing card, it could no longer be fitted to the perceptual categories (star or comet) provided by the paradigm that had previously prevailed (1970:115-116). Although Kuhn based his theory on facts that emerge from the history of scientific discovery in fields such as physics, chemistry, and astronomy, he acknowledges borrowing his theses from other areas of human activity: in literature, music, the arts, and political development, historians have noted revolutionary breaks in style, taste, and institutional structure (1970;208). And certainly the stages Kuhn cites have sounded familiar to scholars outside the pure sciences and in

536

TESOL QUARTERLY

fields close to our own. In the field of the teaching of writing to native speakers, for instance, there is much talk of that profession being in the first stages of a paradigm shift, moving from an emphasis on the written product to an emphasis on the writing process and thus calling for new directions in research, particularly in rhetorical invention (Young 1978). Researchers in this field are busy describing the traditional paradigm, documenting the features of the emerging paradigm, identifying the scholars who are the authorities of the new tradition (Emig 1980), and predicting the future directions of the profession (Hairston 1982). And what of our own field, the teaching of English to speakers of other languages? Is it useful to examine our non-scientific discipline in the light of Kuhns theory? I believe it is, for such an analogy gives us a new vantage point from which to view the nature and development of our discipline and to consider our roles as teachers, researchers, administrators, writers, or publishers. The very process of thinking and reflecting about how our work compares to other disciplines is productive. It gives us a chance to be students of our own profession and to do what Bruner says is essential to student learningto climb on your own shoulders to be able to look down at what youve just done and then to represent it to yourself (1965:101). The analogy with Kuhns theory also gives us a framework in which to examine not just how we teach but all the other features that makeup the paradigm of a discipline, namely what is included in the discipline and what is excluded from it, what is taught and not taught, what problems are regarded as important and unimportant, and, by implication, what research is regarded as valuable in developing the discipline (Young 1978:29). In short, using the analogy forces us to look at what we know as well as what we do. We can begin to do that by scrutinizing our textbooks, our professional literature, and our classroom practices in an attempt to explore the same three questions that scholars in the field of the teaching of writing have been asking: 1. Can we define a traditional paradigm? 2. Are there signs of a paradigm shift? 3. Is a new paradigm emerging? However, the danger of asking the same questions about our discipline is that we might then be seduced into seeing the same patterns and finding the same affirmative answers. To avoid being trapped in a nice neat analogy, parallel with the teaching of writing, I want to add a fourth question: 4. If a new paradigm isnt emerging, where are we headed? TRADITION AND REVOLUTION 537

CAN WE DEFINE A TRADITIONAL PARADIGM? In Kuhns terms, the recognition of a traditional paradigm implies the existence of theoretical presuppositions on which the disciplines activity is based. It also implies that there will be a revolution against tradition. And it is this idea of revolution that has caught the attention of some theorists in our field, though they do not always see it as happening in the same way at the same time or even agree that it happens at all. A past President of TESOL, in referring to Kuhn and cyclical theories, sees no fewer than four revolutions this century, with a new paradigm emerging every twenty-five years (H. D. Brown 1975). We have, he posits, gone from the direct method to the grammar-translation method to the audiolingual method, and now, in the last quarter of the century we are still searching for the ultimate method, though we understand more clearly that there will be no such panacea in teaching a behavior as complex as a second language (H. D. Brown 1980:243). But others have counted differently, and we hear about two major revolutions in methodology during this century, one of which was the post-World War II break from the traditional, or grammar-translation, method, while the other was the proliferation of new affectively-based methods or approaches (Kunz 1980:170). The most recent count is down to no revolutions at all: even though teachers are confronted by a bewildering array of methodologies, they are urged not to conclude that our field has moved away from a generally accepted body of principles as a basis for the organization of language teaching, nor to see current practices as random or radical departures from the mainstream of applied, linguistic thought and practice since the present innovations are but variations on familiar themes (J. Richards and Rodgers 1982:153-154). The problem with these counts is that the scholars are looking only at classroom methodology and not at the underlying intellectual assumptions which generate methods. Most of us would agree, though, on the familiar themes that have characterized classroom instruction in the past: patterns/drills/rules/accuracy/sequence, all of which reflect assumptions in disciplines related to our own. The basic and still frequently practiced classroom technique of imitation and memorization of basic conversational sentences (Lado 1964:6) stems mostly from psychology and linguistics. From psychology comes Skinners behaviorist stimulus-response theory that emphasized positive reinforcement of correct responses to form habits. From linguistics, from Bloomfield, Sapir, and Fries, comes the view of language as an observable set of structures. And it is not only classroom instruction that borrows from psychology and linguistics. Our research, too, takes its emphasis on quantifiable data from these two fields; research studies in our field measure, count, and compare. The prevailing 538 TESOL QUARTERLY

assumptions of our discipline can thus be summed up as rooted firmly in the positivist tradition, which prizes the empirical, the identifiable, the countable, and the verifiable. When guided by such a positivistic governing gaze (Emig 1982:65), a profession like ours, in its desire to be scientific and objective, excludes everything except observable phenomena. So we have emphasized language as structure and form, we have divided language up into discrete units, we have stressed assembling and not creating. It is important to note, too, what we have excluded . . . not taught . . . and regarded as unimportant (Young 1978:290). We have paid little attention to real communication and to language as making meaning. Perhaps some of you are by now muttering, Yes, we know all that. Thats all old hat. Nobody subscribes to that stuff any more. Look at all the communicating, all the sharing of meaning that goes on in our classes. And you are rightup to a point. The authorities who used to dominate our literature and our classroomsFries and Lado and their followerssimply are not our mentors any more. There have indeed been changes. Contrary to what many in our profession would believe, however, I hold that those changes, while evident, have not yet been radical. This leads us to our next question. ARE THERE SIGNS OF A PARADIGM SHIFT? We have been living with dissonance in our profession for many years. Red sixes of spades keep appearing. Many of us will remember being interested some twenty years ago in Chomskys refutation (1959) of Skinners view of language. We were not, however, immediately rattled from the security of our fast-paced, finger-snapping oral-aural drills. Then, in 1966, Hymes raised the issue of real language use, of the importance of context and of the socio-cultural factors that influence any linguistic act. He redefined Chomskys notion of competence, coining a phrase that has since become a new ESL watchword: communicative competence. Suddenly rules of grammar were jostled by rules of use. In that same year a paper on How Not to Interfere with Language Learning was published, which maintained that language is learned a whole act at a time rather than learned as an assemblage of constituent skills (Newmark 1966/1979:163). Methodologies started competing with each other for the one way of dealing with all of these anomalies, methodologies as different from each other as the Silent Way, Community Language Learning, and Suggestopedia. With evangelistic zeal, teachers turned to these, found new mentors in the originators, and saw themselves as in the forefront of a revolution. At the same time, some scholars turned their attention to wider issues than methodology: they examined how we view and divide up language, TRADITION AND REVOLUTION 539

developing new syllabuses to replace the grammatical structure syllabus. The terms notional and functional worked their way insistently into our literature. Outside the classroom things began to move faster in the mid-1970s. Researchers turned their attention to second language acquisition, seeking new theories to explain the nature of language and language learning. The topic index of the TESOL Quarterly shows us how the field is changing. The index for 1967-76 (ten years) lists 29 articles on second language learning. For 1979-80 (two years) it lists 24 articles on what is re-named second language acquisitiona four-fold growth. Insights from the study of first language acquisition (R. Brown 1973, Bloom 1978) now inform our researchers literal re-search for new theories. The work of one of the researchers, Stephen Krashen, has done a great deal to attempt to explain the anomalies that are troubling us. His Input Hypothesis (1981) has emphasized language acquisition through real communication and has provided us with a theoretical basis for questioning instructional techniques that focus on patterns, drills, and rules. In addition, the willingness of Krashen to use selfreport (1981: 122-124, 133-136) and of other researchers to use diaries, introspective journals, and first-hand accounts in interviews to record the language learning process (F. Schumann and J. Schumann 1977, Bailey 1980, F. Schumann 1980, Stevick 1982) has shown us the value of the case study and of ethnographic research as an alternative to the empirical studies that dominate in the positivist tradition. Much of the dissonance and much of the activity in methods, curriculum, and research are revealed in the numbers of new questions that occupy us. We find ourselves now beset by questions that we never even used to ask because the answers were once so clear: Which grammar do we teach? Traditional, transformational, sector analysis, case grammar, none . . .? Which syllabus do we use? Structural, situational, notional, functional, student-initiated, none . . .? What part of the student do we address? The affective or the cognitive domain; the right or left hemisphere of the brain; the Parent, Child, or Adult (Stevick 1976); or the whole person . . .? We also question our own role as we ask: What are we? Guide, model, counselor, judge, carer and sharer, facilitator, or, dare we say it, teacher? All of these questions lead us further away from the common body of assumptions of our tradition. Now controversy and indecision abound in our literature and in our classrooms. And no one is immune, not even well-known and influential figures in the language teaching 540 TESOL QUARTERLY

field. Earl Stevick, for instance, describes deciding to leave Audiolinguia for the new land of Cognitia when something caused [him] to change [his] mind and leave Cognitia, not to return to Audiolinguia but to explore Terra Incognita (1976:104). That Terra Incognita is the place of the paradigm shift, now a heavily populated land, thronging with ESL teachers. Some step into the new unknown territory decisively. Others, more cautious, hold on to traditions as long as possible. Kuhn points out that when scientists are first confronted by anomaly they will devise ad hoc modifications of their theory in order to eliminate any apparent conflict (1970:78). Such ad hoc modifications are abundant in our field, providing evidence of a tentative questioning of tradition and yet reflecting a basic distrust and fear of abandoning it completely. The following are some of the solutions that we can see practitioners of our discipline using. One ad hoc solution is to include the new as a kind of overlay on the old. Textbooks that still divide and sequence the language into grammatical structures deal with the anomaly of What about real language use? by turning to functions. They add to the table of contents the functions covered in each chapter and include a few charts or exercises in the body of the text. A chapter devoted to the grammatical form of the imperative, of the traditional give me that book variety, now includes one short exercise where the students, with no change of context or of audience, are also instructed to use May I have that book? and Excuse me, would you mind handing me that book? (Lanzano and Bodman 1981:63). Some practitioners do not go that far. They just change labels. They incorporate the terminology but not the concepts of the new theories. The 1980 second edition of a book originally published in 1973 retains 20 of the original lessons. Each chapter head, however, now includes a function as well as a grammatical structure as the focus of the chapter: How many; need and want is joined by Expressing need and offering help. But the material is unchanged and students are still instructed to say things like Does the clerk yell? Yes, he does. He yells (Hines 1980:77-79). Another textbook, a composition book, trying to reflect the current focus on process in composition, uses a label of process prominently in its title yet states in its preface that one of its four assumptions is that American university writing is linear and straightforward in structure (Reid 1982:ix). The assumption is sound, but it is not an assumption about process. It is about product. And the pedagogy of the text is matched to this assumption. Yet another ad hoc solution is to stick firmly to the old tradition but to add a component that includes new theory and incorporates it into old practice so that it becomes so disguised as to be unrecognizable as TRADITION AND REVOLUTION 541

new. Our literature urges us in the name of communication to increase the opportunities for interaction in the classroom by asking students to work in pairs or in small groups. In one lesson, the students are given cards. One card says on it: You are in London on a business trip and you have Saturday free. You want to know if the British Museum is open and how much the entrance fee is. Another student has a card which says: The British Museum is open Monday through Saturday, 9:00 a.m. to 5:00 p.m. There is no entrance fee (Paulston 1974:360). The students have to pretend to call each other up and ask for information using the Could you please tell me . . .? questioning form. Cited in a theoretical article as a communicative act, such a technique as this one makes us question the way communicative competence is being viewed. For this technique seems to be only a slight modification of techniques anchored firmly in the positivist tradition, with language viewed as a set of patterns. The students do little more than what they do in a question and answer pattern drill: they address a question they have not made up to a designated person who has to be told the answer and then they wait for an answer that makes no difference to them whatsoever. These ad hoc solutions to problems in our discipline provide effective ways for teachers, theorists, and textbook-writers to cope with the anomalies around them, but they cannot really hide the conflicts or dispatch the anomalies that confront us as a profession. The scene before us in our discipline is still one of competing syllabuses, divergent methodologies, contradictory views of our profession, various attempts to deal with change, and innovative theories of second language acquisition, with all of us teachers trying to find our way in Terra Incognita. Kuhn describes the symptoms of the preparadigmatic stage of a revolution in a discipline as these: the proliferation of competing articulations, the willingness to try anything, the expression of explicit discontent, the recourse to philosophy and to debate over fundamentals (1970:91). Thats a paradigm shift and that certainly sounds like us. IS A NEW PARADIGM EMERGING? I have been able to point to some rather clear parallels between Kuhns theory and our own discipline for the first two questions I posed. I identified a traditional paradigm and saw signs of activity that fit the picture of the paradigm shift. It would not be difficult now to 542 TESOL QUARTERLY

continue the affirmative responses to our questions and to document from our literature ample evidence for the existence of an emerging paradigm. We have only to skim through articles in professional journals, conference presentations, accounts of research and of classroom activities, and new textbooks, and they would all provide abundant evidence for a set of beliefs about language with communication as the cornerstone. From various sources, we could quite easily compile a description of an emerging paradigm as one which: sees language as communication emphasizes real language use, as opposed to usage (Widdowson 1978) recommends a student-centered classroom encourages language acquisition develops humanistic, interpersonal approaches considers the nature of the learner, the learning process, and the learning environment. The old familiar themes of patterns/drills/rules/accuracy/sequence are being replaced by a new set: communication/functions/use/reality/affective approaches. This compilation of features certainly appears revolutionary next to our traditional paradigm. There is no doubt that we are trying to incorporate into our work much that is being discovered about language acquisition and the value of foreigner talk outside the classroom and teacher talk and interlanguage talk inside the classroom in order to provide comprehensible input for the language learner (Krashen 1981:137). Yet however much we learn about the value of interactive, communicative activities, however much we try to change our approach, it is interesting to note how tradition can still claim us in subtle ways. Examining a few instances of something as simple as one common type of classroom activity the asking of questionshelps us realize how singularly unrevolutionary we in fact can be. It has been found, for example, that in informal conversation between a native and a non-native speaker, that is, in a real communicative situation, information questions are asked (Where do you live?) and display questions (Whats the capital of France?) are virtually unknown (Long and Sato 1983). In ESL instruction, however, Long and Sato have found that display questions still predominate and only 14% of the questions we teachers ask are information questions. From data collected from the classes of experienced, informed, and eclectic teachers, it was found that contrary to the recommendations of many writers on second language teaching methodology, communicative use of the TRADITION AND REVOLUTION 543

target language makes up only a minor part of typical classroom activities, The researchers concluded that Is the clock on the wall? and Are you a student? are still the staple diet, at least for beginners (Long and Sato 1983). Jack Richards, too, finds from analyses of interviews, novels and plays, and classroom textbooks that, despite the evidence provided by an examination of real discourse, yes/no questions are regularly treated in the classroom as request for information/answer and seldom as request/grant (1977). So, to the question Are these apples fresh?, students are routinely taught to reply Yes, they are rather than I just bought them, Help yourself (1980:421). Even when a textbook tries hard to pay more attention to communicative aims by asking students to say Do you have any quarters? No, only nickels (Lanzano and Bodman 1981:19), that exercise is still just language practice and not real discourse. The students get no sense of why one speaker might want a quarter, no inkling of why the respondent feels inclined to offer nickels as a substitute. There is no context and no purpose. Interpersonal, yes, but the focus is still primarilyand traditionallyon form and discrete units of language. If no one needs or even pretends to need a quarter, what are these interacting persons really communicating to each other? What we have here is, of course, the paradox of a communicative drill, which surely has more to do with traditional drill than with contextually rich, meaningful communication (Taylor 1982:33). So, even though we can find evidence of an increase of communicative activities in the classroom and of a trend toward more real language use, we have to ask where the many traditional features are coming from. They could be just some practitioners interim ad hoc solutions to the anomalies in our discipline. Or their very existence, frequent and recurrent as it is, might indicate that the changes in our profession have been more in the design and procedure of our language teaching than in our underlying philosophical approach (J. Richards and Rodgers 1982). To discuss an emerging paradigm in the light of these issues, I now need to move from a description of phenomena in our field to speculation about the causes and contexts of those phenomena and about future directions. I propose that there are two reasons why it is not appropriate for us to regard our current emphasis on communication as an emerging paradigm for our discipline. First, it is occurring in a climate still too heavily influenced by our positivist tradition for the result to be truly revolutionary. In language teaching that has people talking to each other about quarters and nickels or about the hours of the British Museum for no identifiable purpose other than to practice a sentence pattern or a function, what is being emphasized is still the message 544 TESOL QUARTERLY

itself (language), as it was in the traditional paradigm. Students can be talking to each other in a language class and the focus can still be on the form of the language itself and not on the context of reality, the speaker, or the audience. Fanselow distinguishes between categories of language teaching content in the classroom: the content can be classified as language itself (with the purpose of practicing grammar, structure, and pronunciation), procedure (classroom administration), life (personal information or discussion of issues), or subject matter (skills and school subjects) (1977:24-25). To fit a new emphasis on communicative procedures into an approach which traditionally views language as the content of the language class is to push communication mm-e into line with drill and usage than with real life language use (Widdowson 1978). The current emphasis on communication has, I believe, been absorbed neatly into our positivist traditional framework. Far from superseding tradition, it has been assimilated into it. It opens up a variety of classroom procedures, such as interactive drills, while for the most part leaving undisturbed the underlying approach to language and language teaching. Our new interests are reflected in the surface features of our teaching and not in the deep structure of our theoretical underpinnings. Second, the view of language as a tool for communication still lacks the dimension of language that was missing from the traditional paradigmthe view of language as it is related to thought and to the mind of the language user. Whether the language used is a first or a second language, a person who uses a language construes the representations of meaning and is creative in that language. Yet, traditionally in second language teaching we have focused on form and have paid little attention to meaning, function, and purpose. With our new emphasis on communication, we are beginning to stress function along with form. But without a simultaneous emphasis on purpose, on the mind and intention of the new language user, language teaching continues to foster learning rather than acquisition, drills rather than discourse, and accuracy rather than fluency. All this doubt about whether Communicatia can really be the land of our new paradigm encourages us to look at alternatives and so leads us straight into our fourth question. IF A NEW PARADIGM ISNT EMERGING, WHERE ARE WE HEADED? Our readiness to espouse communication as a solution might be caused by a desire to establish certainty in our profession again. With our traditional paradigm in place, we relied on an established system of beliefs and we felt secure. We had our answers. Now, with confusion all around, we want to find new answers. Rather like TRADITION AND REVOLUTION 545

children on a long car journey, we keep wanting to know are we there yet? This desire to arrive somewhere drives us naturally to try to find solutions to problems. But we will never be able to make a statement like It isnt a star and it isnt a cornet, so it must be a planet. Nothing so neat, absolute, and supportable can complete the sentence: Language isnt a set of patterns and habits, so it must be . . . Looking for simple answers in a discipline like ours, concerned as it is with the complexities of language and mind, is too reductive. There are, however, some signs that, in spite of the current emphasis on language as communication, we are not settling for an easy answer and a reductive view of language. We are continuing to ask new questions, just like Gertrude Stein, who on her deathbed asked What is the answer? and, when none came, asked In that case, what is the question? (Toklas 1963:173). Both outside and inside our own field, questions are beginning to address the range of complexity of language by linking language with thought. If we turn again to the field of the teaching of writing, we see that scholars there are beginning to look at meaning as well as at form and function. They are asking more and more questions about the heuristic function of language, that function whereby language is used to discover and explore ideas, and not just to communicate them. They are interested in language as a tool to find out anew what we know and what we think. They emphasize, above all, the creative power of language to inform the user of language and not just the audience: Language is itself the great heuristic: words come into being as verbal generalizations . . . Clusters of words turn into syntax; it is the discursive nature of language, its tendency to run alongand thats what discourse meanslanguages tendency to be syntactical brings thought along with it (Berthoff 1981:38). Teachers of writing are, in turn, consulting authorities outside their field who are also addressing issues that link thought and language philosophically. The acknowledged mentors of many in the field of the teaching of writing are Susanne Langer, Michael Polanyi, I. A. Richards, and Lev Vygotsky. Note the titles of their influential works, respectively: Mind: An Essay in Human Feeling (1967, 1972); P e r s o n a l Knowledge (1958); The Philosophy of Rhetoric (1965); and T h o u g h t a n d L a n g u a g e ( 1 9 6 2 ) . Mind, feeling, knowledge, philosophy, a n d thought . . . These are words that suggest that the exploration of language is beginning to move away from positivist approaches and to involve more than communication. Within our field of second language learning there are signs that we, too, are beginning to look beyond formal accuracy and beyond isolated communicative exchanges. In the classroom, content other 546 TESOL QUARTERLY

than just language is a feature of English for Specific Purposes instruction, which teaches language and subject matter together. In immersion programs, particularly those in Canada, children learn the new language primarily through instruction in the content of the school subject matter (Swain 1981). The students thus become real language users while they are still language learners. Currans (1976) CounselingLearning/Community Language Learning also emphasizes real discourse, learner-initiated and with no limits imposed by a teacher. Learners begin forming and expressing ideas and engaging in real conversation in the very first classes. These pedagogical directions run parallel to some of the theoretical work in our discipline. More questions are now being raised about the relationship of language to the users thoughts, feelings, ideas, and knowledgein short, to the language users mind. Halliday, for one, distinguishes two other functions apart from the interpersonal function of language: a textual function, whereby language has to provide for making links with itself, and an ideational function, whereby language serves for the expression of content, that is, of the speakers experience of the real world, including the inner world of his own consciousness (1970: 143). Widdowson, on the other hand, includes the textual within the interpersonal function and makes a case for two principal functions of language: the conceptual and the communicative (1980:235), the latter being dependent upon the former. He describes language as working in two ways: On the one hand it provides for conceptual activity whereby clear and processible propositions are formulated in the mind. On the other hand, it provides the means whereby such propositions can be conveyed in the most effective manner for particular communicative purposes. These two functions, potentially in conflict, have to be reconciled by negotiation on every occasion of social use. This negotiation is realized in discourse (1980:238). He urges us not to neglect the conceptual function of language for it provides the individual with a means of establishing a relationship with his environment, of conceptualizing, and so, in some degree, controlling reality (1980:235). Widdowson is worried about excessive attention being paid to the communicative function of language while we neglect language used for thinking, formulating concepts, [and] fashioning propositions (1980:235). For him, creativity, that human capacity for making sense, is a crucial concept in language learning (1982:211-212). That language users adjust their speech for selfpresentational reasons is stressed by Beebe and Giles (in press), who see that a function of language other than conveying content is the presentation of self for the purpose of maintaining a positive social

TRADITION AND REVOLUTION

547

identity. Frank Smith, too, links language and self. He regrets the tendency, especially widespread in education, to regard language primarily as communication, as a vehicle for transmitting information from one person to another (1982:66). He urges us to pay more attention to the arena of the imagination for there we have a freedom to hypothesize, to test and to explore consequences that is rarely available to us in the world outside (1982:34). An intensive exploration of this arena of the imagination is being made in our field, particularly by scholars who are interested in the process of reading or writing. Studies by Hosenfeld (1976, 1977), Jones (in press), Zamel (1982, 1983), Carrell (1982), and Jones and Tetroe (1983) focus on how writers or readers interact with the text and with each other to create meaning and how that interaction is itself creative. The more we look at language in terms of concepts, thought, creativity, and imagination, the more questions we will continue to raise. And our questions will be philosophical questions rather than questions with quantifiable answers. In research, for example, it would not be surprising if more questions were raised about Krashens Input Hypothesis, which argues that theoretically one can acquire competence in a second language or a first language without ever producing it (1981:107-108). How much, we might ask, of what goes on in our mind, in our own negotiation with meaning, contributes to acquisition? How much of a learners understanding and comprehension comes from interaction with othersa speaker or a piece of writingand how much from the learners own active processes of forming propositions which then might or might not be translated into and adapted for a communicative act? Studies on the reading and writing process are beginning to look at how language users generate and negotiate meaning. Such studies will probably multiply and eventually be joined by studies of how learners acquire and generate concepts as well as morphemes and syntactic structures. We will focus attention not on the discrete features of the language the learners produce (their isolated product) but on how, why, when, and where they produce it (their process in context). The concept of research will also broaden to include contemplative firsthand accounts from teachers and students of what actually goes on when someone learns another language. We are all surrounded by research data, not necessarily data that can be counted and fed into the computer, but valuable data nevertheless. All teachers can be researchers when research includes observing and making connections with our knowledge and experience. In the classroom, teachers will ask questions about content and will seek out materials and methods that will allow their students the time,

548

TESOL QUARTERLY

space, and opportunity to generate and explore ideas by means of language. That could mean that learners will produce real instead of simulated discourse. It could also mean that with more questions being raised about writing as an authentic and valuable mode of learning, we might see writing become an integral part of every ESL program and of every class. With more questions about the use of language to discover and develop concepts, we will begin to pay as much attention to subject matter and to the interpretation of content as to form, function, and accuracy. Our students have been telling us for many years, What we need is vocabulary. They are probably right. They certainly do need vocabulary to engage with content, to turn clusters of words into syntax, and to use those words and that syntax to make sense of what is around them. Finally, all the questions we ask about language, thought, and meaning could have the effect that in our professional literature we will read as much about creativity and imagination as about information gaps, T-unit analysis, and the Spearman rank correlation coefficient. All of these questions that we face can only mean that we are still very far from a common body of beliefs and assumptions that makes up a new paradigm. We are, rather, still very firmly in the middle of a paradigm shift. We have a long way to go before we will be able to establish a canon of beliefs about language and language learning that forms a common core of our discipline and turns its back on the positivist tradition. In the meantime, Terra Incognita will remain our home. Let us look again at how Kuhn described the features of that land of the paradigm shift: the proliferation of competing articulations, the willingness to try anything, the expression of explicit discontent, the recourse to philosophy and to debate over fundamentals (1970:91). That sounds like fertile ground for language teachers. What could be better for a humanistic discipline than debate and philosophy? We can all expect to continue to confront more and more red sixes of spades.

ACKNOWLEDGMENTS
This article is a revised version of a paper presented at the 17th Annual TESOL Convention in Toronto, March 15-20, 1983. The author is grateful to Leslie Beebe, Ann Berthoff, Vivian Zamel, the TESOL Quarterly editor, Barry P. Taylor, and an anonymous TESOL Quarterly reviewer for their valuable suggestions on earlier drafts.

TRADITION AND REVOLUTION

549

THE AUTHOR
Ann Raimes is Coordinator of the Developmental English Program at Hunter College, the City University of New York. She is a frequent presenter at professional conferences, has published several articles on the teaching of writing, and is the author of Focus on Composition and Techniques in Teaching Writing (Oxford University Press), Currently she is a member of the Editorial Advisory Board of the TESOL Quarterly.

REFERENCES
Bailey, Kathleen M. 1980. An introspective analysis of an individuals language learning experience. In Research in second language acquisition, Robin C. Scarcella and Stephen D. Krashen (Eds.), 58-67. Rowley, Massachusetts: Newbury House. Beebe, Leslie M., and Howard Giles. In press. Speech accommodation theories: a discussion in terms of second language acquisition. International Journal of the Sociology of Language. Berthoff, Ann E. 1981. The making of meaning. Montclair, New Jersey: Boynton/Cook. Bloom, Lois (Ed.). 1978. Readings in language development. New York: John Wiley and Sons. Brown, H. Douglas. 1975. The next 25 years: shaping the revolution. In O n TESOL 75, Marina K. Burt and Heidi C. Dulay (Eds.), 80-85. Washington, D, C.: TESOL. Brown, H. Douglas. 1980. Principles of language learning and teaching. Englewood Cliffs, New Jersey: Prentice-Hall. Brown, Roger. 1973. A first language: the early stages. Cambridge, Massachusetts: Harvard University Press. Bruner, Jerome S. 1965. On knowing: essays for the left hand. N e w Y o r k : Atheneum. Bruner, Jerome S., and Leo Postman. 1949. On the perception of incongruity: a paradigm. Journal of Personality 18:206-223. Carrell, Patricia L. 1982. Cohesion is not coherence. TESOL Quarterly 1 6 (4):479-488. Chomsky, Noam. 1959. A review of B. F. Skinners V e r b a l b e h a v i o r . Language 35 (1):26-58. Curran, Charles A. 1976. Counseling-learning in second languages. A p p l e River, Illinois: Apple River Press. Emig, Janet. 1980. The tacit tradition: the inevitability of a multi-disciplinary approach to writing research. In Reinventing the rhetorical tradition, Aviva Freedman and Ian Pringle (Eds.), 9-17. Arkansas: L and S Books for the Canadian Council of Teachers of English. Emig, Janet. 1982. Inquiry paradigms and writing. College Composition and Communication 33 (1):64-75. Fanselow, John. 1977, Beyond Rashomon conceptualizing and describing the teaching act. TESOL Quarterly 11 (1):17-40.

550

TESOL QUARTERLY

Hairston, Maxine. 1982. The winds of change: Thomas Kuhn and the revolution in the teaching of writing. College Composition and Communication 33 (1): 76-88. Halliday, M. A. K. 1970. Language structure and language functions. In New horizons in linguistics, John Lyons (Ed.), 140-165. Harmondsworth: Penguin Books. Hines, Mary Elizabeth. 1973, 1980. Skits in English. New York: Regents. Hosenfeld, Carol. 1976. Learning about learning: discovering our students strategies. Foreign Language Annals 9 (2):117-129. Hosenfeld, Carol. 1977. A preliminary investigation of the reading strategies of successful and non-successful second language learners. System 5:110123. Hymes, Dell H. 1966. On communicative competence. Excerpts in T h e c o m m u n i c a t i v e a p p r o a c h t o l a n g u a g e t e a c h i n g , C . J. Brumfit and K. Johnson (Eds.), 5-24. Oxford: Oxford University Press, 1979. Jones, C. Stanley. In press. Attention to rhetorical information while composing in a second language. Proceedings of the 4th Los Angeles Second Language Research Forum. Jones, C. Stanley, and Jacqueline Tetroe. 1983. Observing ESL writing. Paper presented at the 17th Annual TESOL Convention, Toronto, March, 1983. Krashen, Stephen D. 1981. Second language acquisition and second language learning. Oxford: Pergamon Press. Kuhn, Thomas S. 1970. The structure of scientific revolutions. Second Edition. Chicago: University of Chicago Press. Kunz, Linda Ann. 1980. English as a second language. In Teaching basic skills in college, Alice Stewart Trillin and Associates, 145-207. San Francisco: Jossey-Bass. Lado, Robert. 1964. Language teaching: a scientific approach. New Y o r k : McGraw Hill. Langer, Susanne. 1967, 1972. Mind: an essay on human feeling. Volumes 1 and 2. Baltimore: Johns Hopkins Press. Lanzano, Michael, and Jean Bodman. 1981. Milk and honey: an ESL series for adults, Book 1. New York: Harcourt Brace Jovanovich. Long, Michael H., and Charlene J. Sato. 1983. Classroom foreigner talk discourse: forms and functions of teachers questions. In C l a s s r o o m oriented research on second language acquisition, Herbert W. Seliger and Michael H. Long (Eds.). Rowley, Massachusetts: Newbury House. Newmark, Leonard. 1966. How not to interfere with language learning. International Journal of American Linguistics 32 (1):77-83. Reprinted in 1979 in The communicative approach to language teaching, C. J. Brumfit and K. Johnson (Eds.), 160-166. Oxford: Oxford University Press. Paulston, Christina Bratt. 1974. Linguistic and communicative competence. TESOL Quarterly 8(4):347-362. Polanyi, Michael. 1958. Personal knowledge. Chicago: University of Chicago Press. Reid, Joy M. 1982. The process of composition. E n g l e w o o d C l i f f s , N e w Jersey: Prentice-Hall.

TRADITION AND REVOLUTION

551

Richards, I. A. 1965. The philosophy of rhetoric. New York: Oxford University Press. Richards, Jack C. 1977. Answers to yes-no questions. English Language Teaching Journal 31 (2):136-141. Richards, Jack C. 1980. Conversation. TESOL Quarterly 14 (4) :413-432. Richards, Jack C., and Ted Rodgers. 1982. Method: approach, design, and procedure. TESOL Quarterly 16 (2):153-168. Schumann, Francine M. 1980. Diary of a language learner: a further analysis. In Research in second language acquisition, Robin C. Scarcella and Stephen D. Krashen (Eds.), 51-57, Rowley, Massachusetts: Newbury House. Schumann, Francine M., and John H. Schumann. 1977. Diary of a language learner: an introspective study of second language learning. In On TESOL 77, H. Douglas Brown, Carlos Yorio, and Ruth Crymes (Eds.), 241-249. Washington, D. C,: TESOL. Smith, Frank. 1982. Writing and the writer. New York: Holt, Rinehart and Winston. Stevick, Earl W. 1976. Memory, meaning and method. Rowley, Massachusetts: Newbury House. Stevick, Earl W. 1982. Learning a foreign language: the natural ways. In O n TESOL 81, Mary Hines and William Rutherford (Eds.), 1-10. Washington, D, C.: TESOL. Swain, Merrill. 1981. Target language use in the wider environment as a factor in its acquisition. In New dimensions in second language acquisition research, Roger W. Andersen (Ed.), 109-122. Rowley, Massachusetts: Newbury House. Taylor, Barry P. 1982. In search of real reality. TESOL Quarterly 16 (1) :29-42. Toklas, Alice B. 1963. What is remembered. New York: Holt, Rinehart and Winston. Vygotsky, Lev. 1962. Thoughts and language. Translated by E. Hanfman and G. Vakar. Cambridge, Massachusetts: M.I.T. Press. Widdowson, Henry G. 1978. Teaching language as communication. Oxford: Oxford University Press. Widdowson, Henry G. 1980. Conceptual and communicative functions in written discourse. Applied Linguistics 1 (3):234-243. Widdowson, Henry G. 1982. The use of literature. In On TESOL 81, Mary Hines and William Rutherford (Eds.), 203-214. Washington D. C.: TESOL. Young, Richard. 1978. Paradigms and problems: needed research in rhetorical invention. In Research on composing, Charles R. Cooper and Lee Odell (Eds.), 29-47. Urbana, Illinois: National Council of Teachers of English. Zamel, Vivian. 1982. Writing: the process of discovering meaning. TESOL Quarterly 16 (2):195-210. Zamel, Vivian. 1983. The composing processes of advanced ESL students: six case studies. TESOL Quarterly 17 (2):165-187.

552

TESOL QUARTERLY

TESOL QUARTERLY, VoI. 17, No.4, December 1983

Schema Theory and ESL Reading Pedagogy

PATRICIA L. CARRELL
Southern Illinois University at Carbondale

JOAN C. EISTERHOLD
Northwestern University

This article discusses the important role of background knowledge in a psycholinguistic model of EFL/ESL reading and demonstrates the relevance of schema-theoretic views of reading to the teaching of reading to EFL/ESL students. According to schema theory, reading comprehension is an interactive process between the text and the readers prior background knowledge (Adams and Collins 1979, Rumelhart 1980). Reading comprehension involves ones knowledge of the world, which may be culturally based and culturally biased. Classroom implications of the schema-theoretic view of reading for EFL/ESL reading pedagogy are discussed, with techniques suggested for bringing about reader-centered EFL/ESL reading.

Every act of comprehension involves ones knowledge O f the world as well. (Anderson, Reynolds, Schallert, and Goetz 1977:369) INTRODUCTION The idea expressed by the above quote is certainly not new, but it is one worth reminding ourselves of when we consider comprehension in a second or foreign language, and specifically reading comprehension in EFL/ESL. If, as Immanuel Kant claimed as long ago as 1781, new information, new concepts, new ideas can have meaning only when they can be related to something the individual already knows (Kant 1781/1963), this applies as much to second language comprehension as it does to comprehension in ones native language. Yet, traditionally in the study of second language comprehension (as much as, if not more so than, in the study of first language comprehension), the emphasis has been almost exclusively on the language to be comprehended and not on the comprehended (listener or reader). In this perspective, each word, each well-formed sentence, and every well-formed text passage 553

is said to have a meaning. Meaning is often conceived to be in the utterance or text, to have a separate, independent existence from both the speaker or writer and the listener or reader. Also in this view, failures to comprehend a non-defective communication are always attributed to language-specific deficitsperhaps a word was not in the readers vocabulary, a rule of grammar was misapplied, an anaphoric cohesive tie was improperly coordinated, and so on. Recent empirical research in the field which has come to be known as schema theory has demonstrated the truth of Kants original observation and of the opening quote from Anderson et al. Schema theory research has shown the importance of background knowledge within a psycholinguistic model of reading. The purpose of this article is twofold. Our first goal is to give a brief overview of schema theory as part of a reader-centered, psycholinguistic processing model of EFL/ESL reading. This goal is addressed in the first part of this article, in which we discuss how EFL/ESL reading comprehension involves background knowledge which goes far beyond linguistic knowledge. Our second purpose is to explore the relationship of culture-specific background knowledge and EFL/ESL reading methodology and is taken up in the second part of the article, where we review this relationship as it has been discussed in the extant methodology literature. We illustrate this discussion of the culturally based and culturally biased nature of background knowledge with sample reading passages which have actually caused comprehension problems for EFL/ESL students. Finally, we suggest a variety of techniques and classroom activities for accommodating this phenomenon in a readercentered EFL/ESL reading program. THE PSYCHOLINGUISTIC MODEL OF READING During the past decade, EFL/ESL reading theory has come under the influence of psycholinguistics and Goodmans (1967, 1971, 1973a) psycholinguistic model of reading (see also Smith 1971). Goodman has described reading as a psycholinguistic guessing game (1967) in which the reader reconstructs, as best as he can, a message which has been encoded by a writer as a graphic display (1971:135). Goodman views this act of the construction of meaning as being an ongoing, cyclical process of sampling from the input text, predicting, testing and confirming or revising those predictions, and sampling further. In this model, the reader need not (and the efficient reader does not) use all of the textual cues. The better the reader is able to make correct predictions, the less confirming via the text is necessary, that is, the less visual perceptual information the reader requires:

554

TESOL QUARTERLY

. . . the reader does not use all the information available to him. Reading is a process in which the reader picks and chooses from the available information only enough to select and predict a language structure which is decodable. It is not in any sense a precise perceptual process (Goodman 1973b:164). These views are by now generally well known and widely accepted in our field. Coady (1979) has elaborated on this basic psycholinguistic model and has suggested a model in which the EFL/ESL readers background knowledge interacts with conceptual abilities and process strategies, more or less successfully, to produce comprehension (see Figure 1).

By conceptual ability, Coady means general intellectual capacity. By processing strategies, Coady means various subcomponents of reading ability, including many which are also more general language processing skills which also apply to oral language (e. g., grapheme-morphophoneme correspondences, syllable-morpheme information, syntactic information [deep and surf ace], lexical meaning, and contextual meaning). Coady says little more about the role of b a c k g r o u n d knowledge other than to observe that background knowledge becomes an important variable when we notice, as many have, that students with a Western background of some kind learn English faster, on the average, than those without such a background (Coady 1979:7). Coady also suggests that background knowledge may be able to compensate for certain syntactic deficiencies: The subject of reading materials should be of high interest and relate well to the background of the reader, since strong semantic input can help compensate when syntactic control is weak. The interest and background knowledge will enable the student to comprehend at a reasonable rate and SCHEMA THEORY AND READING 555

keep him involved in the material in spite of its syntactic difficulty (Coady 1979:12). It is this third factor, background knowledge, that has been the most neglected in EFL/ESL reading. Even though the psycholinguistic model of reading is seen as an interaction of factors, it has generally failed to give sufficient emphasis to the role of background knowledge. Recent research indicates that what the reader brings to the reading task is more pervasive and more powerful than the general psycholinguistic model suggests: More information is contributed by the reader than by the print on the page. That is, readers understand what they read because they are able to take the stimulus beyond its graphic representation and assign it membership to an appropriate group of concepts already stored in their memories . . . The reader brings to the task a formidable amount of information and ideas, attitudes and beliefs. This knowledge, coupled with the ability to make linguistic predictions, determines the expectations the reader will develop as he reads. Skill in reading depends on the efficient interaction between linguistic knowledge and knowledge of the world (Clarke and Silberstein 1977:136-137). THE SCHEMA THEORY MODEL The role of background knowledge in language comprehension has been formalized as schema theory (Bartlett 1932, Rumelhart and Ortony 1977, Rumelhart 1980), which has as one of its fundamental tenets that text, any text, either spoken or written, does not by itself carry meaning. Rather, according to schema theory, a text only provides directions for listeners or readers as to how they should retrieve or construct meaning from their own, previously acquired knowledge. This previously acquired knowledge is called the readers background knowledge, and the previously acquired knowledge structures are called schemata (Bartlett 1932, Adams and Collins 1979, Rumelhart 1980).1 According to schema theory, comprehending a text is an interactive process between the readers background knowledge and the text. Efficient comprehension requires the ability to relate the
1 Other

closely related concepts, which are technically distinct from schemata but which may be thought of as part of the same general, cognitive approach to text processing, are scripts, plans, and goals (Schank and Abelson 1977), frames (Minsky 1975, Fillmore 1976, Tannen 1979), expectations (Tannen 1978), and event chains (Warren, Nicholas, and Trabasso 1979). All of these terms emanate from basic research at the intersection of artificial intelligence, cognitive psychology, and linguistics in the new discipline called cognitive science. These terms are not identical or even interchangeable; however, they may all be broadly characterized as part of a schema-theoretical orientation to text processing. Carrell (1983a) gives an extensive overview of schema theory and the relevant theoretical literature as well as the empirical research in first language processing (children and adults) and second language processing (adults).

556

TESOL QUARTERLY

textual material to ones own knowledge. Comprehending words, sentences, and entire texts involves more than just relying on ones linguistic knowledge. As the opening quote from Anderson et al. points out, every act of comprehension involves ones knowledge of the world as well (Anderson et al. 1977:369). According to schema theory, the process of interpretation is guided by the principle that every input is mapped against some existing schema and that all aspects of that schema must be compatible with the input information. This principle results in two basic modes of information processing, called bottom-up and top-down processing. Bottom-up processing is evoked by the incoming data; the features of the data enter the system through the best fitting, bottom-level schemata. Schemata are hierarchically organized, from most general at the top to most specific at the bottom. As these bottom-level schemata converge into higher level, more general schemata, these too become activated. Bottom-up processing is, therefore, called data-driven. Topdown processing, on the other hand, occurs as the system makes general predictions based on higher level, general schemata and then searches the input for information to fit into these partially satisfied, higher order schemata. Top-down processing is, therefore, called conceptually-driven. An important aspect of top-down and bottom-up processing is that both should be occurring at all levels simultaneously (Rumelhart 1980). The data that are needed to instantiate, or fill out, the schemata become available through bottom-up processing; top-down processing facilitates their assimilation if they are anticipated by or consistent with the listener/readers conceptual expectations. Bottom-up processing ensures that the listeners/readers will be sensitive to information that is novel or that does not fit their ongoing hypotheses about the content or structure of the text; top-down processing helps the listeners/readers to resolve ambiguities or to select between alternative possible interpretations of the incoming data. To illustrate the effects of background knowledge, schematic interpretation, and the simultaneity of top-down and bottom-up processing, consider the following mini-text (originally from Collins and Quillian 1972; discussed in Rumelhart 1977:267): The policeman held up his hand and stopped the car. In the process of trying to understand this sentence, we try to relate it to something familiar, some schema which will account for the event described. There are many schemata possible, but perhaps the most likely is the one involving a traffic cop who is signaling to a driver of a car to stop. Notice that when we interpret this mini-text against that schema, a number of related concepts come to the fore which are not SCHEMA THEORY AND READING 557

literally mentioned in the text. In particular, we imagine that the car has a driver and that the policeman got the car to stop through signaling to the driver, who then put on the brakes of the car, which, in turn, caused the car to stop. The proximal cause of the cars stopping is, in this interpretation, the operation of the cars brakes. Further, the significance of the policemans holding up his hand is that of a signal to the driver to stop. This fact is neither stated in the sentence nor is it even in the direct visual perception of such a situation, but is rather a fact in our prior cultural knowledge about the way traffic police are known to communicate with automobile drivers. Notice how the interpretation of the text would change if the policeman were known to be Superman and the car were known to be without a driver. A completely different schema would be required to understand the text. Notice how the relationship of the policemans holding up his hand and the cars stopping takes on an entirely different interpretation when the text is interpreted against the Superman schema. Now, holding up the hand is not interpreted as a signal at all, but rather the direct physical mechanism for stopping the car. In this interpretation, the hand actually comes into physical contact with the car and is the proximal physical cause of the cars halting. The brakes of the car do not come into play in this schema. Notice how sets of inferential reading comprehension questions would receive diametrically opposed answers depending upon which of the two schemata was activated in the mind of the reader:

Mary heard the ice cream man coming down the street. She remembered her birthday money and rushed into the house . . . (1977:265) Upon reading just these few lines, most readers are able to construct a rather complete interpretation of the text. Presumably, Mary is a little girl who heard the ice cream man coming and wanted to buy some ice cream from this ice cream man. Buying ice cream costs money, so she had to think quickly of a source of funds. She remembered some money which she had been given for her birthday and which, presumably, was in the house. So she hurried into the house to try to 558 TESOL QUARTERLY

get the money before the ice cream man arrived. Of course, the text does not say all of this; we readers are inferring a lot of this in giving the text an interpretation. Other interpretations are also possible. Yet, most readers will probably give this text an interpretation quite similar to the one suggested here, and most readers will retain this interpretation unless some contradictory information is encountered. Notice what happens if the reader next encounters the phrase: . . . and locked the door.2 The reader is unable to fit this new piece of textual input information into the developing interpretation. The reader is forced to revise the interpretation in such a way as to make this new information compatible with the previous informationto make the whole text cohere. If there were no such thing as schemata guiding the developing interpretation in a top-down processing mode, causing the reader to make conceptual predictions about the meaning of the text, then why would encountering the added phrase cause the reaction it does in the reader? What has happened, we claim, is that as long as the incoming information being processed through bottom-up processing and the conceptual predictions being made through top-down processing are compatible, we have a satisfactory interpretation of the text. When we encounter a mismatch between the top-down predictions and the bottom-up information, we are forced to revise the interpretation in such a way as to make the two compatible once again. In this example, we must revise our interpretation to accommodate the information about Marys locking the door. Perhaps we infer that for some reason Mary is afraid that the ice cream man might steal her birthday money and that she locks the door to protect it and herself. We believe these two examples vividly demonstrate the existence and operation of schemata in the process of text interpretation. Thus, it seems clear that readers activate an appropriate schema against which they try to give a text a consistent interpretation. To the extent that they are successful, we may say that they have comprehended the text. However, one potential source of reading difficulties may be that the reader has a consistent interpretation for the text, but it may not be the one intended by the author. Nonetheless, the basic point is that much of the meaning understood from a text is really not actually in the text, per se, but in the reader, in the background or schematic knowledge of the reader. What is understood from a text is a function of the particular schema that is activated at the time of processing (i.e., reading) the text.
2 This

example was offered by Charles J. Fillmore in a class lecture at the University of California, Berkeley, in 1980.

SCHEMA THEORY AND READING

559

In seeking to understand the role of background knowledge in reading comprehension, it is often useful to draw a distinction between formal schemata (background knowledge of the formal, rhetorical organizational structures of different types of texts) and content schemata (background knowledge of the content area of a text) (Carrell 1983b). In other words, one type of schema which readers are said to possess is background knowledge about, and expectations of, differences among rhetorical structures, such as differences in genre, differences in the structure of fables, simple stories, scientific texts, newspaper articles, poetry, and so forth. Our schema for simple stories, for example, includes the information that the story should have, minimally, a setting, a beginning, a development, and an ending. Also for simple stories, Mandler (1978) distinguishes between schemata for causally connected and temporally connected stories. For expository texts, Meyer and her colleagues (Meyer 1975, 1977, 1981; Meyer and Rice 1982; Meyer and Freedle, in press) recognize five different types of expository rhetorical organization: collection list, causation cause and effect, response problem and solution, comparison comparison and contrast, and description attribution. Each of these types, they say, represents a different abstract schema of ways writers organize and readers understand topics. In schema theory research, this type of formal schematic knowledge is usually contrasted with content schematic knowledge, which is claimed to be background knowledge about the content area of a text, such as a text about washing clothes, celebrating New Years Eve in Hawaii or Halloween in Carbondale, or about the economy of Mexico, the history of Canada, problems of nuclear breeder reactors, and so forth. A readers failure to activate an appropriate schema (formal or content) during reading results in various degrees of non-comprehension. This failure to activate an appropriate schema may either be due to the writers not having provided sufficient clues in the text for the reader to effectively utilize a bottom-up processing mode to activate schemata the reader may already possess, or it may be due to the fact that the reader does not possess the appropriate schema anticipated by the author and thus fails to comprehend. In both instances, there is a mismatch between what the writer anticipates the reader can do to extract meaning from the text and what the reader is actually able to to. The point is that the appropriate schemata must exist and must be activated during text processing. One of the most obvious reasons why a particular content schema may fail to exist for a reader is that the schema is culturally specific and is not part of a particular readers cultural background. Studies by

560

TESOL QUARTERLY

Steffensen, Joag-dev, and Anderson (1979), Johnson (1981), and Carrell (1981a) have shown that the implicit cultural content knowledge presupposed by a text interacts with the readers own cultural background knowledge of content to make texts whose content is based on ones own culture easier to read and understand than syntactically and rhetorically equivalent texts based on a less familiar, more distant culture. Other research has shown general effects of content schemata on EFL/ESL reading comprehension. Johnson (1982) has shown that a text on a familiar topic is better recalled by ESL readers than a similar text on an unfamiliar topic. Hudson (1982) reports a study showing an interaction between overall linguistic proficiency in ESL and contentinduced schematic effects in ESL reading comprehension. Specifically, that study demonstrates the facilitating effects on comprehension of explicitly inducing content schemata through pre-reading activities, especially at the beginning and intermediate proficiency levels, as compared to two other methods of inducing content schemata (through vocabulary activities and read-reread activities). Finally, Alderson and Urquhart (1983) have found a discipline-specific effect of content background knowledge in measuring reading comprehension in ESP/EST. Several recent studies have shown the effects of formal, rhetorical schemata in EFL/ESL. In a study by Carrell (1981 b), two groups of university-bound, intermediate-level ESL subjects each read a different type of simple storyone type well structured according to a simple story schema structure and the other type deliberately violating the story schema structure. Results showed that when stories violating the story schema are processed by second language learners, both the quantity of recall and the temporal sequences of recall are affected. In other words, when the content is kept constant but the rhetorical structure is varied, second language reading comprehension is affected. Recent studies done in the area of contrastive rhetoric (Kaplan 1966) also demonstrate the effects of formal schemata on both the comprehension and production of written texts in a second language (Ostler and Kaplan 1982). In particular, Hinds research (1982, 1983) shows the contrasting effects on different groups of readers of typical Japanese rhetorical organization and typical English rhetorical organization. Burtoff (1983) has found differences among the typical rhetorical patterns of expository prose produced by writers with different formal schemata according to their native-language/native-culture backgrounds. Thus, a growing body of empirical research attests to the role of both content and formal schemata in EFL/ESL reading comprehension

SCHEMA THEORY AND READING

561

and to the potential cultural specificity of both types of schemata. In the following sections, we focus on the implications of the cultural specificity of content schemata for EFL/ESL readers and EFL/ESL reading methodology. IMPLICATIONS FOR EFL/ESL READERS Given the role of content schemata in reading comprehension, there are obvious implications for the EFL/ESL reader. The background knowledge that second language readers bring to a text is often culture-specific. Hudson notes that the reading problems of the L2 reader are not due to an absence of attempts at fitting and providing specific schemata . . . Rather, the problem lies in projecting appropriate schemata (Hudson 1982:9). Second language readers attempt to provide schemata to make sense of texts, and they do so persistently. However, these efforts will fail if the reader cannot access the appropriate existing schemata, or if the reader does not possess the appropriate schemata necessary to understand a text. Most commonly, accessing appropriate content schemata depends initially on textual cues; the graphic display must be somehow reconstructed by the reader as meaningful language. At this point, general language processing skills are most important. For second language readers, then, obviously some language proficiency is required to activate relevant schemata, and it is not surprising that failures to access appropriate schemata (i.e., comprehend) are often interpreted solely as deficiencies in language processing skills. Consequently, poor readers are encouraged to expand their vocabularies and to gain greater control over complex syntactic structures in order to improve reading comprehension. Indeed, some reading problems are related to such language skill deficiencies. However, as we have noted, reading comprehension depends crucially on the readers being able to relate information from the text to already existing background knowledge. In the EFL/ESL classroom, we must be particularly sensitive to reading problems that result from the implicit cultural knowledge presupposed by a text. A review of the literature in EFL/ESL methodology shows that the role of cultural knowledge as a factor in reading comprehension has been an issue for some time. Fries (1945, 1963) talked about meaning at the social-cultural levelthat is, the meaning that transcends the language code and is related to the background knowledge of the native speakers of that code. Reading comprehension occurs when the total meaning of a passage is fitted into this network of information organized in ways meaningful to a 562 TESOL QUARTERLY

society. The following passage from an ESL reading text3 illustrates Fries concept of social-cultural meaning: By voting against mass transportation, voters have chosen to continue on a road to ruin. Our interstate highways, those much praised golden avenues built to whisk suburban travelers in and out of downtown have turned into the worlds most expensive parking lots. That expense is not only economicit is social. These highways have created great walls separating neighborhood from neighborhood, disrupting the complex social connections that help make a city livable (Baudoin et al. 1977:159). In reading this passage, some ESL students fail to perceive the connection between mass transportation and highways. In the United States, where individual ownership of cars results in an overabundance of highways and a reduced need for mass transportation, this passage makes sense. Sometimes, however, students perceive that highways are built for mass transportation, which renders this passage (and especially the critical reading question which asks whether the author supports the idea of mass transportation) at best illogical, at worst incomprehensible. The social-cultural meaning in this passage relates to the culturespecific schema of the cars/mass transportation opposition. Furthermore, comprehension can also be related to semantic associations available when a schema is accessed. The notion of interstate highways, here referred to narrowly as those in urban areas, invites the semantic associations of crowding, congestion, and rush hour traffic. The meaning of the phrase the worlds most expensive parking lots i s associated with, and can only be understood with reference to, this specific urban highway subschema. Elsewhere in the EFL/ESL methodology literature, Rivers (1968) recommends that the strong bond between culture and language must be maintained for the student to have a complete understanding of the meaning of language. She believes that differences in values and attitudes are one of the main sources of problems in foreign language learning. Culture-specific values can be a significant factor in comprehension if the values expressed by the text differ from the values held by the reader. Devout Muslim students, for example, tend to have problems with the following passage: There is a question about the extent to which any one of us can be free of a prejudiced view in the area of religion (Baudoin et al. 1977:185).
3 The

passages chosen to illustrate reading problems related to culture-specific background information are all drawn from Readers Choice ( Baudoin, Bober, Clarke, Dobson, and Silberstein 1977), a widely used and widely respected ESL reading text. The difficulties noted with the passages from this text are in no way intended as a criticism of the book, which we consider an excellent text based on psycholinguistic principles. Rather, these passages were chosen because they caused actual classroom problems and because they illustrate often subtle or hidden problems which we, as EFL/ESL teachers, may find difficult to identify.

SCHEMA THEORY AND READING

563

While this sentence is excellent for developing critical reading skills, the mention of religion in this context does not coincide with Islamic values. A subsequent exercise requires the student to analyze the relation of the original text to the following sentence: Because we cant be free of prejudice in the area of religion, we should not practice a religion. One student refused to even consider the premise of this sentence; his only comment: For me, its false. More recently, Rivers and Temperley (1978) have emphasized the importance of providing background information, explaining highfrequency but culturally loaded terms, and using illustrations with reading passages to provide additional meaning to texts. The important point is that problems with individual lexical items may not be as pervasive as problems related to the absence of appropriate generalized information assumed by the writer and possessed by a reader sharing that writers cultural background. The relevance of appropriate generalized, underlying information is illustrated by the following text: Although housewives still make up the majority of volunteer groups, male participation is reported on the rise nationwide as traditional distinctions between mens work and womens work begin to fade (Baudoin et al. 1977:184). The phrase volunteer groups requires appropriate underlying information before this sentence can be understood. Although the lexical items volunteer and groups were clearly understood by one student, the concept of volunteer groups (predominantly female, unpaid social workers) was clearly not understood since he wondered if these women had volunteered to be housewives. Paulston and Bruder (1976) also discuss covert information and reading. Proficient readers, they say, must draw on their own experience in order to supply a semantic component to a message. They argue that texts with familiar settings and even specialized lowfrequency vocabulary are appropriate (even though the texts may feel as if they are not appropriate) because they are relevant to the students world (and are, thus, easier to read). Robinett (1978) agrees that covert cultural information is a factor in reading performance and suggests that the teacher facilitate reading by providing specific background experience. When covert information is assumed by the writer, it must be supplied by the reader and is sometimes done so erroneously. I saw by the clock of the city jail that it was past eleven, so I decided to go to the newspaper immediately (Baudoin et al. 1977:83).

564

TESOL QUARTERLY

After reading this sentence, one student was convinced that the writer had been in jail at the time because, as he said, an outside clock is only on a church. He had concluded that the only place the writer could have seen the clock of the city jail was from inside the jail itself. If this sentence merely sets the scene, then this misinterpretation is insignificant. However, if the misreading causes the reader to consider such a scene significant (when it is not), or to dismiss it as in-significant (when it is not), then a serious comprehension problem has resulted. Finally in the methodology literature, Marquardts work (1967, 1969) is representative of the pedagogical approach that holds that reading should be in the literature of the target culture for the express purpose of teaching that culture to foreign students. Such literature, however, must be chosen carefully. Consider the following passage from Cheaper by the Dozen: Mother the psychologist and Dad the motion study man and general contractor decided to look into the new field of the psychology of management, and the old field of psychologically managing a houseful of children. They believed that what would work in the home would work in the factory, and what would work in the factory would work in the home. Dad put the theory to a test shortly after we moved to Montclair. The house was too big for Tom Grieves, the handyman, and Mrs. Cunningham, the cook, to keep in order. Dad decided we were going to have to help them, and he wanted us to offer the help willingly. He had found that the best way to get cooperation out of employees in a factory was set up a joint employer-employee board, which would make work assignments on a basis of personal choice and aptitude. He and Mother set up a Family Council, patterned after an employer-employee board. The Council met every Sunday afternoon, immediately after dinner (Baudoin et al. 1977:91). This text would require considerable background teaching before the text itself could teach anything (even if we considered what it has to teach to be culturally relevant). Using literature to teach culture m a y be the most direct way to teach culture, but it certainly implies thorough background preparation and may, in fact, not be the best way to teach language. There is much in these methods that is of value, but in light of the broad message behind the schema-theoretic view of reading, are they sufficiently sensitive to cross-cultural interference at all levels of meaning? The factors noted above are important not just to ground words and phrases for the reader. Rather, notions such as socialcultural meaning, culture-specific values, and covert information r e f e r to different aspects of the same problem, and that is how to deal with reading difficulties caused by the mismatch of the background knowledge presupposed by the text and the background knowledge possessed SCHEMA THEORY AND READING 565

by the reader. A schema-theoretic view of reading suggests the pervasive effects of such a mismatch and requires our being sensitive to these reading difficulties on a more global level. CLASSROOM ACTIVITIES Our immediate goal as EFL/ESL reading teachers is to minimize reading difficulties and to maximize comprehension by providing culturally relevant information. Goodman puts the issue into focus when he says that even highly effective readers are severely limited in comprehension of texts by what they already know before they read. The author may influence the comprehensibility of a text particularly for specific targeted audiences. But no author can completely compensate in writing for the range of differences among all potential readers of a given text (Goodman 1979:658). Since no author can compensate for the individual variation among readers, especially readers from different cultural backgrounds, this is one of the roles of the teacher in the EFL/ESL reading classroom. As teachers we can approach this problem by manipulating either one of the two variables: the text and/or the reader. Text What can we do with texts to minimize cultural conflicts and interference and to maximize comprehension? For the beginning reader, the Language Experience Approach (LEA) (Rigg 1981) is an excellent way to control vocabulary, structure, and content. The basic LEA technique uses the students ideas and the students own words in the preparation of beginning reading materials. The students decide what they want to say and how to say it, and then dictate to the teacher, who acts as a scribe. LEA works when the students beginning reading materials, developed by them with the teachers help, have the students ideas in their own words. LEA works because students tend to be able to read what they have just said. The students, in effect, write their own texts, neutralizing problems of unfamiliar content. Another way to minimize interference from the text is to encourage narrow reading, as suggested by Krashen (1981). Narrow reading refers to reading that is confined to a single topic or to the texts of a single author. Krashen suggests that narrow reading, and perhaps narrow input in general, is more efficient for second language acquisition (Krashen 1981:23). Reading teachers usually provide short and varied selections which never allow students to adjust to an authors style, to become familiar with the specialized vocabulary of the topic, 566 TESOL QUARTERLY

or to develop enough context to facilitate comprehension. Rather, such selections force students to move from frustration to frustration. However, students who read either a single topic or a single author find that the text becomes easier to comprehend after the first few pages. Readers adjust either to the repeated vocabulary of a particular topic or to the particular style of a writer. Furthermore, repetitions of vocabulary and structure mean that review is built into the reading. The significant advantage from the schema-theoretic point of view is that schemata are repeatedly accessed and further expanded and refined, resulting in increased comprehension. The third possibility of text facilitation is to develop materials along the lines of those proposed by Paulston and Bruder (1976). As we have noted, they suggest using texts with local settings and specialized lowfrequency vocabulary. These materials might be student or local newspapers, pamphlets, brochures, or booklets about local places of interest. English travel guides or National Geographic- type articles from the students own countries are also good sources for the EFL/ESL reader. Finally, Sustained Silent Reading (SSR) is an excellent activity for ESL readers. Through silent reading of texts, students become selfdirected agents seeking meaning. To be effective, however, an SSR program must be based on student-selected texts so that the students will be interested in what they are reading. Students select their own reading texts with respect to content, level of difficulty, and length. What is important, from our point of view, is that readers tend to be interested in reading texts that are relevant to their own experiences. Students who choose their own texts are, in effect, also providing their own appropriate background knowledge for understanding the text. Reader Instead of, or in addition to, text control, we also need to consider what we can do with the readers themselves. Providing background information and previewing content for the reader seem to be the most obvious strategies for the language teacher. We want to avoid having students read material cold. Asking students to manipulate both the linguistic and cultural codes (sometimes linguistically easy but culturally difficult, and vice versa) is asking too much. Providing background information and previewing are particularly important for the less proficient language student (see the findings of Hudson 1982). These readers are more word-bound, and meaning tends to break down at the word level. Thus, less proficient students tend to have vocabulary acquisition emphasized and, as such, are encouraged to do a lot of specific (and less efficient) word-by-word SCHEMA THEORY AND READING 567

processing exclusively in a bottom-up processing mode. Readers who are more proficient in a language tend to receive content previews because they are no longer as susceptible to vocabulary and structure difficulties in reading. As a result, these more proficient students are encouraged to do more global, predictive (and more efficient) processing in the top-down processing mode. One thing we surely want to remind ourselves of, however, is that less proficient readers also need familiar content selections and/or content preview as much as, if not more than, more proficient readers. Illustrations may be particularly appropriate for students with minimal language skills. Providing the semantic content component for low-level readers will free them to focus on vocabulary and structure expressive of that content. Previewing is an important activity in the reading classroom, but it is not necessarily a process of simply providing a preliminary outline of what is to be read. Sometimes, it involves teaching a key concept which is culturally loaded, such as the one in the short story T h e Lottery (Baudoin et al. 1977:140-145). If one does not understand the process or purpose of a lottery, then this short story about one woman who wins and is then killed by her neighbors will be totally incomprehensible. In this case, a discussion of lotteries before assigning the reading would be absolutely necessary. Previewing can also include presenting specialized vocabulary and structures that the teacher predicts will cause difficulties. In the mass transportation passage cited earlier, students who could not come up with the appropriate background information also had difficulty with the phrase road to ruin (a road for ruining, as in an apple to eat?) a n d expensive parking lots (parking will be expensive?). Even a sentence that supposedly contains within it enough experiential context to explain the word mildew is often incomprehensible to many students from arid regions: What could John expect? He had left his wet swimming trunks in the dark closet for over a week. Of course they had begun to mildew (Baudoin et al. 1977:5). Finally, by carefully listening to what our students say about the texts they are asked to read, we can become further sensitized to their hidden comprehension problems. As teachers, we should not respond to what the reader does (right/wrong) as much as to what the reader is trying to do. Given that the reader is trying to make sense of the text (construct meaning), a teacher who listens carefully and responds to a students efforts will become aware of both the background knowledge and the cultural problems that students themselves bring to the text. In any case, the most valuable information is in our students perceptions and not our own. This is the type of information that is gleaned through asking open-ended questions, probing for inferences from the text, and asking students to justify answers to more direct questions 568 TESOL QUARTERLY

(for example, Why do you think so?). In addition, having students provide oral or written summaries will help teachers to discern problem areas in comprehension. CONCLUSION Thus, in achieving our immediate goals in the EFL/ESL reading classroom, we must strive for an optimum balance between the background knowledge presupposed by the texts our students read and the background knowledge our students possess. As we have shown by means of the foregoing classroom activities and techniques, this balance may be achieved by manipulating either the text and/or the reader variable. Of course, our long-range goal as reading teachers is to develop independent readers outside the EFL/ESL classroom, readers whose purpose in learning to read in English as a foreign or second language is to learn from the texts they read.4 But there, too, as Anderson notes, without some schema into which it can be assimilated, an experience is incomprehensible, and therefore, little can be learned from it (Anderson 1977:429; emphasis added). What makes the classroom activities and other techniques we have described valid is their applicability to the real world beyond the EFL/ESL reading classroom. Every culture-specific interference problem dealt with in the classroom presents an opportunity to build new culture-specific schemata that will be available to the EFL/ESL student outside the classroom. In addition, however, and possibly more importantly, the process of identifying and dealing with cultural interference in reading should make our EFL/ESL students more sensitive to such interference when they read on their own. By using the classroom activities and techniques we have described, our EFL/ESL readers should become more aware that reading is a highly interactive process between themselves and their prior background knowledge, on the one hand, and the text itself, on the other.

ACKNOWLEDGMENTS
This article grew out of the authors portions of a workshop entitled Reading in ESL: Insights and Applications from Research, presented at the 16th Annual TESOL Convention in Honolulu, Hawaii, May 1, 1982.
4 We

use the phrase reading to learn from texts in the broadest sense, including reading for academic purposes, reading for survival purposes or for purposes of functioning in society at various levels, and even reading for recreation or entertainment.

SCHEMA THEORY AND READING

569

THE AUTHORS Patricia Carrell is Professor of Linguistics and EFL/ESL at Southern Illinois University at Carbondale. Her current research interests include the role of background knowledge and the interaction of text and reader variables in text comprehension. Her research in this area has appeared in the TESOL Quarterly, Language Learning, Language Learning and Communication, and On TESOL 82. Joan C. Eisterhold is a doctoral student at Northwestern University and a former instructor of ESL at the University of Tennessee at Martin. She holds an M.A. in TEFL from Southern Illinois University at Carbondale.

REFERENCES
Adams, Marilyn J., and Allan Collins. 1979. A schema-theoretic view of reading. In New directions in discourse processing, Roy O. Freedle (Ed.), 1-22. Norwood, New Jersey: Ablex Publishing Corporation. Alderson, J. Charles, and Alexander Urquhart. 1983. This test is unfair: Im not an economist. Paper presented at the 17th Annual TESOL Convention, Toronto, Canada, March, 1983. Anderson, Richard C. 1977. The notion of schemata and the educational enterprise: general discussion of the conference. In Schooling and the acquisition of knowledge, Richard C. Anderson, Rand J. Spiro, and William E. Montague (Eds.), 415-431. Hillsdale, New Jersey: Lawrence Erlbaum Associates. Anderson, Richard C., Ralph E. Reynolds, Diane L. Schallert, and Ernest T. Goetz. 1977. Frameworks for comprehending discourse. American Educational Research Journal 14(4): 367-381. Bartlett, Frederic C. 1932. Remembering: a study in experimental and social psychology. Cambridge: Cambridge University Press. Baudoin, E. Margaret, Ellen S. Bober, Mark A. Clarke, Barbara K. Dobson, and Sandra Silberstein. 1977. Readers choice: a reading skills textbook for students of English as a second language. A n n A r b o r : U n i v e r s i t y o f Michigan Press. Burtoff, Michele. 1983. Organizational patterns of expository prose: a comparative study of native Arabic, Japanese and English speakers. Paper presented at the 17th Annual TESOL Convention, Toronto, Canada, March, 1983. Carrell, Patricia L. 1981a. Culture-specific schemata in L2 comprehension. In Selected papers from the ninth Illinois TESOL/BE annual convention, the first midwest TESOL conference, Richard Orem and John Haskell (Eds.), 123-132. Chicago: Illinois TESOL/BE. Carrell, Patricia L. 1981b. The role of schemata in L2 comprehension. Paper presented at the 15th Annual TESOL Convention, Detroit, Michigan, March, 1981. Carrell, Patricia L. 1983a. Background knowledge in second language comprehension. Language Learning and Communication 2(1):25-34.

570

TESOL QUARTERLY

Carrell, Patricia L. 1983b. Some issues in studying the role of schemata, or background knowledge, in second language comprehension. Paper presented at the 17th Annual TESOL Convention, Toronto, Canada, March, 1983. Clarke, Mark A., and Sandra Silberstein. 1977. Toward a realization of psycholinguistic principles in the ESL reading class. Language Learning 27(1):135-154. Coady, James. 1979. A psycholinguistic model of the ESL reader. In Reading in a second language, Ronald Mackay, Bruce Barkman, and R. R. Jordan (Eds.), 5-12. Rowley, Massachusetts: Newbury House Publishers. Collins, Allan M., and M. ROSS Quillian. 1972. How to make a language user. In Organization of memory, Endel Tulving and Wayne Donaldson (Eds.), 310-351. New York: Academic Press. Fillmore, Charles J. 1976. The need for a frame semantics within linguistics. In Statistical methods in linguistics, 5-29. Stockholm: Sprakforlaget Skriptor. Fries, Charles C. 1945. Teaching and learning English as a foreign language. Ann Arbor: University of Michigan Press. Fries, Charles C. 1963. Linguistics and reading. New York: Holt, Rinehart and Winston. Goodman, Kenneth S. 1967. Reading: a psycholinguistic guessing game. Journal of the Reading Specialist 6(1):126-135. Goodman, Kenneth S. 1971. Psycholinguistic universals in the reading process. In The psychology of second language learning, Paul Pimsleur and Terence Quinn (Eds.), 135-142. Cambridge: Cambridge University Press. Goodman, Kenneth S. 1973a. On the psycholinguistic method of teaching reading. In Psycholinguistics and reading, Frank Smith (Ed.), 177-182. New York: Holt, Rinehart and Winston. Goodman, Kenneth S. 1973b. Analysis of oral reading miscues: applied psycholinguistics. In Psycholinguistics and reading, Frank Smith (Ed.), 158176. New York: Holt, Rinehart and Winston. Goodman, Kenneth S. 1979. The know-more and the know-nothing movements in reading: a personal response. Language Arts 55(6):657-663. Hinds, John. 1982. Contrastive rhetoric: Japanese and English. Paper presented at the 16th Annual TESOL Convention, Honolulu, Hawaii, May, 1982. Hinds, John. 1983. Retention of information using a Japanese style of presentation. Paper presented at the 17th Annual TESOL Convention, Toronto, Canada, March, 1983. Hudson, Thom. 1982. The effects of induced schemata on the short circuit in L2 reading: non-decoding factors in L2 reading performance. Language Learning 32(1):1-31. Johnson, Patricia. 1981. Effects on reading comprehension of language complexity and cultural background of a text. TESOL Quarterly 15(2):169181. Johnson, Patricia. 1982. Effects on reading comprehension of building background knowledge. TESOL Quarterly 16(4):503-516. Kant, Immanuel. 1963. Critique of pure reason. (lst ed. 1781, 2nd ed. 1787, N. Kemp Smith, translator.) London: MacMillan Publishing Co.

SCHEMA THEORY AND READING

571

Kaplan, Robert B. 1966. Cultural thought patterns in inter-cultural education. Language Learning 16(1-2):1-20. Krashen, Stephen D. 1981. The case for narrow reading. TESOL Newsletter 15(6):23. Mandler, Jean M. 1978. A code in the node: the use of a story schema in retrieval. Discourse Processes 1(1 ):14-35. Marquardt, William F. 1967. Literature and cross-culture communication in the course in English for international students. The Florida Foreign Language Reporter 5(1):9-10. Marquardt, William F. 1969. Creating empathy through literature between members of the mainstream culture and disadvantaged learners of the minority cultures. The Florida Foreign Language Reporter 7 ( 1 ) : 1 3 3 - 1 4 1 , 157. Meyer, Bonnie J. F. 1975. The organization of prose and its effects on memory. Amsterdam: North-Holland Publishing Co. Meyer, Bonnie J. F. 1977. The structure of prose: effects on learning and memory and implications for educational practice. In Schooling and the acquisition of knowledge, Richard C. Anderson, Rand J. Spiro, and William E. Montague (Eds.), 179-208. Hillsdale, New Jersey: Lawrence Erlbaum Associates. Meyer, Bonnie J. F. 1981. Basic research on prose comprehension: a critical review. In Comprehension and the competent reader: inter-specialty perspectives, Dennis F. Fisher and Charles W. Peters (Eds.), 8-35. New York: Praeger. Meyer, Bonnie J. F., and Roy O. Freedle. In press. The effects of different discourse types on recall. American Educational Research Journal. Meyer, Bonnie J. F., and G. Elizabeth Rice. 1982. The interaction of reader strategies and the organization of text. Text 2(1-3):155-192. Minsky, Marvin. 1975. A framework for representing knowledge. In T h e psychology of computer vision, Patrick Henry Winston (Ed.), 211-277. New York: McGraw-Hill Publishing Co. Ostler, Shirley E., and Robert B. Kaplan. 1982. Contrastive rhetoric revisited. Paper presented at the 16th Annual TESOL Convention, Honolulu, Hawaii, May, 1982. Paulston, Christina Bratt, and Mary N. Bruder. 1976. Teaching English as a second language: techniques and procedures. Cambridge, Massachusetts: Winthrop. Rigg, Pat. 1981. Beginning to read in English the LEA way. In Reading English as a second language: moving from theory, C. W . T w y f o r d , William Diehl, and Karen Feathers (Eds.), 81-90. Monographs in Language and Reading Studies 4. Bloomington, Indiana: Indiana University. Rivers, Wilga M. 1968. Teaching foreign language skills. Chicago: University of Chicago Press. Rivers, Wilga M., and Mary S. Temperley. 1978. A practical guide to the teaching of English as a second or foreign language. N e w Y o r k : O x f o r d University Press. Robinett, Betty Wallace. 1978. Teaching English to speakers of other languages. New York: McGraw-Hill.

572

TESOL QUARTERLY

Rumelhart, David E. 1977. Understanding and summarizing brief stories. In Basic processes in reading: perception and comprehension, David LaBerge and S. Jay Samuels (Eds.), 265-303. Hillsdale, New Jersey: Lawrence Erlbaum Associates. Rumelhart, David E. 1980. Schemata: the building blocks of cognition. In Theoretical issues in reading comprehension, Rand J. Spiro, Bertram C. Bruce, and William E. Brewer (Eds.), 33-58. Hillsdale, New Jersey: Lawrence Erlbaum Associates. Rumelhart, David E., and Andrew Ortony. 1977. The representation of knowledge in memory. In Schooling and the acquisition of knowledge, Richard C. Anderson, Rand J. Spiro, and William E. Montague (Eds.), 99-135. Hillsdale, New Jersey: Lawrence Erlbaum Associates. Schank, Roger C., and Robert P. Abelson. 1977. Scripts, plans, goals and understanding. Hillsdale, New Jersey: Lawrence Erlbaum Associates. Smith, Frank. 1971. Understanding reading: a psycholinguistic analysis of reading and learning to read. New York: Holt, Rinehart and Winston. Steffensen, Margaret S., Chitra Joag-dev, and Richard C. Anderson. 1979. A cross-cultural perspective on reading comprehension. Reading Research Quarterly 15(1):10-29. Tannen, Deborah. 1978. The effect of expectations on conversation. Discourse Processes 1(2):203-209. Tannen, Deborah. 1979. Whats in a frame? Surface evidence for underlying expectations. In New directions in discourse processing, Roy O. Freedle (Ed.), 137-181. Norwood, New Jersey: Ablex Publishing Corporation. Warren, William H., David W. Nicholas, and Tom Trabasso. 1979. Event chains and inferences in understanding narratives. In New directions i n discourse processing, Roy O. Freedle (Ed,), 23-52. Norwood, New Jersey: Ablex Publishing Corporation.

SCHEMA THEORY AND READING

573

TESOL QUARTERLY, Vol. 17, No. 4, December 1983

Student-Teacher Working Journals in ESL Freshman Composition

RUTH SPACK
Tufts University and Boston University

CATHERINE SADOW
Northeastern University and Boston University

Recently, first language composition researchers have shown that by using a traditional composition teaching method which focuses on the form and correctness of a finished product, teachers ask students to produce writing which does not reflect the actual writing process. Findings indicate that most school-sponsored writing does not involve the self-motivation, contemplation, exploration, and commitment which characterize real life writing. These researchers recommend that in addition to being taught expository writing, students should have more opportunity for expressive writing in their writing courses to allow them to become better academic writers. According to some second language composition researchers and teachers, these recommendations are applicable to ESL college students because their composing strategies are similar to those of native English-speaking college students. Ungraded, uncorrected journals can provide a non-threatening way for students to express themselves in written English. However, the student-teacher working journals which we describe in this article are unlike student personal journals in two important ways: first, the topic of working journals is not personal, but is rather an outgrowth of the writing class; and second, the teacher regularly writes journals to the class on the same subject and includes, in those journals, selected student journal entries. The advantages of this approach are that a group awareness develops around issues relevant to ESL composition, that students come to see writing as a way to generate ideas and to share them, and that teachers become participants in the writing process. Since 1980 our students have been regularly writing ungraded, uncorrected working journals for our ESL Freshman Composition courses, and we have been regularly writing working journals too. Journal writing is an integral part of our classes, which are special sections of composition for international students and which cover the 575

two semesters of the college writing requirement. During both semesters students are required to write several expository essays and to become familiar with research techniques; in the second semester students also write a long research paper. These courses have two major goals: to prepare students for the level of academic written work established at the university in which they are enrolled, and to make students aware that writing is a process of using language to discover meaning and to communicate it (Murray 1978). The students we teach are primarily college freshmen, although upperclassmen and graduate students occasionally take the courses. Students represent every academic field at the university and come from all over the world. As is typical of foreign students studying in the United States, their English proficiency varies; nevertheless, all of these students carry a full academic load, can communicate orally with some ease, have adequate reading and listening skills, have studied formal grammar, and have some experience writing in English. Their TOEFL scores are at least 550. Some have just arrived in the United States after having studied English only as a class in high school, while others have studied in high schools in the United States or in Englishspeaking high schools overseas; all need strong writing skills in order to compete academically with their American counterparts. Until 1980, we followed the traditional composition teaching model presented in most textbooks: we asked our students to fit a topic into a rhetorical form, to imitate a model essay, to outline their main points before writing, and to pay careful attention to the correctness of their grammar, punctuation, and sentence structure. We then corrected, commented on, and graded the finished products. But our students often stiff prose, repetition of mistakes, inadequate development of ideas, lack of involvement in the topics, and lack of confidence in their ability to write well in English convinced us that this method was not meeting their needs or the goals of the course. Because research on ESL composition was almost nonexistent at that time (Zamel 1976, 1982), we looked for clarification and guidance to those writers, teachers, and researchers in native English-speaking contexts who were challenging the traditional method of teaching composition. The focus of research and classroom practice, it turned out, had recently shifted from the written product to the writing process. We discovered that requiring our students to formulate their thoughts beforehand and to fit those ideas into a rhetorical framework according to someone elses model was an unrealistic assignment. Murray (1968), for example, maintains that since writers do not know exactly what they are going to say before they start, they find it necessary to see their words on paper in order to explore their thoughts and to learn what their main points are. The act of writing itself can 576 TESOL QUARTERLY

change preconceived notions. Emig (1971), in a study of the composing processes of native English-speaking twelfth graders, discovered that students lack commitment to teacher-initiated topics when schoolsponsored writing, with its models and rules, does not reflect real life writing, which involves a self-motivated, explorative, contemplative process of composing. Perls study of the composing processes of unskilled native English-speaking college writers confirms that composing does not take place in linear fashion; rather, writing is a backand-forth movement in which writers write, reflect upon, change, and develop their ideas (1979). Students who are required to write in an outline-imitate-write order are therefore likely to produce artificially structured writing. Judy (1980) points out that by studying forms first, students come to view form as a pre-existent organizational mold into which content is poured; instead, he suggests, they should be taught that the form an essay will take is determined by the writers purpose and message. In fact, it has been found that idea development can be seriously inhibited by concern for correct form (Perl 1979) and likewise by fear of error (Shaughnessy 1977). Similarly, Britton, Burgess, Martin, McLeod, and Rosen (1975) have found that an approach which emphasizes the final product leads student writers to expect an evaluation of their writing rather than a response to their writing and further leads them to view writing as a test of their ability to master prescribed technical skills rather than as a body of work to be read for itself. All of these findings suggest that students should be made aware that writing is an act of discovery. Writing can be viewed, too, as problem solving or as a thinking problem whose goal is to communicate ideas to someone else by adapting the intentions of the writer to the needs of a reader (Flower and Hayes 1977:449). Emigs practical suggestion is to allow students more time and opportunity to practice reflexive writing which allows them to immerse themselves in the topic as they write and reformulate and rewrite in the process of discovering, exploring, and developing what it is they have to say. Britton et al. (1975) recommend that teachers assign more expressive writing, which entails informal, loosely structured language. According to Britton et al., the importance of expressive writing lies in its educational value as a matrix from which, in favorable circumstances, transactional writingwriting to inform, instruct, or persuadeis developed. Reflexive and expressive writing, then, can lead students toward better academic writing. Some ESL writing teachers and researchers have also begun to focus on the composing process. Zamel (1976, 1982) has observed that the findings and recommendations of first language composition researchers are applicable to ESL students because the composing strategies of STUDENT-TEACHER WORKING JOURNALS 577

such students are similar to those of native speakers. Even though many ESL students have less language proficiency than their native English-speaking counterparts, they nevertheless need to practice the non-linguistic intellectual and cognitive skills that constitute writing (Taylor 1976, 1981). Language does not have to stand in the way of ESL composition, and an ESL student need not have to be able to generate a perfect, eloquently flowing piece of writing the first time around to be a successful academic writer (Taylor 1981:9). Teachers do not have to initially look for and comment on error in an ESL students piece of writing; they should, at all levels, look at a composition as a message conveying the writers ideas (Raimes 1979). ESL teachers cannot overlook the conceptual function of language (expression of self) while they stress its purposeful function (interaction with others) (Raimes 1979). These are the issues which led us to restructure our writing classes. Our students are still taught throughout the course how to organize ideas, but they are no longer asked to fit a topic into a form, imitate a model essay, outline before they write, or attend to error in the early stages of their writing. Instead, they are given a choice of topics whose premises have grown out of ideas discussed in class based on a series of readings with a common theme. These writing assignments simulate authentic writing situations, including those the students face in the academic setting and will face in technical and professional fields. Students are then taught to use invention techniques, such as those described by Flower and Hayes (1977) and Daubney-Davis (1982a, 1982b, 1982c), in and out of class, in order to discover what they know about the topic and to generate more ideas. As they develop their topics, students begin to think about the appropriate audience for the subject and to tailor their first draft for that specific audience, whether it be the teacher, fellow students, or some hypothetical authority. Trial drafts are shared orally with the class; students discuss ideas and struggle for clarity. The drafts are then reworked, sometimes reread in class or in conference, during which time students are guided toward accuracy of expression; and then a paper is handed in. We read the paper for ideas and make suggestions for further development and changes in content, organization, and vocabulary. Students make changes and turn in a revised paper which is read for error and returned. Students then correct the paper in class, under supervision. Because they are writing for (simulated) authentic purposes, students understand that formal accuracy is important. They are shown how to say precisely what they mean. Editing of errors takes place in the final stage of writing after ideas are fully developed. At that point, student and teacher together solve the individual technical problems which arise out of the writing. Finally, 578 TESOL QUARTERLY

a fully revised and corrected copy is handed in, with all written brainstorming and drafts, for a grade. It is in the context of this course that working journals are introduced. Until very recently, only isolated teachers of native English speakers such as Macrorie (1968) and Elbow (1973), who recognized the need for more expressive writing in composition courses, recommended that students write journals. Today, student journal writing is accepted as an integral part of the curriculum in many regular freshman composition programs. There is general agreement that one of the main reasons student journals succeed is because they establish a dialogue between students, who write about what interests them, and teachers, who respond non-judgmentally. What is created is a real life writing situation, with a purpose, an audience, and a message, which allows students to feel comfortable when they write (Wood 1978, Staton 1980, Staton, Shuy, and Kreeft 1982, Kelly 1981). The act of writing the journal itself generates ideas (Fulwiler 1980, Cloud 1981). ESL teachers, too, have begun to introduce journal writing into their courses (Root 1979, Taylor 1981, Staton 1981, Zamel 1982). The focus of most of the journals proposed by these teachers is on personal issues raised by the student. There are difficulties with this approach, however. For example, personal writing encourages unskilled writers to write egocentrically without anticipating the readers needs (Flower and Hayes 1977, Perl 1979). It also causes some ESL students to feel dissatisfied precisely because they do not feel that the topic of their school writing should be personal (Zamel 1982). The teacher, too, may find that some students journals cover extremely personal subjects which are difficult to respond to. And even though personal journals can help students in a number of ways and can strengthen some student-teacher relationships, which are particularly important for lonely international students, they do little for the overall development of the course. In a writing class whose goal is to lead students toward writing reader-based prose (Flower and Hayes 1977:459), a less egocentric, more academicyet still self-expressivejournal assignment is called for. PROCEDURE The journal plan which we present in this article is based on a program developed for native speakers (Cowan and Cowan 1980), which we have adapted for non-native speakers. Our approach is a departure from the students personal diary. Rather, we prefer to set up a teacher-student interaction in which the teacher and the students regularly write and exchange journals which have a common general focus of ESL writing class issues and individual reactions relevant to STUDENT-TEACHER WORKING JOURNALS 579

them. The sharing of selected journal entries with the group by means of the teachers journal extends and enriches the interaction and leads to a more dynamic group awareness and exchange of ideas. One advantage of this format is that students need not feel compelled to write about themselves if that makes them uncomfortable. Since the topic of the journals is the English course, students may write about class discussions, readings, essays, research, journals, exercises, other students ideas, or the process of writing itself. This provides a focus for the journals, a broad enough subject which gives students plenty of leeway, yet which limits the number of extremely personal issues that border on diary entries. Students are guided away from egocentric writing toward writing which meets the needs of the readers. Another advantage of the student-teacher working journal plan is that the teachers journal gives students another perspective on the writing process and another view of informal written English. They can view writing teachers not only as critics but also as participants in the writing process (Murray 1968). That is, teacher journals give writing teachers credibility. Just as important, they also give teachers an opportunity to write informally to and for their students. Emig reports that teachers who do not write under-conceptualize and oversimplify the process of writing (1971:98) and notes that all teachers of writing need frequent opportunities for writing. One of the best ways to learn about the writing process is to write (Kelly 1981, Applebee 1981). How do we begin? A typical first journal from teacher to students on the first day of class explains the procedure and might look like this: Welcome to Freshman Composition for International Students. This term each of us will write a journal entry for each day of class, and this is my first. It is difficult for me to sit herein front of my typewriter composing a journal to a group of students I havent yet met. When you write your first journal, you will have the advantage of having met me and the rest of the class. It may be easier for you to write, but I doubt it. The first few journals are a challenge. After a while, you will probably find that they are easier to write. The subjects of your journals should be something that grows out of the class, for example, from the discussions, the readings, or the writing assignments. You may use the journals to explore topics for papers, to examine your own writing progress, or to discuss the process of writing itself. Hand in a journal of any length at the beginning of every class on looseleaf paper, and I will respond to it by commenting on the bottom and then I will hand it back to you at the beginning of the next class. I read for ideas only: I never correct or grade journals. My journals, which I will mimeograph and hand out at the beginning of each class, will be a response to the class, an outgrowth of discussion, 580 TESOL QUARTERLY

reading, writing. When I think that something you wrote will interest your classmates, Ill publish it within my journal. At the end of the semester, I am sure you will find the journal project a valuable one, for it will open up a dialogue that is not possible in the classroom, and it will give you the opportunity to use written English informally to explore, develop, and communicate your thoughts. I look forward to reading your first journals. Our journals, the length of which varies from half a page to two pages, consist of encouraging statements about writing, reactions to what has transpired in class, reinforcement of concepts covered in class, quotations from famous writers, and most important, excerpts from our students journals. Students like to know what other students write in their journals; not only does this help them overcome the feeling that they are writing privately, in isolation, but it also gives them ideas for their own journals (Fulwiler 1980). Understandably, ESL students are astonished to see their own words printed in a teachers journal. But all are pleased to be published once they get over the initial shock: Today in English class I saw my own diary was on the hand out. I was really surprised, because I did not expect teacher would put junk like my diary on an hand out. I settled down a little bit after I read the diary over. It was not as bad as I though it was. (unedited) Because our students journals are, of course, full of second language errors, we edit them before we publish them in our journals. We make corrections by using ellipsis when we omit words and brackets when we fill in missing words; we correct all spelling errors. The above student journal would be changed, when published, to: Today in the English class I saw my own diary . . . on the handout. I was really surprised because I did not expect that [the] teacher would put junk like my diary on a handout. I settled down a little bit after I read the diary over. It was not as bad as I thought it was. (edited) Even as we present corrected English, we make our ESL students aware that fellow classmates make mistakes, a normal and understood result of second language writing (Taylor 1981). But our major concern is to make it possible for students to read one anothers ideas without the mistakes, for, as Shaughnessy points out, errors demand energy without giving any return in meaning (1977:12). Although we strongly state at the outset that we never correct journals, we inevitably get one or two student journals at the beginning of each term which question the lack of correction: The best way to learn is by making mistakes and we obviously make a lot of them. Spelling, sentence structure, even grammar mistakes. It would be STUDENT-TEACHER WORKING JOURNALS 581

very helpful if you could show us our mistakes in the journal. Maybe by underlining a misspelled word so that we could look it up in the dictionary. On the other hand it shouldnt be too much work for you. Just a few corrections would already be a great help for us. In addition to responding individually to the student, we include such a comment in a subsequent teacher journal and reply to the whole class by saying that there is a lot more to writing than learning to correct errors. When unskilled writers hunt for errors in the early stages of writing, a breakdown of the rhythms generated by thinking and writing occurs (Perl 1979:333). We want the students to be able to write in their new language without fear of red marks (Raimes 1979). They should be able to experiment with words and phrases without the pressure of grades (Macrorie 1970). We tell them that sometimes students make the mistake of trying to write grammar when they should write ideas (Shaughnessy 1977). By the same token, sometimes teachers make the mistake of correcting papers when they should be reading them (Raimes 1979). Writing teachers should value writing which is not evaluated. Through working journals, students can learn to put the emphasis on correctness in its proper place. One student described this realization in this way: Journals have definitely helped us write. Basically I feel that they have helped me to quickly get my thoughts down in writing. It may not [be] grammatically revolutionary, but that can come later, once I have my thoughts written down. We find that the extensive time set aside for the detailed correction and grading of other writing assignments adequately attends to the need for accuracy in our students use of language. Because we make simple comments on each entry we can spend as little as twenty minutes reading through a group of fifteen journals. We read them quickly, looking for meaning only, and make short, responsive, non-judgmental comments. When spelling or sentence structure is grossly incorrect, we can use misspelled words or convoluted syntax correctly in our responses (Staton 1980). However, aware that the native English-speaking students resistance to school-sponsored writing is combined in the ESL student with a resistance to the language itself, which presents an even more formidable challenge than that of the students linguistic problems (Raimes 1979:3), we provide only positive feedback in our responses. Our goal is to teach students to overcome a fear of being misunderstood in a second language and to feel confident in their ability to communicate. They can use t h e language rather than be concerned with its usage (Widdowson 1978). Because the fear of error and criticism is removed from journal assignments, students can become comfortable expressing their ideas 582 TESOL QUARTERLY

in English even if the grammar is faulty. We encourage them to experiment with new vocabulary, idiomatic English, and recently learned structures in their journals (Root 1979). The journals give them invaluable experience in getting an idea out of their heads, onto a piece of paper, and into someone elses head. They see, through our comments, that even in English they can invoke a response with their written words. When ESL students write journals regularly and often, many say that it becomes easier for them to write journals (Root 1979). As one student wrote: This is my sixth journal to be passed up. I never knew that I can write six journals in less than one month. (I just cant forget when I got the journal assignment . . . I thought Mrs. Spack must be mad asking us to write journals twice a week. ) Now I am convinced that by writing journals you have more confidence . . . no more hesitate, no such thing as I have run out of ideas or any other excuses when I cannot write an article. I never realize the benefit of writing journals, and I always feel I am forced to write in my first few journals. But now (I guess the magic has work!) I am always ready to write about anything willingly. One of the benefits for teachers is the continuous feedback provided in student working journals, which helps us to adjust our teaching approaches throughout the semester. It is, of course, gratifying to learn what students are gleaning from our classes and when they are enjoying them: it is useful when they ask clarifying questions if there is something they have not understood (Staton 1980) or when they make suggestions about the course. Even more beneficial is learning about specific language problems our students encounter when they work with a new language. A student who was reluctant to speak in class shared in her journal her thoughts about the difference between speaking and writing: IS writing more difficult than speaking? Im skeptical about it. Because, the advantages of speaking, which are pointed in our textbook, are, for me, not meritorioussome of them . . . vice versa. Examples are as follows: x audience presentso I might be shy x immediate feedbackIm scared x The listeners heard the words spoken, and the tone, volume, pitch, speed, recognize the words and meaningscould it be possible even if the speakers language is very poor? x subjects will fit the person and the situation (so will the words)Im not sure On the contrary, writing benefits me to communicate with foreigners. (1) In writing, I can spend enough time to think about how I should explain my ideas in English. STUDENT-TEACHER WORKING JOURNALS 583

(2) I can use any dictionary and erase or correct errors of my sentences. (3) As there is no audience face to face waiting my words, I feel at ease. For nobody cares about my punctuation, intonation, and speed, etc. There are many ways to establish good relationships with ESL students, but establishing a writing relationship seems especially appropriate in a composition class. There are students who are culturally quieter or students who are uneasy speaking out in English. These students who are afraid to speak in class discover that they can use the journals to express and share their ideas in writing in a nonthreatening format. The journals, in a sense, liberate them. Our students use working journals to explore their conscious and unconscious thoughts about other class-related writing. When used in that fashion, the journal entries themselves are a type of invention technique. Many students approach their topic in a personalized fashion to get themselves started on a writing assignment. For example, after being given a choice of topics which grew out of a unit on the Thirties in America, one student described in her journal the discovery of the specific subject for her paper: The other day I went to see the movie Manhattan. It was excellent and I was very impressed by Woody Allen. The theme song of the movie was Rhapsody in Blue. It gave a nice atmosphere throughout the whole movie. I was in a special mood when this music came through the screen, and it caught my mind. May be that because it reminded me that I once played the music on the piano during the period that I was really sick of my lessons in classical music. I wanted to try something different. My teacher suggested me to try something from the field of jazz. I chose music sheets by George Gershwin, and I loved his music. As George Gershwins music attracted me, my interest in his life and himself increased. That is why I picked the topic about Porgy and Bess for my paper. Over the course of several days of class discussions on readings concerned with fairy tales, students were encouraged to respond to the readings in their journals. The following student journal, which is shown as it appeared in a teacher journal, demonstrates how one student used the working journal to discover more about the topic and to develop a focus for the paper assigned: Id like to share Yosefs journal with you because in it he focuses on the readings which we will discuss in todays class. Yosef may have found his subject and thesis while writing this journal. Journals can be very useful for findings things to write about. Unfortunately for children, fairy tales are now disappearing to make room for new modern heroes coming out of space and science fiction. I remember, for instance, when I was a child in Lebanon, my grandmother used to tell me the same fairy tale. And at the end of every fairy tale I 584 TESOL QUARTERLY

used to beg my grandmother for one more time, until I fell asleep and at last my grandmother could take a breath. Unfortunately, nowadays, the irreplaceable way of telling fairy tales has almost disappeared especially in the United States. Television is taking the place of the grandmother and instead of giving to the child the softness of the fairy tales, it gives tough and shining new extra-terrestrial or science fiction heroes. It is, I think, not only an injury to the customs and traditions of our grandparents, but also a big loss in our childrens culture. A teachers individual response to such journals serves to give the students encouragement and feedback on their topics, possibly by discussing the content and guiding the students to explore the subject matter further. Although at this stage in the writing the topic is closely linked to the students experience, the final paper is not necessarily written in such a personal mode. The paper which grows out of the early stages of writing is often written objectively, and teachers can, through journal responses, direct the student to approach the material in an objective fashion, if that is the purpose of the assignment. Conversely, the paper may turn out to be a personal piece of writing, and what is written in the journal may even appear verbatim as a paragraph in the essay. The point is that this exploratory stage, which gets students involved in their subjects and shows them that they can learn through writing, is highly valued. Some students may write about other class-related assignments only once; others continue to discuss their work in each journal until the assignment is due. The latter practice is more common during the second semester, when students spend a great deal of time creating a long research paper. Often, each stage of research is discussed in each journal and there is a give-and-take between student and teacher as problems arise and solutions are suggested. In the following exchange between student and teacher, the student begins her journal with a response to the teachers comment on a previous journal and then goes on to raise a question about the development of a thesis sentence for her research paper. The teachers response is included. Student journal: I finally did it. Last night I went to the library, and spent the whole evening reading and taking notes from those books about Modern Times. Managed to get what I want except three books were missing. I prefer to do it this way. So that I know what Im doing. Rather than bringing the book everywhere I go and read it only when Im free. Then I wont know or remember whats what. This is what I mean when I told you I was going to the library to do my research. There goes the answer to the mystery. I have to apologize for written such a terrible thesis statement. I really take it as a temporary one. When I wrote it, I never intend to use it. Its just a sentence to tell you where Im heading. In fact, I find it difficult to have a STUDENT-TEACHER WORKING JOURNALS 585

thesis statement. What Im aiming at is to use this silent movie as a mirror to reflect as much about the 30s as I could. IS this a valid research? Or must a research paper be like an argumentative paper where the writers view is clearly stated in the thesis statement? Teacher response: It doesnt have to be argumentative, Your thesis could be The silent movie is a mirror which reflects the Thirties. You may find that this is still too broad a topic; more research and writing will help you narrow your focus. However, youre on the right track. Not only can students use their working journals to search for and develop topics and theses for papers, they can also discuss in them the final stages of putting a paper together: I am experiencing a very nice mood this week since it became clear that my research paper has turned out decently. Before this past weekend I spent a lot of time on reading and thinking. Most of the reading was no dice and most of the thinking was dismal. I believe I did not get much out of my reading because I turned up too few sources. I had to read almost every sentence on a page where I would merely skim in Russian. The research was so upsetting that my life lost its usual balance and rhythm for a month. I did not pay the bills in time, did not reply to letters, hated a regular How are you? question addressed to me. No question about normal sleep or enjoying a movie. When I put everything down, it turned out to be enough and it wasnt so bad. And I almost had two more weeks to polish it. When the first draft was almost done, I hoped to make it more consistent by adding more to the content. And then I saw the daylight. My confidence came back and I changed my mind about adding more material (I also felt a space constraint). I wrote a conclusion and stressed a few points there were in the paper. Now I like it in general. It makes sense. I get back to it every night, but for the last few days I have not changed anything. Statistics is not a very pleasing topic for an English composition class and it may take some effort to read it. I definitely failed to make the topic enchanting but I hope I described it satisfactorily and showed my point of view. I dont have it typed yet, but life is blossoming again. When we publish such student journals in our journals, we make the whole class aware that writing is a process of discovery; all writers go through several stages of composing, often in a back-and-forth movement, as they attempt to clarify meaning (Murray 1978). As teachers, we become aware that each of our students papers has its own history: we learn about the difficulties and successes they have encountered from inception to completion. This awareness has influenced the comments we write on the papers we do correct and grade; we do not evaluate only the final product but also take into consideration the knowledge that writing is an enormously complex activity, especially for second language learners. 586 TESOL QUARTERLY

Although some students have trouble keeping journals regularly, we find that the majority hand them in on time. We believe that the teacher journal adds a dimension to the English class which inspires students to take the journal assignment seriously, as it is part of a pact made between the teacher and the students; we all have the same assignment. Most students feel that because the teacher writes a journal, it is an obligation on their part to do the same. One student wrote: Today I would like to write about my teachers journal that I get whenever I give her mine. First of all, I think it is a good idea that she writes a journal like her students, too. So I dont consider it just as homework but something that she likes to read it as well as I like to read hers. I personally think her journal is like a message to me that makes me to give it an answer. It makes me to write because I feel responsible to let her know what I got and am getting from her or her class. Writing our own journals takes very little time. Teacher journals can be ready to hand out to a class in as few as ten minutes, if we handwrite a short message directly onto a ditto master and then run it off on the mimeograph machine; in as many as twenty minutes if we write or type a longer message and/or include one or several quotations from student journals. Of course what we write on any given day depends on inspiration and time. We find the more experience we have writing teacher working journals, the easier it is to write them. When we first wrote, we were not always sure what to say to our students. Even at the beginning of each new course finding something to write can be difficult because we do not know each student and the class does not initially have a personality of its own. We sometimes find ourselves writing stiffly in clichs, and using expressions such as in this course you will discover inner talents, and of course feel dissatisfied with the results. At this early point in the term we sometimes regret having made the commitment to write. But every term, without fail, something happens, something almost magical. Our writing loosens up. We begin to look forward to writing our journals. We sometimes have to stop ourselves to keep within a self-imposed one-page limit. We often go over the limit because we are brimming with quotations from student journals we want to share and with ideas we want to express. We also use our journals to delineate and formulate our own ideas. In the journals we communicate information to the students and discuss our own ideas about the readings and/or what has transpired in class. The following teacher journal, for example, was written in response to several class discussions. Drawing on a personal experience, the teacher was able to share a thought related to the classroom experience: STUDENT-TEACHER WORKING JOURNALS 587

Do you ever wonder what makes some class discussions stimulating and others less than that? I usually consider that subject from the point of view of a teacher. Last week I reconsidered it as a student. I am participating in a literature discussion class at the Cambridge Adult Education Center; we are reading classic short stories: OConnor, Singer, Joyce, Kafka, Chekhov, etc. I embarrassingly admit to you that I didnt do the reading for last weeks class. I thought of skipping the class, but decided against that choice since I usually enjoy the intellectual stimulation of the experience of discussing literature with my peers. But I didnt enjoy myself at all. For one thing, I didnt know what anyone was referring to, so I felt left out. For another, I couldnt contribute to the conversation, so I felt frustrated. However, what bothered me most was the apparent strain the teacher felt as she tried to evoke responses from the class members. One normally expressive voiceminewas silent and the teacher had to work harder than usual to make the discussion lively. Every time her questioning eyes met mine, I felt uncomfortable. After a while, I just kept my head down, trying to give the impression that I was deeply engrossed in the book. In fact, I did get a chance to speed read my way through the rather long short story, but it was too late to contribute any words of wisdom. I wanted to explain myself at the end of class, but that seemed absurd; the class is a voluntary one, there are no grades, no tests, no reprimands. So I simply went home and crawled into bed. I learned more about teaching from this incident than I did about learning. Im not going to tell you all that you MUST do the reading and you MUST participate in class. I just want you to realize that the relationship between a teacher and her class is a mercurial one. It depends not only on a teachers preparedness or on her ability to lead an invigorating discussion, but also on the preparedness of the students and their ability or willingness to participate. Since these factors often do not coincide, it is obvious that not all classes can possess the same quality. When they do, discussions are energizing and will be remembered for a long time. They make it all worthwhile. This teacher journal had an impact on the class as a whole. After students read it and had time to reflect upon it, they came to class better prepared and the discussions improved dramatically. However, although some of the teacher journals are attempts at the same kind of creative activity the students are undertaking, not all of them have a serious message. The next teacher journal was written while students were working on papers concerning college-related problems: One of our readings is about the problem of procrastinationthe putting off or postponing of things which need to get done, I know all about procrastination. Procrastination and I are familiar companions, especially when it comes to correcting papers. Some of you may feel that correcting papers is easier than writing them, but having been both a student and a teacher, Im not convinced that that is true. I have mastered the art of

588

TESOL QUARTERLY

postponing the correction process, and have become better at it than I ever was at postponing the writing process. For example: Last Monday, you gave me your completed papers. As always, I gave myself one week to correct them. Since I also teach on Tuesday, Wednesday, and Thursday, I could not, of course, correct them on those days or on the nights before those classes, since I spent my time reading and writing journals and preparing the lessons. Although I dont teach on Friday, I could not correct papers on Thursday night because I myself take a class then. And since Friday is my only day off, I needed the time to relax, shop, and cook. And that brought me to Friday night. I began reading the essays, and thought, oh these are wonderful; theyll be easy to correct. So I corrected two of them and then decided to finish them Saturday morning. Saturday morning I did four more and then started feeling sorry for myself. They werent so easy to correct. They took a lot of concentration and required that I, as always, write my own little essays at the end. So I took the rest of the day off. And, of course, I never work on Saturday night. Sunday morning, I did some more papers, all the while thinking, is this any way to spend a weekend? So I went with the family to buy pumpkins, visit a turkey farm, browse through some antique shops, and generally, to have fun. Sunday night. Sunday night! Sunday night? I thought I had left that Sunday-night-pit-in-the-stomach-sensation behind me when I graduated. How could I avoid that feeling? Procrastinate, of course. Watch Masterpiece Theater and read. Of course I had to getup early this morning to face the task. I did the rest of the papers (frantically, but with alacrity), then sat down to write this journal. I have just enough time to eat, drive to work, run this off on the machine, and be on time for class. And, oh dear, on Wednesday, I will have another meeting with Procrastination. Problems, problems . . . We have learned a lot since discovering the teacher working journal. We empathize with our students; we know how difficult it can sometimes be to produce writing by a certain date, and to write even when we are not inspired. We have learned about the writing process by writing. We have learned, as they have, to find something to say just by writing. We have learned to write honestly to a specific audience within a specific context, to write frequently and easily, to experience the delight of seeing our own words in print, and to experience the satisfaction that comes with writing to be read and acknowledged. Perhaps the significant feature of student-teacher working journals is that they have enabled us to sustain an exchange of ideas with the whole class; no one is writing in isolation. This frequent practice is done for fifteen other living, breathing, interesting and interested

STUDENT-TEACHER WORKING JOURNALS

589

human beings. The journals establish a special spirit in the class, a camaraderie. We learn from one another throughout the course. We are not writing to a blank piece of paper; we are writing to one another, trying to communicate within a context which is meaningful to all of us. By reading our journals and by writing theirs, students become more aware of the way most people really write, of some ways in which they can learn to write, and of writing as a way to learn. CONCLUSION We and our students perceive some changes in their writing that might be attributable to the journal project. However, we make no claims about improvement in ESL student writing; the major thrust of this article is a pedagogical one. Some research has been done on the journal writing of native speakers, the focus of which is on amount and frequency (Hull 1981) and on childrens communicative competence and language functions (Staton, Shuy and Kreeft 1982). Our experience with student-teacher working journals has raised questions about how journals affect ESL college students writing attitudes, behaviors, and ability. We perceive, for example, that journal writing has had a positive impact on ESL students writing attitudes and habits. As we discussed earlier in this article, Raimes (1979) finds that ESL students are resistant to classroom writing and to the English language itself. Does the writing of journals in fact help to break down the barriers which exist between the ESL student and school-sponsored writing and/or between the ESL student and the use of the written language? Are students reluctant to write journals at the beginning of the course, and do they find that journals are easier to write as the semester progresses because they write journals regularly and often and/or because the teacher also writes journals regularly and often? Are students afraid to write because they are afraid of making mistakes, and do they lose their fear of writing in the journals because the journals are not corrected? Are students hesitant about experimenting with their writing, and do they use the journals to take risks because the journals are not graded? If they find writing in journals easier and less intimidating, does their new-found confidence affect their attitude toward other class-related writing? Do students become more committed to writing as a way of communicating ideas to others as a result of journal writing, and does this commitment carry over to other writing assignments? Do students generate more words faster in the journals as the term progresses, and does this increase in amount and speed carry over to other class-related writing because of the journals?

590

TESOL QUARTERLY

How can such attitudes and behaviors be determined? One way is through before-and-after questionnaires. Some instruments for measuring attitude toward writing and writing anxiety already exist (see Daly and Miller 1975, Bloom 1980). A case study approach which includes interviews with students, observations of students in the process of composing, and an analysis of their written work (see Zamel 1983) might be the most effective way to examine writing behaviors. We also perceive that students better understand the purpose of writingto explore, develop, focus, organize, and finally to share ideas with othersbecause they write journals. This, of course, is difficult to measure. It has generally been the case that our students journals contain a focused and somewhat organized message which they want to communicate to us and to which they expect a response. To what extent does the expected teacher feedback, the awareness that the student journal might be published in the teacher journal, and the reading of the teacher journal itself contribute to the students sense of the need for organization and purpose? Another complete study could focus on the differences between writing samples of students who have taken a writing course which includes the student-teacher working journal component and students who have taken a similar course which does not include this journal component. Writing samples could be taken before, during, and after the courses. These writing samples could be studied to determine whether there is a difference in the ways in which the students incorporate principles of organization and the rhetorical conventions of writing such as audience, purpose, and thesis into their own writing. Do students who write working journals exhibit more skill in their approach to these principles and conventions than students who do not write working journals? Can it be said that those students who have a better grasp of these principles and conventions are more competent expository writers? Of course, each question raises other questions; our suggestions are not meant to be exhaustive or definitive. The working journals, it seems clear, would be a rich resource for investigation.

ACKNOWLEDGMENTS This article expands on a demonstration given at the 17th Annual TESOL Convention, in Toronto, March, 1983.

STUDENT-TEACHER WORKING JOURNALS

591

THE AUTHORS
Ruth Spack is Adjunct Lecturer/Special Instructor for Foreign Students in the English Department at Tufts University and Lecturer in English in the English Department at Boston University. Her professional background includes ESL consulting and ESL teaching in secondary schools and in adult education programs. She currently teaches Freshman Composition and Intermediate Composition. Catherine Sadow, a Coordinator at the English Language Center at Northeastern University and a Lecturer in English in the English Department at Boston University, teaches Intensive English and Freshman Composition for International Students. She has taught English and ESL at the elementary and secondary levels and has had extensive experience teaching immigrants in adult education programs.

REFERENCES
Applebee, Arthur N. 1981. Looking at writing. Educational Leadership 3 8 (6):458-462. Bloom, Lynn Z. 1980. Teaching anxious writers: implications and applications of research. Composition and Teaching 2:47-60. Britton, James, Tony Burgess, Nancy Martin, Alex McLeod, and Harold Rosen. 1975. The development of writing abilities (11-18). London: Macmillan. Cloud, Geraldine. 1981. The student journal: improving basic skills. Clearing House 54:248-50. Cowan, Elizabeth, and Gregory Cowan. 1980. Instructors manual to accompany Writing. New York: John Wiley and Sons. Daly, John A., and Michael D. Miller. 1975. The empirical development of an instrument to measure writing apprehension. Research in the Teaching of English 9:242-249. Daubney-Davis, Ann E. 1982a. Using invention heuristics to teach writing. Part I: An introduction to heuristics. TECFORS Newsletter 5 (2):1-3. Daubney-Davis, Ann E. 1982b. Using invention heuristics to teach writing. Part II: A look at methods. TECFORS Newsletter 5 (3):1-3. Daubney-Davis, Ann E. 1982c. Using invention heuristics to teach writing. Part III: More methods to try. TECFORS Newsletter 5(4):5-7 Elbow, Peter. 1973. Writing without teachers. London: Oxford University Press. Emig, Janet. 1971. The composing processes of twelfth graders. U r b a n a , Illinois: National Council of Teachers of English. Flower, Linda S., and John R. Hayes. 1977. Problem-solving strategies and the writing process. College English 39 (4):449-461. Fulwiler, Toby. 1980. Journals across the disciplines. English Journal 69:14-19. Hull, Glynda A. 1981. Effects of self-management strategies on journal writing by college freshmen. Research in the Teaching of English 1 5 (2),135-148.

592

TESOL QUARTERLY

Judy, Stephen. 1980. The experiential approach: inner worlds to outer worlds. In Eight approaches to teaching composition, Timothy R. Donovan and Ben W. McClelland (Eds.), 37-51. Urbana, Illinois: National Council of Teachers of English. Kelly, Lou. 1981. Learner-teacher dialogues and writing that is learning. English Journal 70:26-9. Macrorie, Ken. 1968. Writing to be read. Rochelle Park, New Jersey: Hayden Book Company, Inc. Macrorie, Ken. 1970. Uptaught. New York: Hayden Book Company, Inc. Murray, Donald M. 1968. A writer teaches writing. Boston: Houghton Mifflin Company. Murray, Donald M. 1978. Internal revision: a process of discovery. In Research on composing, Charles Cooper and Lee Odell (Eds.), 85-103. Urbana, Illinois: National Council of Teachers of English. Perl, Sondra. 1979. The composing processes of unskilled college writers. Research in the Teaching of English 13 (4):317-336. Raimes, Ann. 1979. Problems and teaching strategies in ESL composition. Language in Education: Theory and Practice 14. Arlington, Virginia: Center for Applied Linguistics. Root, Christine B. 1979. The use of personal journals in the teaching of English to speakers of other languages. TESL Reporter 12 (2):3-5. Shaughnessy, Mina P. 1977. Errors and expectations. New York: Oxford University Press. Staton, Jana. 1980. Writing and counseling: using a dialogue journal. Language Arts 57 (5):514-18. Staton, Jana. 1981. Literacy as an interactive process. The Linguistic Reporter 24 (2):1-5. Staton, Jana, Roger Shuy, and Joy Kreeft. 1982. Analysis of dialogue journal writing as a communicative event. Vol. I. Final Report to the National Institute of Education. Washington, D. C.: Center for Applied Linguistics. Taylor, Barry P. 1976. Teaching composition to low-level ESL students. TESOL Quarterly 10 (3):309-319. Taylor, Barry P. 1981. Content and written form: a two-way street. TESOL Quarterly 15 (1):5-13. Widdowson, H. G. 1978. Teaching language as communication. O x f o r d : Oxford University Press. Wood, Hugh B. 1978. Opening the classroom through the use of student journals. Improving College and University Teaching 26:25-8, 30. Zamel, Vivian. 1976. Teaching composition in the ESL classroom: what we can learn from research in the teaching of English. TESOL Quarterly 1 0 (1):67-76. Zamel, Vivian. 1982. Writing: the process of discovering meaning. TESOL Quarterly 16 (2):195-209. Zamel, Vivian. 1983. The composing processes of advanced ESL students: six case studies. TESOL Quarterly 17 (2):165-187.

STUDENT-TEACHER WORKING JOURNALS

593

TESOL QUARTERLY, Vol. 17, No. 4, December 1983

Increasing Learner Involvement in Course Management

ANDREW PETER LITTLEJOHN
University College of Bahrain

This article explores ways in which learners can be brought into a more central role in making decisions about the organization and direction of their language courses. As an introduction, a review of some of the arguments which support increasing learner involvement is presented. Constraints in introducing such involvement are then discussed, concentrating on learners prior experiences and expectations, and it is argued that what is required is a gradual approach toward relinquishing the teachers dominant role. This argument is taken up in a third section, where some materials and tasks are presented which are designed to bring about such a relinquishment.

In recent years, there has been considerable debate over the need for more learner-centered teaching approaches. For some, such approaches refer to the design of syllabuses and course materials that more accurately reflect the interests of learners and the situations in which those learners are likely to require use of the target language. In terms of this article, however, such approaches are not fully learner-centered since they still involve a person other than the learner determining what is to be learned and how. For that reason, the issue that is dealt with here is how we may encourage learners to take more control over the management of their own study both inside and outside the classroom. The role of the teacher would therefore be that of a learning adviser (i.e., someone who is experienced in the teaching/ learning process) or a knower (i.e., someone who can provide a ready source of the language and, if necessary, can correct errors). THE CASE FOR INCREASED LEARNER CONTROL Among the many arguments in favor of allowing learners greater control in the management of language courses, the following are offered as an overview.
595

The Risks of Teacher Dominance Teaching a language course involves making a large number of decisions concerning a wide variety of topics. These include determining which samples of the target language to present, how much guidance to offer, what the long term and short term teaching and learning goals should be, how evaluation should be handled, which methods and task types to implement, and what the general standards in both target language attainment and classroom behavior should be. According to Allwright (1978, 1981), the complexities involved in managing these areas and making these decisions necessarily entail a number of risks that can threaten to destroy the value of the classroom experience for the learner. For example, there is the danger that learners may feel spoon-fed if the language is broken down into too many minute parts; they may be demoralized by standards that are set too high or too low; they may be frustrated by an inappropriate pace or teaching direction; they may be made to feel dependent on the teacher for help or as a source of intelligible target language input; and they may be confused by inadequate, improvised explanations or by an inconsistent treatment of errors. These risks are so significant that if one person tries to cope with the very considerable complexities of managing everything that needs to happen in the classroom . . . [then] a serious weakening of the value of the classroom experience for the learners is virtually inevitable (Allwright 1978:105). A teacher who assumes direct and exclusive responsibility for course management is, according to Allwright, professionally irresponsible. Similarly, Stevick (1976), using terms from Bernes (1964) theory of transactional analysis, argues that classroom activities often involve a Parent-Child relationship between the teacher and the learners, where the latter have abdicated their rights and responsibilities as Adults in the face of the teacher, who is always right. In this situation, any learning that takes place is more likely to be defensive, as learners seek to protect themselves from the possibility of being exposed or embarrassed. But this learning has, for the most part, no depth; it is like a suit of armor and is a burden, to be worn as little as possible and cast off entirely (i. e., forgotten) at the first safe opportunity (Stevick 1976:110). Involving learners more in the management of their courses might thus conceivably lead to a reduction of risks involved in conducting exclusively teacher-directed classes and, at the same time, could contribute to the development of a classroom atmosphere more conducive to deeper or, as Stevick terms it, receptive learning.

596

TESOL QUARTERLY

The Nature of Language Learning Our understanding of the process of language learning is still far from complete. Some teachers and course designers have, however, tended to assume that there is a set route through the learning task, both in terms of linguistic content and teaching/learning methods. Yet, it seems reasonable to suggest that we should not expect every student to learn in the same way, at the same rate, or to have the same interests and abilities as everyone else. Rather, there may in fact be as many approaches to language learning as there are language learners. Seen in this light, the traditional teacher-led classroom can only be a partially successful arrangement. Since the content and organization of a lesson may not necessarily be appropriate for each individual learner, there is a possibility that such teacher-led classes may actually do more to hinder language learning than to facilitate it (see, for example, Krashen [1982:68-70] on how a grammatically sequenced course may obstruct language acquisition). For a more efficient and effective use of resources, therefore, learners need to be encouraged to share in managing the learning task. We need to move away from the teachers I think you need this and more toward the learners I know I need this. Such a shift implies some kind of learner training, the issue discussed in the second half of this article. Communication as a Goal and a Method One of the most significant changes in English language teaching recently has been the movement toward so-called communicative approaches. In almost all cases this simply refers to the teaching of items of language use (functions/notions) rather than of language form (grammar) and to an increased amount of oral work. Breen and Candlin (1979, in press), however, argue that if our goal is to develop communicative skills, then our method should itself be communicativethat is, it should involve the exchange and negotiation of ideas and feelings about the learning process, with the teacher as coparticipant, not dominant, in the group. The classroom, in this case, is no longer a pale representation of some outside communicative reality (1979:98) where learners are engaged in rehearsing for a performance at some later time and place. It offers instead an opportunity for realistically motivated communication as learners share their views about the learning process. Motivation Experiments conducted by Beach (1974) with tutorless groups of college students in psychology, and by Littlejohn (1982a; discussed INCREASING LEARNER INVOLVEMENT 597

further in 1982b) with tutorless groups studying beginning Spanish, have found that small-group independent study can lead to increased motivation to learn. The evidence presented by Beach (based on observer reports) and by Littlejohn (based on participants responses to a questionnaire) identify similar factors leading to this increased motivation. Participants felt free to speak, to make mistakes, and to contribute their own experiences, all of which gave them a feeling of being supported in their difficulties with learning. In contrast, the participants had often felt inhibited or intimidated in teacher-led classes, either by the presence of the teacher-expert or by the presence of other students with whom they were working in competition, rather than cooperation. An increase in motivation was also reported by Fitz-Gibbon and Reay (1982) after conducting a slightly different, although related, experiment in an inner-city school in England. Prior to the experiment, in which 14-year-olds were involved in tutoring 1l-year-olds in French, three quarters of all of the pupils reported hating the language and considered it a useless subject (1982:40). The pupils were asked to rank school subjects in terms of how much they liked them and, after the experiment, the researchers found a statistically significant positive shift in the ranks assigned to French (1982:42). Quality and Quantity of Subject Matter Learned According to Allwright, encouraging learners to become more involved in course management should bring about a direct improvement in their language learning since they would take much more responsibility for identifying and repairing their errors (1981: 11). There appear to be few, if any, experiments that specifically test this claim in relation to language learning. Evidence available from other areas, however, is supportive of the belief that increased learner involvement leads to increased subject matter mastery. Beach, in the experiment noted above, reported an increased ability to apply principles studied (1974: 198) and found that tutorless groups scored higher in a final achievement test than did tutor-led control groups. Similar results have been reported by Faw (described in Rogers 1969), Hovey (1973), and Webb and Grib (1967). Other Desirable Effects In addition to the gains described above, students who are more deeply involved in controlling their own learning characteristically develop in other ways as well. Beach found that tutorless groups showed increased interpersonal skills, sense of responsibility for ones 598 TESOL QUARTERLY

growth and learning, improvements in critical thinking and lasting curiosity aroused by the learning (1974:198). Similarly, Gruber and Weitman concluded that placing a major responsibility on the students for their own education could lead to developing attitudes which result in the students continuing search for knowledge after the formal experience is over (1962, reported in Beach 1974:192). It is interesting to note that some significant gains in this respect can be made in what would generally be termed educationally difficult circumstances. Williams (1930), in one of the earliest experiments with Self-Directed Learning, reported on a 6-month project with delinquent children in the United States. After administering pre- and post-tests, he concluded that a group of delinquent boys of varying ages and capacities will, if given the opportunity and supervision, improve more in educational age when left alone than they will under ordinary schoolroom conditions with formal instruction (1930:718). While it may be the case that formal instruction today is quite different from what it was in 1930, it is worth recalling that Fitz-Gibbon and Reays work in an inner-city school in England in 1982 (described above) also found gains among learners as a result of increasing learner involvement.

The case for involving learners more in the management of their courses thus seems quite compelling. It is, however, clear that we cannot simply abandon our learnersespecially when they have taken a deliberate step in enrolling for a language class. Probably the greatest constraint in applying notions of learner control is the learners themselves, an issue to which we now turn. CONSTRAINTS IN IMPLEMENTATION There is a widespread belief that in order to learn one has to be taught. Learners normally place considerable expectations on the teacher to organize their exposure to the language and show them how to study. Such expectations reveal the existence of dominant assumptions about the most effective modes of learning, so we should not be surprised if learners view any attempt to involve them more in course management as either very threatening or irresponsible. Part of such a reaction is, of course, quite understandable. Learners, not normally being called upon to think about the planning and implementation of a course, often have rather naive views about the nature of language and language learning. Our demanding that they take more control over course management might thus leave them INCREASING LEARNER INVOLVEMENT 599

feeling exposed and uncertain. In setting up the teacherless Spanish groups noted above, typical initial comments of the students were But we dont know how Spanish should be taught and How will we know if were making mistakes? (Littlejohn 1982a:5). Perhaps it is in situations such as these that the full force of Stevicks Parent-Child analysis can be seen. However, in desiring that our learners assume more control, we are not demanding that they possess a specific body of knowledge, but rather some definite personal qualities. These may include the ability to tolerate ambiguity, to take risks, to study alone, and to suspend doubts. Interestingly, it is just these qualities that are said to be found in good language learners (see, for example, Naiman, Frohlich, Stern, and Todesco 1977, Rubin 1975), which may partly explain why experiments with students who volunteer for small-group independent learning often prove so successful. Introducing such learner-centered approaches as a general course requirement would almost certainly bring different results. We need, therefore, to move very cautiously. Given the difficulties involved and the general, if understandable, naivet of learners in regard to issues of course management, we need to view any attempts at increasing learner involvement as a process involving the gradual and continual refinement of the learners ability to perceive and manage the learning task. We should not, therefore, be so much interested in what learners say about the content and form of their course as in the process by which they arrive at their opinions. In terms of practical implementation, this seems to suggest the devising of open-ended tasks that stimulate learners to think more deeply about how their language course is being conducted and gradually to take a more meaningful role in directing its scope and method. It is the design of such tasks that now concerns us. IDEAS FOR COURSE AND MATERIALS DESIGN The ideas that are presented here stem from experiments carried out at the University College of Bahrain with two groups of Arab students with lower intermediate ability in English who, having failed the preparatory year, were required to repeat a semester in General English, six hours per week for 14 weeks. These students (24 in all) were generally considered to be lacking in motivation and had had little or no experience with communicative approaches or methods involving group work or pair work. They saw language learning as the study of grammar and vocabulary and the roles of teacher and learner as clearly separated. They were in no way specially selected for the purpose of the experiment.

600

TESOL QUARTERLY

Since the desire was to involve the students more in decisions concerning course conduct, it would have been inappropriate to draw up a linguistic syllabus (e.g., a list of structures/functions to be covered) prior to the commencement of the course. Clearly, however, some direction or focus was needed, particularly in the light of the learners previous experience. The approach taken, therefore, was to devise a set of principles which classroom activities were to work toward fulfilling. From the points set out in support of more learner control, the following principles seemed to suggest themselves: 1. Language would be presented and experienced as communicative behavior, not as an abstract system of formal rules. 2. The course would not only attend to the end-of-course needs of the students but also to their wants as they arose during the course (i. e., learning would proceed in a direction and at a pace appropriate for the learners). 3. The course would take as its starting point the prior experience of the students as language learners and as communicators, and this experience would be mobilized and exploited in the classroom. 4. The course would seek to involve and interest the students and to maximize motivation. 5. The course would seek to develop self-sufficiency in the students (i.e., they would be encouraged to make their own decisions over what and how to learn and to identify and correct their errors). 6. The teacher would be co-participant and co-communicator in the group, seeking not to determine how to learn but to stimulate and advise. In terms of practical implementation, it was decided that the course would consist of three components: Language Learning Workshops, which would be concerned with developing the students awareness of the what and how of language learning; Formal Linguistic Input, concerned with the presentation and practice of linguistic knowledge, such as grammar, functions/notions, and vocabulary; and Activities, which would give the students greater experience in the range of possible classroom methods. Obviously, these components were not discrete categories, since any attempt to develop an awareness of the what and how of language learning would have to be carried out through some kind of activity. These components, rather, reflected the concern that the students progress toward exercising greater control should not be constrained by their lack of experience with possible ways of approaching language learning. The ideas that follow, therefore,

INCREASING LEARNER INVOLVEMENT

601

represent some of the tasks that were given to the students to widen their experience and gradually bring them into a more central role.1 MATERIALS AND TASKS The Questionnaire

This questionnaire was given to the students in the first session of the course. After briefly discussing the type of answers one could give, the teacher asked the students to complete it privately in English. Responses to each question were then compiled on the board and served as the basis for a general discussion. The purpose of the questionnaire was to demonstrate from the beginning that the students experiences and opinions were to be drawn upon, and to encourage them to start considering the relevant issues. Review of Previous Textbook

In this task the students were split into small groups, each being allotted sections of grammar work from their last textbook. They were then asked to examine each section in terms of what it required them to
1 It

should be noted that the purpose of these tasks was not to provide opportunities for practicing/communicating in the foreign language itself, although this, of course, may have been an added bonus.

602

TESOL QUARTERLY

do and how difficult or easy they found it. This activity encouraged a considerable amount of discussion but, since the class was composed entirely of Arab students, the discussion was mainly in Arabic. This was not felt to be a problem, however, since demanding that the task be done in English would have prevented effective communication and defeated the purpose of the activity. The results of the group discussions were collected and a list of areas of grammar was drawn up in descending order of difficulty. An additional purpose of the activity was to give the students experience in working cooperatively without ongoing teacher direction. Students as Teachers Working from the list of grammar topics which was produced, the teacher asked for volunteers to research an area of grammar, present their findings to the group as a whole, and provide exercises, tasks, games, and other activities for practice. For this activity, the researchers were given advice and guidance and were provided with relevant reference texts (grammars, dictionaries, and other textbooks). Once in the class, however, the teacher sat among the students and only gave assistance when called upon to do so. The purpose of these activities was to encourage the students to listen to each other and to become involved in thinking more deeply about organizing their learning. Sessions with a student as teacher had a characteristically more relaxed atmosphere than teacher-led sessions, and the students felt much freer to make mistakes, correct each other, and ask questions. Initially, the students showed a considerable range of abilities in leading such sessions, but, as the course progressed and they developed a clearer idea of what was expected, they became more expert in formulating their research findings and devising interesting and unusual practice activities. It was significant that in those sessions where the researchers clearly had not prepared sufficiently, the others in the class were, nevertheless, eager to contribute ideas.

INCREASING LEARNER INVOLVEMENT

603

After presenting the students with two different ways of making a request in Arabic, a discussion was started to encourage the students to think about factors that might influence the form of an utterancefor example, the roles of the speakers, the topic, and the situation. The purpose of this activity was to demonstrate that knowledge of a language not only consists of rules of grammar and related formal features but also of rules of useillustrated by the plus symbol (+) in the above diagram. The conclusions of this discussion were referred to periodically as various functional items of English were introduced in the course (e.g., how to agree/disagree, ask for permission, and so forth). Group Error Correction

The students were divided into small groups, each group being given a tape recorder in order to record their version of a roleplay. During the recording, students frequently stopped the tape and sought help from each other. As a follow-up to this activity, the groups were asked to listen to their recording, stop the tape where they thought they heard a mistake, and then discuss the correct form. The task sheet (above) was supplied to give them a focus. The purpose of this task was, once again, to encourage the students to make use of each other, drawing on the teacher only when necessary, and to think more deeply about the type of errors they typically made. General Error Correction In line with the purpose of the above task, the errors which students made in their homework were not corrected by the teacher. Instead, a mark was made only to show that there was a mistake, sometimes indicating the type (e.g., spelling, tense). Additionally, students were encouraged to exchange their homework and discuss it with each other.

604

TESOL QUARTERLY

Student-Directed Lessons The introduction of student-directed lessons marked the final step toward learner control of course management. After approximately the eighth week of the course, when it was felt by the teacher that the students could profitably begin to organize themselves and draw upon their experiences with the tasks set out above, two of their six hours in English a week were designated as student-directed. For this, the class was divided into groups of five or six, each group planning and carrying out its own learning activities together, with the teacher giving advice or correction only when called upon to do so. The remaining four teacher-led hours were used as an opportunity to continue to introduce students to other possible activities or approaches or to remind them of ones they had already experienced. As specific preparation for this last step, the class had previously been divided into smaller groups and assigned the task of compiling a list of 1) all the activities they could remember having done in a language class (e.g., dictations, roleplays, listening comprehension activities), 2) all the activities they had found useful or enjoyable, and 3) their ideas about what they felt they wanted or needed to learn in English. These lists thus formed a ready source of ideas for their student-directed lessons. Once the groups had experienced a few sessions by themselves, a general discussion was initiated to consider the problems of learning without teacher direction. From this discussion, groups were encouraged to draw up their own set of rules or code of conduct for their student-directed lessons. Typically, the following rules were agreed upon: 1. Only speak English. Only speak another language if it is very necessary. 2. Help each other and correct each others mistakes. 3. Only ask the teacher after you have asked the others in the group. 4. At each session, make a different person responsible for preparing something to bring to the next meeting. In their groups, students carried out a full range of the methods and activities commonly used by a teacher (dictations, grammar, explanations, pair work, dialogue building, communication games, and so forth). The atmosphere in these lessons was, as noted before, very relaxed and open, and the students freely helped each other and explained points to each other. Whereas it had always been difficult to persuade the students to speak in English when the teacher was in full control, it was particularly noticeable that in these sessions English was spoken more frequently than Arabic. The students were also much more prepared to use reference books than they had previously been. INCREASING LEARNER INVOLVEMENT 605

FINAL COMMENTS From the remarks made throughout the description of the above tasks, it should be clear that the students responded very positively to a movement toward placing more control in their hands. For students who had been described as very heavy going, they began to display considerable energy and enthusiasm for their student-directed lessons, the fruits of which became readily apparent. Three other traditional teacher-led groups were run simultaneously with students of the same type and ability who had similarly failed the preparatory year. At the end of the semester all students were required to re-take the examination they had failed the previous semester. On the average, the students in the experimental groups showed an improvement in their scores equal to or above that of the other students, confirming a point made by Beach that self-directed small group study does not result in any decrement in subject matter mastery (1974:197). Something, however, that the test did not (and could not) reveal is that these students had begun to develop skills and attitudes that went far beyond just the learning of English. The value of the approach lay particularly in the sense of responsibility that the students developed and in their subsequent change from a passive to an active role in the classroom. The requirements of the student-directed lessons brought the students into greater involvement with the course texts and stimulated them to take the initiative to look beyond those texts on their own, It would, however, be incorrect to suggest that there were no difficulties in implementing the ideas presented in this article. The students, coming from a very traditional background in almost all respects, initially showed considerable resistance to or lack of comprehension of the purpose behind the tasks, uppermost in their minds being the examination which they knew they were to re-take. Students in the teacher-led classes could comfort themselves in the belief that the teacher was aware of the contents of the final examination and would therefore teach accordingly. Such comfort, however, gradually became unavailable to the students in the experimental groups as they were required to take more control over the form and content of the course. These students were, therefore, not only forced to work harder to discover facts about the language, but they were also required to tolerate a high degree of uncertainty about the relevance of those facts. Thus, although there was no noticeable decline in morale among these particular students, one can speculate that the conflict between a learner-centered approach, as described in this article, and a formal examination-based system could lead to student demoralization.

606

TESOL QUARTERLY

The argument put forward in this article has been that increased learner involvement in course management can offer benefits in terms of depth of learning, motivation, and attitude toward studying. It was noted, however, that learners prior experience and expectations often make it difficult to introduce such a learner-centered approach. For this reason, one might expect that the ideas outlined here would prove more effective in cultures that place less emphasis on the authority of the teacher/leader and more on contributions by individuals. Similarly, one might also expect that the approach would be suitable for students who are already in tertiary education and thus, to some extent, experienced in self-directed study. Yet, the relative success of this experiment, given the background of the students involved, points to an interesting conclusion: that if adopted in a careful and gradual way, learner-centered approaches can offer significant gains among otherwise passive, teacher-dependent students. s

THE AUTHOR Andrew Littlejohn teaches at the University College of Bahrain and is Coordinator of Student Self-Access Facilities. He holds an M.A. in Linguistics for English Language Teaching from the University of Lancaster, England. REFERENCES
Allwright, Richard L. 1978. Abdication and responsibility in language teaching. Studies in Second Language Acquisition 2 (1):105-121. Allwright, Richard L. 1981. What do we want teaching materials for? English Language Teaching Journal 36 (1):5-18. Beach, Leslie R. 1974. Self-directed student groups and college learning. Higher Education 3:187-200. Berne, Eric. 1964. Games people play. New York: Grove Press. Breen, Michael P., and Christopher N. Candlin. 1979. Essentials of the communicative curriculum. Applied Lingutitics 1 (2):89-112. Breen, Michael P., and Christopher N. Candlin. In press. The communicative curriculum in language teaching. London: Longman. Fitz-Gibbon, C. I., and D. G. Reay. 1982. Peer-tutoring: brightening up FL teaching in an urban comprehensive school. British Journal of Language Teaching 20 (1):39-44. Hovey, D. E. 1973. Effects of self-directed study on course achievement, retention and curiosity. Journal of Educational Research 56:346-351. Krashen, Stephen D. 1982. Principles and practice in s e c o n d l a n g u a g e acquisition. Oxford, England: Pergamon Press.

INCREASING LEARNER INVOLVEMENT

607

Littlejohn, Andrew P. 1982a. Teacherless language learning groups: an experiment. Manuscript, University of Lancaster, U.K. Littlejohn, Andrew P. 1982b. A procedural guide for teacherless language learning groups. M.A. dissertation, University of Lancaster, U.K. Naiman, N., M. Frohlich, H. Stern, and A. Todesco. 1977. The good language learner. Toronto: Ontario Institute for Studies in Education. Rogers, Carl. 1969. Freedom to learn. Ohio: Merrill. Rubin, J. 1975. What the good language learner can teach us. T E S O L Quarterly 9 (1):41-51. Stevick, Earl W. 1976. Memory, meaning and method. Rowley, Massachusetts: Newbury House. Webb, N. J., and T. G. Grib. 1967. Teaching process as a learning experience: the experimental use of student-led discussion groups. W e s t D e P e r e , Wisconsin: St. Norbert College. Williams, H. P. 1930. Experiments in self-directed education. School and Society 31:715-718.

608

TESOL QUARTERLY

TESOL Quarterly, Vol. 17, No 4, December 1983

The Least You Should Know About Arabic: Implications for the ESL Writing Instructor
KARYN THOMPSON-PANOS
University of Colorado

MARIA THOMAS-RUZIC
Fulbright Foundation

A concern for the particular needs of university-bound Arabicspeaking students has been shared by many in ESL. Out of this concern, and especially out of concern for the characteristic English writing deficiencies of many Arab ESL students, various aspects of written Arabicfrom orthographical conventions to rhetorical devicesare discussed. These contrasting features have been identified as potential contributors to observed error production and weaknesses in some reading skills, but most particularly in writing skills. Ideally, a better understanding of the language background of Arab students can aid the ESL specialist in better addressing the special needs of these students through supplemental curricular objectives and appropriate exercises. Although contrastive analysis is no longer seen as a foundation for instructional programs (Schachter 1974), it can be a useful tool in understanding characteristic language-learning weaknesses demonstrated by a particular language group. Implicit in this article is the assumption that a familiarity with some salient contrasting features of written Arabic and English may prove valuable to those in ESL concerned with addressing noted weaknesses in Arab students English writing skills. DIGLOSSIA IN THE ARAB WORLD Altoma (1969) and others have described the phenomenon of diglossia, which exists throughout the Arab world. There are at least two distinct languages in every Arab country: Classical Arabic, also known in slightly modernized forms as Literary or Modern Standard Arabic, and colloquial Arabic. It is colloquial Arabic in its many regional varieties that is the first language for all Arabs. It is learned without formal instruction and, until very recently, was virtually never 609

written or read (there is now some drama written in the colloquial). Modern Standard Arabic, on the other hand, is the language of written communication and the form used for speeches, lectures, and on radio and television. It is, in a sense, a second language learned only through formal education. Large differences in vocabulary, pronunciation, and grammar exist between Modern Standard and the various colloquial. These differences, coupled with the basic phenomenon of diglossia itself, pose obvious problems in trying to address the notion of contrastive features and native-language sources of error. One hypothesis which has been proposed for further research, for example, is that errors in writing result from interference from Classical Arabic, while speech interference comes from the colloquial (Scott and Tucker 1974). WRITING SYSTEM Because poor handwriting can detract from the writers ideas, organization, and style, supplemental instruction in handwriting for students whose native alphabet is not Roman can result in improved legibility and the elimination of a major distraction to the reader. Middle Eastern students in particular have been found to have more difficulty in handwriting in English than students from other language backgrounds (Nevarez, Berk, and Hayes 1979). Simple recognition of certain basic mechanical features of written Arabic can assist the ESL teacher in addressing this problem. First of all, the Arabic alphabet, which is also used in a number of non-Arab countries, differs significantly from the Roman alphabet. There is little distinction between printing and script, and there are no capital letters. The Arabic alphabet contains twenty-eight letters, some of which have as many as four different shapes depending on their position in a word. In addition, Arabs write through the lines on their paper, rather than on top of them. However, perhaps the most notable contrast between the Arabic and Roman writing systems is that Arabic moves from right to left. This feature of Arabic, in addition to the very different appearance of its letters, which are formed by a series of strokes rather than a continuous flow, can pose problems for Arab learners when they form letters in English writing tasks. In reading, moreover, recognition of letters and words, handwritten as well as printed, can be a very deliberate and time-consuming process for the Arabic speaker decoding an unfamiliar alphabet. Not only the special difficulties of working in a foreign alphabet, but also the purely physical problem of adapting to an opposite direction of movement in reading and writing should be considered when the Arab learner is compared to faster, more proficient readers and writers 610 TESOL QUARTERLY

in the mixed ESL classroom. Speed, then, becomes a major obstacle in many reading and writing tasks, such as skimming, scanning, dictation, and note-taking. SPELLING Foreign learners of English, as well as many native English speakers, are plagued by difficulties in spelling. Although the morphophonemics qualities of the English spelling system may be recognized by some linguists (C. Chomsky 1970, N. Chomsky and Halle 1968, Schane 1970), its regularities are certainly not appreciated by most second language learners. While it may be reassuring to learn that nearly eighty-five percent of all English words have a regular spelling (Miller 1973), the number of rules and exceptions is considerable and the nature of the rules relatively complex. These aspects of English spelling, when combined with the fact that the Arabic writing conventions and vowel system are vastly different, contribute to particular spelling difficulties for Arab learners. Spelling errors constituted the single most common error found in a study of freshman compositions at the University of Petroleum and Minerals in Saudi Arabia (Beck 1979). In fact, errors in spelling were so pervasive that ninety-eight percent of all paragraphs examined contained at least one spelling error, with the majority containing many more. Ibrahim (1978) examined spelling errors in the written work of undergraduates at the University of Jordan and found that the majority of errors fell into the following (sometimes overlapping) categories: 1. Errors caused by the non-phonetic nature of English spelling, such as the inconsistency in spelling the weak vowels (*husbund, *biginner) 2. Errors caused by differences between the sound systems of English and Arabic, such as the substitution of the letter b for p as in *beoble (Arabic does not have two distinctive bilabial plosives, only the voiced /b/) and hypercorrected spelling that represents both b and p as p (*hapit, *compination) 3. Errors attributed to analogy, such as *languidge (compare knowledge), *maney (compare money), and *toled (compare liked) 4. Errors attributed to the somewhat inconsistent spelling in English word derivation, such as high/ *hight and speak/ *speach 5. Transitional errors resulting from ignorance or overgeneralization of a spelling rule. From our personal experience with Arabic speakers in the classroom, we would suggest an additional category of errors caused by differences THE LEAST YOU SHOULD KNOW ABOUT ARABIC 611

in the orthographic representation of the common vowel sounds in the two languages. We have observed a notable inconsistency in the way many Arabic speakers represent vowels in written English. For example, we discovered that one student spelled his name Muh-a m-a-d on some occasions and M-o-hame-d on others. His records revealed still another spelling M-o-h-a-mm-ad. These variations point to an obvious conclusion: the systems for representing vowels in the two languages lack close correspondence. In written Arabic the are articulated, they are not represented orthographically. An examination of the vowel system in Arabic can help to clarify this fact. Modern Standard Arabic makes use of a six-vowel system; there are three pairs of vowels, the vowels in each pair distinguished by length. first members of these pairs, the long variants, are the only ones always represented in writing. The shorter variants are not; rather, they are indicated only in childrens books, the Quran, and special texts (e. g., texts for foreigners) by diacritical markings placed above the consonant that precedes the vowel sound. The triangular system of Arabic vowels has broken down in the spoken colloquial, the short vowel variants having taken on a more centralized, shwa like quality. Nonetheless, in terms of writing, the conventions for vowel representation have important implications for spelling in English. First, it is not uncommon for syllables and even whole words in Arabic to be spelled therefore, that vowels in English words are often omitted altogether by Arab students. Second, the English vowel system is relatively complex. Spoken English distinguishes eleven or more (depending on dialect) distinctive vowel phones, but in writing makes use of only six vowel graphs (including y). The meaningful vowel variations found, for example, in beat, bit, bait, bet, pot, bat, bought, boat, put, but, boot, and bout can be difficult for the Arab to discriminate in listening, speaking, and reading, as well as in writing. Take, for example, the boot, boat, and bought. All of these distinct sounds in English would be allophonic variants for the Arab, that is, not meaningfully distinct. In written Arabic these sounds would all be represented by the same vowel graph. Many transliterated Arabic words and names illustrate the lack of a one-to-one correspondence with English: Muslim a n d Moslem, Quran and Koran, Muhamad and M o h a m e d . The instructor should be sensitive, then, to the difficulties all students face in the task of English spelling, especially in the matter of vowels. Furthermore, in the case of the Arab learner, the instructor who is aware of the very different writing and spelling convention 612 TESOL QUARTERLY

which characterize written Arabic will be better able to provide useful assistance. A systematic presentation of the meaningful vowel variations found in spoken English and their most common representations in writing is perhaps the best way to approach such spelling problems. In addition, the relationship of spelling to word families, roots, and derivations should be pointed out in order to help students to recognize the regularities that do exist in English orthography. VOCABULARY Insofar as we are concerned here with writing and word-level problems, certain remarks can be made with reference to English vocabulary and the Arab learner. Arab and Asian students in the mixed ESL class are often at a disadvantage in terms of vocabulary when they are compared with students from Romance and Germanic language backgrounds, which feature many cognates. Limitations of vocabulary can be an obstacle in all language skill areas. ESL instructors might observe that their Arab students characteristically make relatively little use of a dictionary. Although there are cultural and educational reasons for the observed lack of good dictionary skills among Arab students, it is also the case that using a dictionary in Arabic is a difficult task because words in an Arabic dictionary are arranged according to their word root. The three-consonant word-root system, which is the basis of most of the lexicon, is one of the most outstanding features of Arabic and other Semitic languages. For example, the verb to study has the root dr-s in Arabic. Related nouns, verbs, and adjectives such as to teach, teacher, studious, studies, school are formed by adding different prefixes, infixes, and suffixes to the root. Early Arab lexicographers began the practice of entering all words in the dictionary under the root, and this has remained as the basis of organization for Arabic dictionaries. Using the dictionary in Arabic can be understandably difficult (somewhat comparable to looking up the English word misconceived under cept). Many of our Arab students, for these and other reasons, have not acquired good dictionary habits for reading and writing. For the writing instructor trying to aid students in making greater use of varied vocabulary, two implications can be seen from the Arabic word-root-based lexicon: 1) additional instruction and practice in developing English dictionary skills should be provided, and 2) the concept of word derivation should be exploited. By this we mean that analyzing a word and using its different forms (e. g., to criticize, critic, criticism, critical) is a familiar concept for Arabic speakers, who can learn to express themselves better in writing by applying high-frequency derivational forms in English. Once again, systematic THE LEAST YOU SHOULD KNOW ABOUT ARABIC 613

presentation and practice of selected affixes enables the students to learn an entire system of vocabulary rather than individual words. It should be noted, however, that knowing the various word forms of a given lexical item does not necessarily mean that students will be able to use these correctly. Beck (1979) found that word-form errors were the sixth most common written error in his study of Arab students writing, occurring in approximately fifty percent of all paragraphs studied. Arabic and English have differing distributions of nouns, verbs, and adjectives, and often errors in word use are related to semantic and distributional differences in use rather than to structural dissimilarities. SYNTACTIC FEATURES Basic familiarity with some of the structural and syntactic differences between Arabic and English can be quite useful to the typical ESL teacher faced with a heterogeneous, multi-cultural group of foreign students, each with his or her own special needs and weaknesses. An error analysis by Scott and Tucker (1974) found the four most problematic grammatical features of English for Arabic-speaking students to be verbs, prepositions, articles, and relative clauses, Studies done by Mukattash (1981) and Beck (1979) support this observation. Preposition misuse is largely a lexical problem, involving significant Arabic interference and infrequent miscommunication (Scott and Tucker 1974), and therefore will not be treated in this article. First language (Ll) interference also accounts for at least half of the errors with articles, a significant portion of which result from omission of the indefinite article (Scott and Tucker 1974). Although there is an indefinite morphological marker in Arabic, it is usually unspoken and unwritten, so indefiniteness is indicated by the absence of a definite article. Extension of this principle in learning a second language frequently results in the Arab students overlooking the indefinite article in English. It must be realized, however, that the use of articles in English is problematic for students from many language groups and should be addressed as difficulties arise. Because verb usage and relative clause formation are more complex problems and are particularly pervasive in Arabic students written English, they merit fuller discussion here. Verbs To begin with, one of the most frequent verb errors among Arab students is the omission of the copula. Even advanced students on occasion make such errors as *he absent or *my teacher very angry. 614 TESOL QUARTERLY

These sentences, although ungrammatical in English, would be well formed in Arabic. They are examples of equational sentences, which correspond to English sentences with be in the present tense affirmative. The major difference here is that there is no surface-structure copula or verb present in Arabic sentences of this type. If transferred to English, errors such as those above result, with the copula omitted. This structure persists in being problematic for many Arab students, as Beck (1979) and Scott and Tucker (1974) have demonstrated. In addition, Arabic tends to use verbs to describe states more frequently than English. Bateson (1967) cites the example of Arabic preferring h e grieves or he rages to he is unhappy or he is angry. Perhaps a more complex problem to analyze and address in the writing classroom is the actual misuse of verb forms in English. Because the rules governing the use of verbs in Arabic sentences are quite distinct from those in English, an examination of the Arabic verb system is essential if the students difficulties are to be understood. First of all, Arabic is a highly aspectual language. While English can combine a number of tenses with simple, perfective, and progressive aspects, Arabic makes two basic distinctions: the perfect and imperfect aspects. The perfect is used to describe a completed action (frequently in the past), while the imperfect describes a situation not yet completed (often in the present or future). However, since these aspects derive their meaning from the point of view of the completion or incompletion of an activity rather than the time of completion or incompletion, both aspects may be used to describe an action in the past, present, and future. For example, Abboud et al. (1975) indicate in their textbook on Modern Standard Arabic that the imperfect is used to describe a past habitual, past progressive, or past future activity, while the perfect is used to denote a completed event or to describe actions that would require the present perfect or past perfect in English. Since the forms of the Arabic verb have little actual time reference (in the English sense), certain invariable particles and conjugated auxiliary verbs can be employed to lend various modal and temporal meanings which might not be clear from context. These particles, which in form resemble English modals or other auxiliary verbs, are used with fully conjugated perfect or imperfect verbs. Problems in English can result if AUX + participle or MODAL + base-form verb are confused with the PARTICLE + perfect or imperfect verb construction in Arabic. For example, the following errors produced by Arab students are not uncommon: *I didnt went to school, *She might didnt understand. In addition, progressive aspect in Arabic is frequently indicated by adjectives derived from the roots of verbs, in which case the sentence becomes equational rather than verbal. The question Where are you THE LEAST YOU SHOULD KNOW ABOUT ARABIC 615

going? in English uses the present progressive form of the verb go. The same question in Arabic, however, would employ a dynamic adjective derived from the verb, as in Where you going?, in which g o i n g functions as an adjective. Another serious problem for Arab learners of English is the misuse of verb tenses (Mukattash 1981). Although the task of correctly forming the various verb tenses is in itself manageable, control over the distribution rules for the usage of the English tenses seems to be a late acquisition in the learning of the verb system (Scott and Tucker 1974). One frequently observed source of inconsistency and difficulty for Arab students is the sequence of tenses across clauses. In Arabic, temporal clauses are frequently in the imperfect (present) tense. The meaning and time reference of the verb in a subordinate clause are derived from the time of the verb in the main clause. Therefore, an imperfect verb in a subordinate clause following a perfect verb in the main clause refers to past time (Abboud et al. 1975). For example, the following sentence, translated directly from Arabic, shows how the imperfect tense in a subordinate clause refers to the same time as the verb in the main clause: The minister arrived (perfect) while he carries (imperfect) an important letter from the president (Abboud et al. 1975:435). Here, the imperfect tense denotes an action taking place at the same time as the main verb. In English, the same idea would be expressed by the following sentence: The minister arrived carrying an important letter from the president. The following sentence from Arabic is an example of the use of the perfect tense with a particle in the subordinate clause: The reporter returned (perfect) to his country while he (particle) talked (perfect) with the president (Abboud et al. 1975:437). The use of the perfect aspect in subordinate clauses indicates a completed action, and the particle clarifies the sequence of events. The most unambiguous translation in English would be: The reporter returned to his country after having talked with the president. It is this type of construction that indicates maturity of style to native English speakers. The differing rules governing tense distribution in subordinate clauses of Arabic and English can result in ambiguous or poorly formed complex sentences if Arabic speakers do not recognize them. To deal with problems of English tense usage, meaningful exercises illustrating both tense and aspect discrimination should be provided. Particular emphasis should be given to temporal clauses and 616 TESOL QUARTERLY

the progressive and perfective aspects, which are expressed in significantly different ways in Arabic. It is noteworthy that the English perfect tenses were found to be especially problematic for the Jordanian students in Mukattashs study of question formation (1980). The ESL teacher and materials developer should focus on the instruction and practice not only of the formation of the various verb tenses but, just as importantly, on their distribution rules and usage as well. Relative Clause Formation Another sentence-level feature of Arabic that differs in a number of important respects from English is relative clause formation. Because relative clauses are such a frequent and important construction in English, and because Arabic-speaking students have so much difficulty with their formation (Scott and Tucker 1974), although not with their use, an examination of their counterpart in Arabic should prove insightful. Superficially, relative clause formation in Arabic is very similar to that of English. For this reason, in a study conducted by Schachter (1974) on the use, misuse, and avoidance of relative clauses by Arab, Persian, Japanese, and Chinese students, Arabic speakers were predicted to have less difficulty than Japanese and Chinese students, whose native languages have a significantly different syntax for relative clauses. The results of the study indicated that Arab and Persian learners, seeing similarities between their L1 and English, transferred their L1 strategy to English, thereby producing errors. Japanese and Chinese students, on the other hand, seeing radical differences between L1 and English, approached relative clauses with extreme caution or avoided their use altogether. There are three major differences between English and Arabic relative clause formation. First, there is no relative pronoun in Arabic. Rather, a relative particle, part of neither clause, links two complete clauses. This particle is present only when the antecedent is definite, as in the sentence (translated literally from Arabic) I saw the boy who he has red hair. When the antecedent is indefinite, however, no relative particle occurs, as in I saw a boy he has red hair. Omission of the relative pronoun in English in sentences where it is the subject of the clause is directly attributable to Arabic interference, according to Scott and Tucker (1974). Another important difference is that the antecedent clause and relative clause in Arabic are both complete sentences; neither is subordinate, at least not in the surface structure. In fact, if there is a pause in reading, or if there is written punctuation, the result is two

THE LEAST YOU SHOULD KNOW ABOUT ARABIC

617

independent sentences (Abboud et al. 1975). We can therefore see that the relative clause construction in Arabic is coordinate, rather than subordinate as in English. The most serious source of error production for Arabic speakers learning English relative clauses is the presence in Arabic of a relator in the relative clause. This is a second word or affix that serves as the subject or object of the clause and refers to the antecedent. When transferred to English, the repetition of referents results in aberrations described by some as Middle Eastern clauses (e.g., *The girl who she was pretty came and *This is the record which I bought it). S c o t t and Tucker (1974) note that the object-deletion rule is acquired later than the rule for subject deletion, indicating that object deletion will require more attention and practice in the ESL classroom. Repetition of the object was the most frequent of all relative clause errors analyzed in the Scott and Tucker study, again attributable to L1 interference. Yorkey (1977) proposes a sequence of exercises that will prepare students for relative clause formation. For example, three initial activities in teaching the formation of object clauses include 1) simple pronoun substitution for nominals in dependent clauses, 2) insertion of which, that, or w h o ( m ) and deletion of object pronouns, and 3) deletion of the object relative pronoun and object pronoun, as seen in the exercises below: 1.My father smokes cigars. My father gets the cigars from Havana. them He 2. My father smokes cigars WHICH he gets ( ) from Havana. 3. My father smokes cigars ( ) he gets ( ) from Havana. In addition, sentence-combining exercises which give practice in forming relative clauses are useful here. For example: Nearly all the teachers (the teachers taught there) were experienced. The first school (I went to that school) is situated on the main road. Children (the childrens fees were overdue) were given letters to their parents. (Tadros 1979:327) Eventually, students should perform less controlled sentence-combining exercises where the clause to be embedded occurs in a variety of positions and the students must determine its proper placement, such as: The first school is situated on the main road. I went to that school. The teachers brought sandwiches with them. Their homes were far from the school. (Tadros 1979:328) 618 TESOL QUARTERLY

STYLE Other possible obstacles to good English writing are certain stylistic devices for assertion and exaggeration in Arabic which may be transferred into written English. Much of the literature on Arabs and their language points out that it is part of Arabic linguistic tradition that main points are overasserted and exaggerated (Patai 1976, Hamady 1960, Shouby 1951). For example, Arabic uses special word endings, ways to double consonants, and rules for redundant pronouns ( M y professor he is funny), as well as other stylistic and rhetorical devices, to achieve exaggeration. Literary studies of Arabic poetry, investigations of Arab and American judgments of written messages (Prothro 1955), and analyses of political speeches (Suleiman 1973) indicate that the Arabs greater use of exaggeration and assertion, relative to English, extends to all language communication. In both spoken and written Arabic, repetition, increased use of the superlative, and frequent rewording and restatement are devices used to communicate ideas clearly. Shouby (1951) writes that Arabs stand a good chance of being misunderstood, in Arabic, unless they overassert and exaggerate; he goes further in noting that if Arabs say exactly what they mean without the expected exaggeration, other Arabs may not only miss the point but may interpret the message to mean just the opposite. The ESL instructor may find it useful to be aware of these stylistic differences, especially in writing tasks involving personal opinion and argumentation. PARAGRAPH AND COMPOSITION Advanced students who have adequately mastered intermediatelevel grammar and have little difficulty with sentence-level grammar are often frustrated and perplexed to find comments such as awkward, lacks organization, and out of focus on their papers. As Kaplan (1966) points out, sentence-level mastery is not the ultimate goal for ESL students; sentences occur in a larger context. Foreign students who have mastered English syntax must learn its logic and rhetoric in order to relate syntactic elements within a paragraph and to relate paragraphs within a total context. Kaplan suggests that contrastive rhetoric be researched and applied to the teaching of writing. It is important to keep in mind here that Arabic and English use different organizational styles and that Arabic speakers benefit from having these differences pointed out to them. It is incumbent upon the ESL specialist both to be familiar with the differences in organization between the students native language and English and to make the students aware of these differences so that they are better prepared to meet the expectations of their readers. THE LEAST YOU SHOULD KNOW ABOUT ARABIC 619

While paragraph development in Arabic and other Semitic languages can be seen as a series of parallel constructions, with parts of sentences connected by coordinating conjunctions, maturity of style in English is measured by the degree of subordination rather than coordination (Kaplan 1966). As Cowan notes, Linguistically speaking, Arabic as a language compounds and is associative. College English skills require analysis and subordination of thought, Arabic requires synthesis and coordination (1978: 11). In fact, infrequent use of subordination and overuse of coordination, particularly coordinating conjunctions at the beginning of sentences, comprise the chief characteristic of Arabic speakers written English (Yorkey 1977). This is largely because Arabic sentences emphasize sequences of events and balance of thought, which favor coordination. When transferred to English, they also frequently lack the types of structures (such as participial phrases and adverbial clauses) that ESL teachers look for in writing classes and university professors expect on campus. For example, Kaplans examination of an Arab students English composition considered the following sentence, which was built largely on parallelism: I was very astonished and a little bit frightened, and when I saw my father and mother a little bit confused, I tried to be courageous, and I went out to see what was happening (1967:14). Kaplan suggests that this sentence, which was revised from an earlier draft, could have been rewritten in, for example, one of the following ways: I was very astonished and frightened, and when I saw my father and mother a little bit confused, trying to be courageous, I went out to see what was happening. I was very astonished and frightened when I saw my father and mother a bit confused. Trying to be courageous, I went out to see what was happening. Kaplan continues with the following suggestion for ESL teachers: If it is desirable in English to relate certain ideas in terms of main and subordinate structures, then the foreign student needs to be taught how to determine which structure he should make subordinate and why. In the same way that this must be accomplished at the syntactic level through the teaching of structures of modification, it needs to be accomplished at the rhetorical level by teaching the larger structures of modification; that is, the kinds of paragraphs which are intended to advance the thought of the whole essay as well as the kinds of paragraphs which are intended to go back over ground already covered and supply the necessary support, analogy, metaphor, illustration, etc. (1967:15). Areas needing particular emphasis in ESL instruction are adverbial 620 TESOL QUARTERLY

clauses of time, place, result, concession, cause, and purpose as well as distinctions between cause and effect, real and unreal conditions, and main ideas and supporting ideas (Yorkey 1977). Activities that address the teaching of organization include completion of paragraphs with sentences omitted, rearrangement of sentences presented in a scrambled order into the correct order, outlining, and providing students with topic sentences to be developed into a composition. SUMMARY Familiarity with students typical errors and problem areas is a responsibility of all ESL instructors. This article has addressed salient features of written Arabicorthography, spelling, vocabulary, sentence grammar, style, and rhetorical organizationand how these features contribute to weaknesses which have been observed in the reading and writing skills of Arabic-speaking students engaged in university-level tasks. Greater sensitivity to these issues can help the ESL specialist to assess and address the needs of Arab students whose writing must meet university standards. By alerting the students to specific areas deserving attention, the writing instructor can be instrumental in promoting effective self-monitoring and the development of individualized objectives on the part of each student who is struggling with the task of learning to write well in English.

ACKNOWLEDGMENTS
This article is a revised version of a workshop given at the, 1981 TESOL Convention in Detroit. We wish to thank Jean Engler, the Director of the International English Center at the University of Colorado, for her support and editorial comments.

THE AUTHORS
Karyn Thompson-Panos is an instructor and curriculum coordinator at the International English Center, University of Colorado. She received her M.A. in TEFL from the University of Michigan and has taught in Barcelona, Spain. She has presented workshops on the Arabic language with Maria Thomas-Ruzic at TESOL conventions and at NAFSA and SIETAR conferences.

Maria Thomas-Ruzic has just completed two years as a Fulbright lecturer in Yugoslavia. She received her M.A. in Linguistics from the University of Colorado, where she was an instructor at the International English Center. She has also taught ESL in Italy and as a Peace Corps volunteer in Cameroon. THE LEAST YOU SHOULD KNOW ABOUT ARABIC 621

REFERENCES
Abboud, Peter F., Najm A. Bezirgan, Wallace M. Erwin, Mounah A. Khouri, Ernest N. McCarus, Raji M. Rammuny, Zaki N. Abdel-Malek, and George N. Saad. 1975. Elementary modern standard Arabic: part one. R e v i s e d Edition. Ann Arbor, Michigan: University of Michigan. Altoma, Salih J. 1969. The problem of diglossia in Arabic. C a m b r i d g e , Massachusetts: Harvard University Press. Bateson, Mary Catherine. 1967. Arabic language handbook. W a s h i n g t o n , D.C.: Center for Applied Linguistics. Beck, Robert Daniel. 1979. An error analysis of free composition. Teachers of English: Arabian Monthly 30:20-26. Chomsky, Carol, 1970. Reading, writing, and phonology. Harvard Educational Review 40(2) :287-309. Chomsky, Noam, and Morris Halle. 1968. The sound pattern of English. N e w York: Harper and Row. Cowan, James W. 1978. Factors influencing Arab and Iranian students incountry and in the United States. In Students from the Arab world and Iran, Gary L. Althen (Ed.), 1-13. Washington, D. C.: NAFSA. Hamady, Sania. 1960, Temperament and character of the Arabs. New York: Twayne Publishers. Ibrahim, Muhammid H. 1978. Patterns in spelling errors. English Language Teaching Journal 32(3):207-212. Kaplan, Robert B. 1966. Cultural thought patterns and inter-cultural education. Language Learning 16(1-2):1-20. Kaplan, Robert B. 1967. Contrastive rhetoric and the teaching of composition. TESOL Quarterly 1(4):10-16. Miller, Wilma H. 1973, Diagnosis and correction of reading difficulties in secondary school students. New York: Center for Applied Research in Education, Inc. Mukattash, Lewis. 1980. Yes/no questions and the contrastive analysis hypothesis. English Language Teaching Journal 34(2): 133-145. Mukattash, Lewis. 1981. Problems in error analysis. In Papers and Studies in Contrastive Linguistics 13, Jacek Fisiak (Ed.), 261-274. Washington, D. C.: Center for Applied Linguistics. Nevarez, Hector, Virginia Berk, and Curtis W. Hayes. 1979. The role of handwriting in TESOL. TESOL Newsletter XIII(1):25-26. Patai, Raphael. 1976. The Arab mind. New York: Charles Scribner's Sons. Prothro, Terry E. 1955. Arab-American differences in the judgment of written messages. Journal of Social Psychology 42:3-11. Schachter, Jacquelyn. 1974. An error in error analysis. Language Learning 24(2):205-214. Schane, Sanford. 1970. Linguistics, spelling, and pronunciation. T E S O L Quarterly 4(2):137-141. Scott, Margaret Sue, and G. Richard Tucker. 1974. Error analysis and Englishlanguage strategies of Arab students. Language Learning 24(1)59-97. Shouby, E. 1951. The influence of the Arabic language on the psychology of the Arabs. Middle East Journal 5:284-302.

622

TESOL QUARTERLY

Suleiman, Michael W. 1973. The Arabs and the West: communication gap. In Intercommunication among nations and peoples, Michael H. Presser (Ed.), 287-303. New York: Harper and Row. Tadros, A.A. 1979. Arabic interference in the written English of Sudanese students relativisation. English Language Teaching Journal 33(3):234-238. Yorkey, Richard. 1977. Practical EFL techniques for teaching Arabic-speaking students. In The human factors in ESL, James E. Alatis and Ruth Crymes (Eds.), 57-85. Washington, D, C.: TESOL.

THE LEAST YOU SHOULD KNOW ABOUT ARABIC

623

TESOL QUARTERLY, Vol. 17, No. 4, December 1983

Recent Language Research and Some Language Teaching Principles

KARL J. KRAHNKE Colorado State University MARY ANN CHRISTISON Snow College

The underlying assumption of this article is that linguistic and communication research can support principles on which to base revisions in the content and method of language teaching. The article first reviews the results of research in language acquisition, interfactional analysis, pragmatic, repair, error, and social and affective factors. Four general principles are then extracted from this research, principles relating to acquisition activities in the classroom, the importance of affective factors, the communicative capacity of learners, and the nature and treatment of error.

INTRODUCTION
Over the last ten to fifteen years, the language teaching profession has been provided with a wealth of new information on language and language use from several areas of linguistic and language-related research. As is usual, the problem for designers and providers of language instruction has been to determine how to assimilate, evaluate, and apply this new knowledge. The process of application really involves two questions: the effect of research on what aspects of language behavior we choose to teach and the effect of research on how we teach. Much recent work has focused on the former question, especially the work done for the European Unit/Credit system (van Ek 1975) and much of the work done in Britain on communicative language teaching. Methodological questions have been addressed in the work of Curran (1968, 1972, 1976, 1978), Gattegno (1972, 1976), Lozanov (1979), Asher (1977), Winitz and Reeds (1973), and Terrell (1977, 1982). None of these, however, has attempted to draw systematically on the results of broad-based research in developing their methodologies. Stevick (1976, 1980, 1982) has done so in a personal and practical way, but his recommendations are often more eclectic than systematic. Krashen (1981, 1982) has begun to address questions of 625

methodology and approach, but his view is tied quite closely to the model of second language acquisition which he has developed. All of this work is instructive and useful. The pedagogical recommendations of Krashen (1982) and Krashen and Terrell (1983), in particular, come closest to bridging the gap between theories of language and language acquisition and actual classroom techniques. This article does not attempt to address that same issue but rather focuses on how an examination of recent language-related research can suggest principles on which effective language teaching can be based, principles which relate to approach, design, and procedure in language teaching (Richards and Rodgers 1982). What this article consists of, therefore, is a review of research that relates to questions of instructional methodology and the suggestion of four principles which can be used to guide specific methodological decisions. ACQUISITION AND LEARNING A useful theoretical construct in language acquisition research is the now widely accepted distinction between learning and acquisition (Krashen 1978). Krashens hypothesis is that adult language learners have two different means available to them to gain knowledge of a second language learning and acquisition. In brief, learning results from deliberate, structured, and conscious attempts to learn the language, such as through drills and rule memorization. Acquisition, on the other hand, results from natural and meaningful interaction with other speakers in the target language. According to this hypothesis, learning is very different from acquisition in that it only functions to correct or monitor (Krashen 1978) the language being produced. These corrections or changes can take place before or after a sentence is actually spoken or written, but in order to make the correction the learner must recognize the error, know how to correct it, and have sufficient time to do so (see Krashen 1978 and 1982 for a more detailed consideration of this point). In immediate, face-to-face or pen-topaper, communicative encounters, these conditions can rarely be met, so it is argued that learning may not be as important in developing the ability to communicate fluently, verbally or in writing, as we previously believed. Rather, according to the hypothesis, fluency develops gradually as a result of communicative experience in the target language, with progress occurring when the learner is able to understand most of what is contained in the communication but when that communication also contains language material the learner needs and is ready to acquire. Krashen (1982:9) refers to this understandable language as comprehensible input and defines it as language that is comprehended by the learner but which still contains and, therefore, 626 TESOL QUARTERLY

provides new structures or other language material that the learner has not yet acquired. That is, it is hypothesized that learners can acquire language when they have to use it to communicate their wants and needs, when the focus is on what they need to say rather than on how they should say it. More research is being done to clarify the relationship between learning and acquisition, but our present understanding of the distinction seems sufficient to suggest an explanation for the evident lack of positive transfer to actual performance in communicative situations in the target language of much language instruction in which the emphasis is on drills and rule memorization. What the learning/acquisition hypothesis suggests is that activities which result in learning may have a limited role in an overall language instruction program, a role of assisting learners in developing learning- and monitoring-sensitive abilities in such areas as spelling, punctuating, and test taking, for example. But, the hypothesis suggests, if effective use of the language in realistic settings is the goal of the program, then learning-fostering activities should not dominate the language classroom to the exclusion of acquisition-fostering activities which help students develop spoken and written fluency in the target language. LANGUAGE ACQUISITION Research in language acquisition has focused on three different types of acquisition: 1) first language acquisition in children, 2) second language acquisition in children, and 3) second language acquisition in adults. Data have been collected in formal and in informal environments from all three types of subjects. From theories and research in first language acquisition we can draw conclusions for second language teaching and learning. There are, undoubtedly, important cognitive, affective, neurological, and physical differences between the acquisition of a first and a second language and between children and adults. At the same time, however, there are important similarities; essentially, both children and adults acquire languages from similar types of experiences and at essentially similar rates when conditions are optimal. Although it has been suggested that children are more likely to be able to take advantage of these experiences than adults (Asher and Garcia 1969), or at least have the advantage of receiving real intake while adults may not (Wagner-Gough and Hatch 1975), the question for language pedagogy is not the degree of similarity or difference between first and second language acquisition but rather the recognition of the similarities so that those similarities, and not just the differences, may be taken advantage of to increase effectiveness in second language instruction. LANGUAGE RESEARCH AND TEACHING PRINCIPLES 627

There are three basic ways in which first and second language acquisition are similar. They are summarized as follows: 1. Natural order. One area of research which has received considerable attention in the past few years is the study of the order of acquisition of various morphemes. Brown (1973) first demonstrated that there was a natural order in which children acquired some of the morphemes of their first language. Browns longitudinal findings were confirmed by de Villiers and de Villiers (1973). Studies of the sequence in which adults acquire grammatical morphemes in a second language began with the work of Bailey, Madden, and Krashen (1974), who confirmed that adults also showed a regular order in acquiring the eight morphemes studied. Dulay and Burt (1974), using the Bilingual Syntax Measure (Burt, Dulay, and Hernandez 1975), established a similar order for children learning a second language. Andersen (1976) and Krashen, Butler, Birnbaum, and Robertson (1978) reported natural orders in the written production of adults. Although there are some problems with these studies, specifically concerning the statistical procedures used, whether the natural ordering is an artifact of the Bilingual Syntax Measure, a n d whether or not morphemes in obligatory contexts reveal enough about the language acquisition process (Larsen-Freeman 1975, Porter 1977, Hakuta and Cancino 1977, and Rosansky 1976), there is no counter-evidence to the hypothesis that the existence of a natural order is part of the creative construction process in both adult and child language acquisition. 2. Comprehension. Children acquiring first languages learn to comprehend before they learn to speak. Winitz and Reeds (1973) claim that in first language acquisition, comprehension outpaces production by about a year. They claim that this sequence of development is a functional property of the brain and cannot be reversed or greatly modified. Recent studies support the hypothesis that delaying speech in second language instruction, while at the same time providing active listening, causes no delay in attaining proficiency in a second language (Gary 1975, Postovsky 1977). In fact, according to Gary (1974), early oral performance may not be profitable for children and adults studying second languages in formal settings. Terrell (1982) has also found strong evidence to support a preproduction phase for classroom instruction. 3. Linguistic input. Language acquirers in natural situations receive what we call comprehensible input (see above). In first language acquisition this type of input is called caretaker speech, and in second language acquisition it is sometimes called foreigner talk. Hatch (1979) summarizes some of the important characteristics of simplified input: reduced rate of speech (words and phrases are not 628 TESOL QUARTERLY

stretched, but more frequent pauses are inserted at possible boundary points), shorter utterance length, greater use of redundancy (repetition and gesture), simplification of syntax, and more frequent use of specific discourse phenomena (e. g., yes/no questions and tag questions). All of these modifications to normal adult speech assist learners in making useful guesses about the language and in developing an internalized working grammar of it. Snow (1979) reports that children figure out underlying structures with the aid of simplified input. Rubin (1975), Terrell (1982), and others have suggested that the same processes operate for adults. From the point of view of research, the role of conscious learning in the language acquisition process remains unclear. Much of present language pedagogy is based on the assumption that language acquisition can be quantitatively and qualitatively improved through formal instruction in the language, especially instruction in the grammatical regularities of the language. Lee, McCune, and Patton (1970), however, have provided evidence that strongly suggests that students do not pay much attention to repetitive drills after a few repetitions, and Hendrickson (1976) has noted the lack of effect on student writing of explicit correction of formal errors. But, at the same time, it is obvious that many language learners have somehow benefited from being exposed to explicit grammatical instruction. Krashen (1982) argues that the role of conscious learning in language acquisition is limited almost entirely to being a source of input for those learners who are open to it, and that there is little or no transfer from formal, conscious learning to communicative performance. Others (Bialystok 1978, Stevick 1982) have posited a greater role for conscious learning but with little evidence to support their views. Given the strength of the acquisition hypothesis, it is difficult to support a view that calls for a central role for conscious learning in language instruction. Since the input which leads to language acquisition is ultimately not defined by the type of language available to the learner but rather by how the learner is able to utilize it, the acquisition hypothesis presents the stronger case that acquisition is an incidental outcome of instruction. Instruction may certainly serve as input for some learners, but it is an additional empirical question whether instruction intended to lead to conscious learning is the most direct route to a useful competence in a new language. Conscious learning may positively affect the learning of some routines and patterns (Krashen 1981:96-99) and limited aspects of language behavior, such as editing written work, but there is strong evidence that extensive communicative experience with a new language is necessary for communicative competence to develop (Carroll 1967, Saegert, Scott, Perkins, and Tucker 1974). What remains to be better determined is the nature LANGUAGE RESEARCH AND TEACHING PRINCIPLES 629

of the relationship between learning and acquisition and the nature of classroom experiences that can foster communication and acquisition. INTERACTIONAL ANALYSIS Since acquisition is hypothesized to be a product of meaningful interaction, it would be useful at this point to review what has been discovered about the form and content of conversation and verbal interaction. The sociological perspective on which much of this work is based was set by Goffman (1967, 1971), but the formal or structural analysis of conversation is a relatively new topic of interest. Most of the work has been done by the late Harvey Sacks and his colleagues, Schegloff and Jefferson. Their work (especially Sacks, Schegloff, and Jefferson 1974) has established the turn as the basic unit of verbal interaction and has characterized conversations as being governed by a turn-taking system. Their analyses are purposely focused on the structural properties of this system, on what behavioral devices are used to establish and hold turns and to open and close conversations. They are less concerned with which linguistic forms are used to accomplish these ends. Their work has established that competent speakers of a language use a system for structuring verbal interaction which is somewhat independent of the form and meaning of the language used in the interaction. There are, in short, expectations and regularities that typically operate in conversation even when, through error, misunderstanding, or perversity, a speaker produces an utterance which does not seem to perform an appropriate function. Coulthard cites an extreme example of a doctor and schizophrenic patient maintaining the turn-taking system even though the patient does not fill his turn appropriately: A: What is your name? B: Well, lets say you might have thought you might have had something from before, but you havent got it anymore. (Coulthard 1977:63) A: Im going to call you Dean. While this example is bizarre, the careful observation of turn-taking formalities in it demonstrates that conversation can be governed by structural constraints, at least at a basic level. Some details regarding how turn selection is done seem to differ from culture to culture, but the system of taking turns and the expectancy on the part of speakers to do so is universal enough that it constitutes a basis for verbal interaction by language learners in the new language well before they are competent enough to participate skillfully in that interaction (Conrad 1982), The learners come to the new language with the expectation that it will be organized into turns and that filling turns is a minimally adequate way of sustaining conversation. They also assume 630 TESOL QUARTERLY

a system of openings and closings, the details of which must be learned or acquired. The concept of turn taking has been elaborated with the notion of adjacency pairs (Sacks, cited by Coulthard 1977:70), a frequently occurring but specialized pair of turns that are related in that the nature of the second turn is predicted by the first, Question-answer is one obvious pair; greeting-greeting is another. A small list is given by Richards (1980:421), with the suggestion that language instruction address this phenomenon more directly. But the phenomenon is a discourse universal and, as such, constitutes pre-existing knowledge of the structure of discourse. Language instruction can use this preexisting competence as a basis for instruction in the specific forms that are used in adjacency pairs in the language being learned. Gordon and Lakoff (1975) and Grice (1975) have identified several principles or postulates that, they claim, govern the conduct of conversation. These principles are the sincerity condition, the reasonableness condition, and the cooperative principle. Taken together, these principles assert that participants in a conversation assume, unless there is evidence to the contrary, that their conversational partners will be sincere and reasonable in the conduct of their conversational behavior and that they will be clear and orderly in what they say. These are basic notions, to be sure, but their suspension predictably requires participants in interaction to use more complex interpretation strategies to process utterances in which the principles are violated. We may offer the following example: A: How was your date last night? B: I got a good nights sleep. By not relating the answer directly to the question, A forces B to do more complex work to interpret the answer. In most normal conversations, however, conversational principles are strictly observed, giving second language learners another set of tools to use in deciphering the language that is directed to them. If they assume that the interaction they are involved in is governed by universal principles of conversation, then they have a structured predisposition and set of expectations as to how to interpret the speech presented to them. PRAGMATIC T O the familiar division of language analysis into phonology, syntax, and semantics, pragmatic has been added to account for the phenomena that relate a given utterance to its functions or meaning in a given instance of conversational interaction. Its cold in here may be used to LANGUAGE RESEARCH AND TEACIIING PRINCIPLES 631

inform, but it may also be used to get someone to close a door or window. Pragmatic encompasses a wide range of contextual factors including, among others, social and physical circumstances, identities, attitudes and beliefs of participants, and the relations that exist among participants. Pragmatic competence gives speakers the ability to correctly interpret sentences such as Its cold in here in the way they are intended, to choose the appropriate form of address for another person in a given situation, and to fulfill many other interfactional functions in between. When there are problems with the pragmatic system, misunderstandings can be quite serious, as in the case of the Japanese student who almost abandoned a new friendship because the new friend had remarked, in haste, Im in a hurry nowIll call you sometime. The Japanese girl interpreted this as a permanent dismissal. There has been considerable research into the pragmatics of English from the point of view of how the pragmatic system is used by competent speakers. In addition, Fraser, Rintell, and Walters (1980) have carried out some initial work into the acquisition of pragmatic by second language speakers and have presented a suggestion for how research should proceed. Their work is based on three assumptions (1980:78): 1. There is a basic set of speech acts common to all languages. 2. The same set of strategies for performing speech is available in all languages. 3. Between languages there is a significant difference between when a speech act is performed and what strategy is used to perform it. Borkin and Reinhart (1978), Fraser (1981), Fraser and Nolen (1980), Manes and Wolfson (1981), and Schmidt and Richards (1980) are other examples of specific studies concerned with the acquisition of pragmatic in second languages. This area of research is still too new for well-founded conclusions to be drawn, but what is clear is that pragmatic is such an integral and universal part of language behavior that it must be addressed by language teachers from the beginning stages of classroom teaching and should not be left until later or for outside the classroom. The conventions of pragmatic vary greatly from one language to another and are of great subtlety and complexity, but since much of message design is pragmatically determined, and since setting and interaction provide much of the contextual basis for pragmatic interpretation, the need for specific interfactional activities in language teaching programs is clear. Kramschs handbook (1981) is useful and practical and provides an overview of pragmatic and other

632

TESOL QUARTERLY

discourse phenomena as well as some excellent suggestions for classroom activities to aid learners in acquiring them. Given the complexity and abstractness of pragmatic rules and the dialect and individual variation that accompany their use, it is doubtful that the recommendation of Taylor and Wolfson (1978) that pragmatic rules should simply be taught to students is of any value other than for Krashens Monitor over-users (Krashen 1981:4). Of more value is the position taken by Coulthard that the language teacher cannot hope to explain discoursal meaning in the traditionally regarded cut-anddried way of teaching grammatical rules (1977:xii) and that ways of teaching should shift from teacher-telling to learner interpreting . . . Learners need to become analysts of discourse themselves (Coulthard 1977: xiii).

REPAIR Some of the most interesting work on verbal interaction has been conducted to determine how speakers repair problems that arise in conversation. The early work on repair by native speakers was done by Schegloff, Jefferson, and Sacks (1977), who defined repair as any conversational move which occurs because there is some real or perceived difficulty in the conversation. Repair is a more general phenomenon than correction, but errors in form or meaning are one kind of trouble, and correction is one type of repair. Schegloff, Jefferson, and Sacks refer to repair as a self-righting mechanism for the organization of language use in social interaction (1977:381). Repair in conversation between native or competent speakers is overwhelmingly self-repair, where it is the speaker who attempts to repair the conversational trouble, whether the trouble was self-perceived or perceived by another participant in the interaction (Schegloff, Jefferson, and Sacks 1977:376). Gaskill (1980) has studied correction between native speakers and non-native speakers of English, and Schwartz (1980) has considered repair between pairs of non-native speakers of English. Both have concluded that self-repair and self-correction are far more frequent than other-repair (repair done by a participant other than the speaker), although in Schwartzs non-quantified analysis other-repair was also frequent. Conrads still unpublished thesis (1982) is a detailed study of many of the factors involved in the interaction of native speakers of English with non-natives who are at different levels of language proficiency. Conrad demonstrated that initiation of repair, or calling attention to the need for repair, was done more frequently by the non-native member of a conversational dyad who was at a low level of LANGUAGE RESEARCH AND TEACHING PRINCIPLES 633

language proficiency. The imbalance was reversed at higher levels of proficiency, where it was the native-speaker member who more frequently initiated repair. As for the accomplishment of the repair itself, however, self-repair was carried out by non-native speakers at all levels of ability and increased as ability increased. These facts, plus a number of other observations on context, the relationship of interactants, and the types of moves used in repair, paint a rich and detailed picture of the conversational interaction of native and non-native speakers. If there is a conclusion beyond the instructive ones reached by Conrad himself, it is that language learners have a demonstrated ability to utilize non-language-specific techniques of interaction maintenance which also facilitate their comprehension and, we can assume, their acquisition of the new language. Even relatively non-proficient learners have this ability, and it is most productively used in the context of extended and sincere personal relationships with native speakers. ERROR Probably no aspect of language pedagogy has been the subject of more interest and misunderstanding than that of learner error. Although a great deal of research has been done on the matter of error, basic questions remain as to: 1. 2. 3. 4. the the the the source of learner error characterization and classification of error effects or gravity of learner error treatment of error in the classroom

Source Early work on error asserted that error in second languages resulted largely from differences between the learners first language and the language being learned (i.e., from transfer or interference), or from inadequate or unmastered instruction. Recent studies (summarized in Dulay, Burt, and Krashen 1982, on which this review is based) have established that interference from the learners first language accounts for much less of learner error than had previously been supposed. Until now, studies of the sources of learner error have focused on syntactic error. Phonological error is probably much more the result of first language influence (Dulay, Burt, and Krashen 1982:97-98) as, possibly, are the relatively unstudied areas of semantic, pragmatic, and interfactional error. Studies of the sources of second language error (as summarized in Dulay, Burt, and Krashen 1982:163-172) have identified four types of error. The first of these are developmental errors, errors which are 634 TESOL QUARTERLY

similar to the errors made by children learning the language as their first language. Developmental errors are assumed to be a natural product of gradually developing ability in the new language and, in the studies so far carried out, developmental errors make up the majority of errors exhibited by second language learners. Interlingual errors, or errors clearly attributable to first language influence, make up a maximum of 25% of learner error, with the percentage being higher for adults than for children. The third and fourth types of error are, first, errors that could be either developmental or interlingual (i.e., ambiguous errors) and, last, other errors, which are neither developmental nor interlingual. While researchers have made some progress in identifying the source of learner error, the processes by which these errors are produced are still poorly understood. Transfer and interference a r e terms that are more metaphoric than empirical since very little is known about what actual linguistic or psychological processes learners are employing when errors are produced. Classification Classifying syntactic error according to linguistic type is a more familiar way of categorizing error and is possibly of more immediate use to classroom teachers. In this approach to error, the sources or effects of errors are ignored in favor of classifying the errors according to the part of the linguistic system which is ill-formed (e.g., the phoneme /o/, third person singular verb endings, article omission, inappropriate use of excuse me, and so on). There are several such classifications which have been done, the most useful being that of Burt and Kiparsky (1972). Classification by linguistic type can be a useful analytic procedure and can provide a useful basis for instructional intervention as long as the classification is not mistaken for a psychologically real analysis of the process by which the errors are produced or for a hierarchy of the communicative effect of errors. Effect It has long been recognized that not all errors have the same communicative effect, significance, weight, or gravity, but few studies have been done to measure that significance or to determine the factors affecting it. One study of error significance was done by Burt and Kiparsky (1972) in conjunction with their linguistic classification of syntactic errors. Although their methodology was not rigorous, they established the distinction between global and local error, the former interfering with communication more than the latter. Global errors, LANGUAGE RESEARCH AND TEACHING PRINCIPLES 635

which affect overall sentence organization and confuse the relations among the constituent clauses (1972:6), include disordering of sentence elements, using connectors inappropriately, and regularizing exceptions to syntactic rules. Local errors include errors in noun and verb inflection, articles, and auxiliaries. For example, the globally erroneous sentence Why like we each other? was judged less comprehensible than the locally erroneous Why we like each other? ( D u l a y , Burt, and Krashen 1982:192). Johansson (1978) has produced the most comprehensive study on error gravity, in which he points out that it is not possible to assign specific weights or values to specific errors because factors such as receiver (or listener) characteristics (age, experience, education, and so on), the context of the communication, and the social roles or status of the speakers all play a role in the success or failure of particular communicative acts. In the research reviewed and carried out by Johansson, several clear conclusions emerge, which are summarized in Figure 1.

Johansson makes two other important observations. One is that a learners status as a non-native speaker contributes both to success and to failure in communication. The failure is accentuated by the faultier 636 TESOL QUARTERLY

form of the message produced by the non-native, but because nativespeaker interlocutors work harder to understand such faultier messages, the overall success of the communicative effort may be facilitated (1978:8). The second observation is that topic or content adequacy can override errors in form, and for most native-speaker interlocutors, comprehension of content is the overwhelming consideration (Johansson 1978:17). This last consideration is strongly echoed by Ludwig in her review of research on error significance (1982), in which she examines research on irritation as well as on comprehensibility. The distinction is sometimes confusing, since the two are said to be covariant, but Ludwig says that irritation is the result of the form of the message intruding on the interlocutors perception of the communication (275). In irritation studies, vocabulary errors or errors combining problems in vocabulary and syntax have been found to result in the most irritation (Ludwig 1982:276), but there is a wide range of difference in the results obtained by the studies Ludwig cites. She concludes her review with some implications for teaching, recommending, in general, a pedagogy which calls for a greater use of realistic language-use activities in the classroom. Treatment There have been few controlled studies of the treatment of learner errors in or out of the classroom. Allwright (1975) has demonstrated the inconsistency with which errors are generally treated in the classroom. Methodological recommendations have ranged from correction of all discernible errors to a relatively tolerant attitude toward learner errors. Learners themselves tend to state a preference for a great deal of overt correction (Cathcart and Olsen 1976, Conrad 1982). Hendrickson has done some research into the effectiveness of specific intervention techniques (1976, 1977) and has summarized his work and that of others (1978). He points out that communicatively interfering, stigmatizing, and high frequency errors are the most likely candidates for treatment (1978:342). Most direct methods of intervention, in which a teacher points out and explains or corrects the error, have been shown to have little or no effect on the production of error by learners (1978:393). At present, no methods of intervention have been demonstrated to have a significant effect in decreasing learner error, although both Hendrickson (1978:396) and Ludwig (1982:281-2) have made recommendations. Conrads work, referred to earlier, demonstrates that even low-level learners are capable of using interactionally embedded repair techniques to correct both form and content.

LANGUAGE RESEARCH AND TEACHING PRINCIPLES

637

In summary, error is inevitable in the process of language acquisition. Most of the errors that learners produce result from incomplete but gradually increasing control over a new linguistic system. Other errors probably result from some kind of interference from the learners first language through processes not yet well understood. There are marked differences in the amount of interference that different errors and error types contribute to communication. Phonological errors are probably the most serious and syntactic the least (Johansson 1978:47 and 68), but context and the compounding of error play important roles in the communicative effect of error. While learners often request overt correction of most or all errors, there is evidence that such direct intervention is not very effective. Some feedback on formal adequacy is almost certainly required for acquisition to continue, and there is some evidence to suggest that learners are able to utilize meaning-based, interactionally modulated repair techniques, which acquisition theory suggests are the most useful form of feedback. Affective factors also play a significant role in the persistence of error (i.e., fossilization) and in the effectiveness of various intervention techniques. SOCIAL AND AFFECTIVE FACTORS Following the fundamental work of Gardner and Lambert (1972 and 1973) and Larsen and Smalley (1972), a good deal of speculation, and some research, has focused on the role of attitudes, motivation, and affective factors in language acquisition and learning. Gardner and Lambert established that learners attitudes toward the society or culture connected with the language being learned can strongly affect success in learning the language. Motivation, which is difficult both to define and to measure, may play a role in the success of formal learning more than in informal acquisition (Krashen 1981:28). Gardner and Lamberts claim that integrative motivation (learning so as to become similar or closer to the new culture and society) would be a greater contributor to success in learning a second language than instrumental motivation (learning so as to achieve some goal through a possibly limited use of the new language) has been brought into question by Lukmanis (1972) research. The work on social and affective factors in the last ten years can be viewed through the taxonomy developed by Schumann to categorize the many factors he has identified as playing a role in the success or failure of learners he has studied (J. Schumann 1976 and elsewhere). The categories relevant here are: 1. social factors, such as the relative size of language groups, and social attitudes between groups 638 TESOL QUARTERLY

2. affective factors, such as language and culture shock, and motivation 3. personality factors, such as self-esteem, and sensitivity to rejection. While there are many other factors that affect language acquisition, it is generally agreed that the above three are the major ones. The problem has been to evaluate these factors quantitatively. Oller (1981) attempted to do this for one factor, motivation, and demonstrated that adequate measuring instruments are lacking. J. Schumann (1978) suggests a number of experimental studies, none of which, to our knowledge, has been undertaken. Guiora and his colleagues (Guiora, Brannon, and Dull 1972, Guiora, Paluzny, Beit-Hallahmi, Catford, Cooley, and Dull 1975) have done work on empathy and what they call ego boundaries. They suggest that language teaching make a greater effort to reduce these boundaries to facilitate language learning, and Clarke (1976) provides a useful theoretical perspective on the relationship of language learning to culture shock. There is general agreement within the field that social and affective factors have a major effect on who learns languages and how well they learn them. While the present authors have heard an anecdotal argument that educated learners can overcome attitudinal and affective barriers through the force of intellect, recent diary studies (F. Schumann 1980, F. Schumann and J. Schumann 1977) and the work of Stevick (1976, 1980, 1982) suggest that attitudinal and affective factors play a lesser role in formal learning than in informal acquisition but that, for all language learners, affective factors influence the ability to use new languages spontaneously and effectively. Many recent developments in approach and methodology have been motivated by a recognition of the need to reduce affective barriers to language acquisition, what Krashen (1978) calls the affective filter after Dulay and Burts affective delimiter (1977:99). These developments range from methods such as Suggestopedia (Lozanov 1979) and Counseling-Learning (Curran 1968, 1972, 1976, 1978) to many of the valuable techniques and materials produced by humanistic language teachers (Moskowitz 1978, for example), and to more general suggestions offered by Stevick (1976, 1980, 1982).

CONCLUSIONS Historically, fashions in language teaching have often been based on the prevailing perspectives of linguistics and psychology. It is only recently that research in second language acquisition has developed as an independent field. Just as the oft-repeated mistake of the past was to rush into the classroom with the latest results of linguistic research, LANGUAGE RESEARCH AND TEACHING PRINCIPLES 639

so it would be a mistake now to assume that all recent language-related research has a direct application in the classroom. However, evidence from empirical research and from informal observation strongly suggests that language instruction can be made more effective in the sense that learners can be led to an improved ability to use or apply knowledge of the new language and to an improved ability to continue learning on their own. Rather than go directly to the classroom, we offer here a set of principles, based on the research reviewed above, that can help guide decisions regarding classroom practice. Principle 1. Language instruction which has as its goal functional ability in the new language should give greater emphasis to activities which lead to language acquisition than to activities which lead to formal learning. It is undeniable that instruction in the forms of a new language can result in language learning and, to some extent, language acquisition. But the shortcoming of formal instruction is its relative inefficiency in helping students become effective users of the language outside of instructional settings. If recent research is correctly interpreted, instructional activities and experiences that foster language acquisition can overcome much of this shortcoming. When complemented with formal learning activities, classroom activities leading to language acquisition, we are suggesting, can help learners become more competent users of a new language and more open to acquiring more of the language independently. Traditional classroom learning activities have tended to have correct production and even conscious ability to describe particular language forms or structures as goals. Acquisition activities concentrate on input and interaction. Instructors who are attempting to foster language acquisition must concentrate on providing experiences for the students in which there is a great deal of language input, language which is both comprehensible and meaningful and which is presented in an activity context which encourages attention and involvement. Meaningfulness is facilitated by the reality of the subject matter (Taylor 1982) and by having life rather than language learning as its content (Fanselow 1977). Interaction allows students to modulate the input they receive to increase its comprehensibility and to test the hypotheses they are forming about the new language. It is the opinion of the authors that the reason why more acquisition activities are not being used in instructional programs is that few wellstructured activities have yet been developed. While there are many collections of activities that attempt to make instruction more active 640 TESOL QUARTERLY

and student-centered, or that attempt to increase the quantity of learner production, few instructional materials have been produced which meet the criteria of providing comprehensible input and supporting it with meaningful interaction. Recent work by Krashen and Terrell (1983), Taylor (1983), and the collection of work edited by Johnson and Morrow (1981) is beginning to shape our understanding of what such materials should look like, but there remains a dearth of actual instructional material. Principle 2. Because negative affect, in the form of the affective filter, seems to be a major impediment to success in language acquisition and learning, instruction should make the minimizing of such affective interference one of its primary goals. Along with the goals of providing input and interaction, language instruction must attempt to lower students affective resistance to acquiring the new language. There are two aspects to this problem the methodology of instruction and the content of instruction. Stevick has addressed the methodological issue at length (1976, 1980, 1982), and a valuable review of the role that methodology can play in lowering affective resistance has been provided by Taylor (1983). Specific instructional activities have been developed by Moskowitz (1978), Christison and Bassano (1981), and others. In general, an affectively positive or supportive classroom is one in which students are encouraged to initiate and sustain real communication and to take risks in using the new language in that communication. The encouragement is provided by carefully defining the instructional task or activity, preparing the students for it, and assisting and supporting the students in carrying out the activity by providing resources and feedback as they need them. Principle 3. Language instruction must make greater use of the learners own abilities to acquire language from natural interaction. Language instructors have tended to mistrust or undervalue the role that natural, non-instructional interaction plays in the language acquisition process. Instructors may mistrust natural interaction because they feel it does not provide enough feedback on the accuracy of student production. Instructors may also believe that meaningful interaction is not possible in a second language until a sound basis in the forms of the language has been achieved. But meaningful interaction is crucial if students are to develop functional ability in the new language (Carroll LANGUAGE RESEARCH AND TEACHING PRINCIPLES 641

1967, Saegert, Scott, Perkins, and Tucker 1974), and there is considerable evidence that such experience is possible and beneficial, even at very low levels of second language ability. What research into the structure of interaction, the turn-taking mechanism, and the techniques of repair supports is the view that learners have a predisposition to interact under the proper circumstances (usually not a traditional language classroom), that they are well aware of the general structure of interaction, and that they can make use of the repair mechanisms available to competent speakers to increase the comprehensibility of the input being provided to them and to improve the accuracy of their own production. By structuring interaction so that the settings, the purpose, and the participants roles are as natural as possible, so that the informational content is the focus of the interaction, and so that the interactants are willing and able to provide feedback in the form of repair, natural interaction can be made a beneficial part of a language instruction program. Principle 4. Error produced in the process of acquiring a second language should be viewed as a natural product of the acquisition process, as a source of information on learner strategies, and as a problem best addressed through more input and interaction rather than through correction and drill. T O concentrate on developing students abilities to monitor their production, or to enforce correction while students are engaged in interaction or production, should be regarded as counter-productive. In any behavior, there is a balance between how much is done and how or how well it is done. Traditional language teaching has emphasized the latter at the expense of the former by equating success with accuracy or mastery of form. In language or in communication, success is determined just as much by how rich and flexible ones linguistic resources are. In language learning, there is increasing evidence that accuracy develops more efficiently when it is viewed as one element in a communicative effort, an overall effort which includes meaningfulness, appropriateness, creativity, and reality. To treat accuracy as usage, the socially approved choice of language forms, is less valuable to learners than teaching accuracy as instrumental to the successful accomplishment of a larger communicative task. Some specific recommendations which we believe are called for in the treatment of error are: 1. Teachers should attempt to analyze students behavior to determine which errors are signs of incorrect learning, which errors seriously interfere with comprehensibility, and which errors are inappropriate at the learners stage of development. These errors should be 642 TESOL QUARTERLY

2.

3.

4.

5.

distinguished from those errors which are simply signs of incomplete learning, those which do not interfere with communication, or those which are persistent for all learners. The first types are candidates for treatment, the second probably not. Because language users ability to monitor for formal accuracy is limited in scope and effect, teachers should expect conscious control only over those few features of the language that are subject to monitoring (i.e., a few salient surface features, generally only in the written mode). Because direct methods of error correction (identification, explanation, correction) have not been shown to be effective, teachers should encourage more indirect methods of error improvement, possibly employing peers, and generally assisting and encouraging students to use communicatively and interactionally modulated repair and clarification techniques to improve accuracy. Teachers should be aware that many aspects of error are affectively determined. Pronunciation errors especially, but many other errors as well, serve to mark the learners identity as a non-native speaker, a status that is often psychologically reassuring and interactionally advantageous. Teachers must also realize that they are often unrealistically harsh judges of accuracy (Ludwig 1982:280) and that naive listeners tend to focus on content more than on form. One mistake should be avoided. In Ludwigs words, Too often students and teachers mistakenly relate linguistic accuracy with increased social acceptance, thinking that if they could speak and write with fewer mistakes, then people would like them more . . . However, the data indicate that cultural stereotypes aside, each L2 user is regarded as an individual with his or her own personality and opinion, neither of which is judged on linguistic criteria in the mind of the native speaker (1982:282). Teachers can identify for learners those modes of language use in which formal accuracy is more important (e.g., academic writing, possibly) and teach editing skills and the use of reference works and feedback resources to assist the learners in achieving accuracy in those contexts.

The conclusion to be drawn from this consideration of error is that formal correctness should be accorded a place somewhat lower on the instructional scale than it has traditionally hadnot abandoned, but viewed as secondary to instructional experiences which call for independence and risk taking on the part of students, and greater openness on their part to feedback from others.

LANGUAGE RESEARCH AND TEACHING PRINCIPLES

643

In conclusion, practice in language teaching has characteristically developed from theory or from accumulated classroom experience. Both are excellent sources of direction. What we have attempted to do in this article is to emphasize a third basis for direction and evaluation in language teachingempirical research into language acquisition and language use. In concluding, we return to the comment made earlier that not all of what is discovered about language and language learning is directly applicable to the classroom. By suggesting four general principles to guide language teaching, we hope that evolving practice will help determine and refine empirically sound and coherent teaching practice aimed at developing functional ability in second language learners.

THE AUTHORS Karl J. Krahnke is Assistant Professor of English and Director of ESL Teacher Training at Colorado State University. He has taught and directed ESL programs in Afghanistan, Iran, Utah, and Washington. He is presently coordinating an annotated bibliography of ESL tests and is working on both a reading and a methodology text. Mary Ann Christison, Assistant Professor and Director of the English Training Center at Snow College in Utah, is the author of several reference resource books for ESL teachers. She is also on the TESOL Newsletter editorial staff and edits the Affiliate/ Interest Section page. REFERENCES
Allwright, Richard L. 1975. Problems in the study of the language teachers treatment of student error. In On TESOL 75, Marina K. Burt and Heidi Dulay (Eds.), 96-109. Washington, D. C.: TESOL. Andersen, Roger W. 1976. A functional acquisition hierarchy study in Puerto Rico. Paper presented at the 10th Annual TESOL Convention, New York, March, 1976. Asher, James. 1977. Learning another language through actions: the complete teachers guidebook. Los Gatos, California: Sky Oaks. Asher, James, and R. Garcia. 1969. The optimal age to learn a foreign language. Modern Language Journal 53 (5):334-341. Bailey, Nathalie, Carolyn Madden, and Stephen Krashen. 1974. Is there a natural sequence in adult second language learning? Language Learning 24 (2):235-243. Bialystok, Ellen. 1978. A theoretical model of second language learning. Language Learning 28 (1):69-83.

644

TESOL QUARTERLY

Borkin, Ann, and Susan Reinhart. 1978. Excuse me and Im sorry. T E S O L Quarterly 12 (1):57-70. Brown, Roger. 1973. A first language. Cambridge, Massachusetts: Harvard University Press. Burt, Marina, Heidi Dulay, and Eduardo Hernandez. 1975. The bilingual syntax measure. New York: Harcourt, Brace, Jovanovich, Burt, Marina K., and Carol Kiparsky. 1972. The gooficon: a repair manual for English, Rowley, Massachusetts: Newbury House Publishers, Inc. Carroll, John B. 1967. Foreign language proficiency levels attained by language majors near graduation from college. Foreign Language Annals 1 (2):131-151. Cathcart, Ruth L., and Judy E.W.B. Olsen. 1976. Teachers and students preferences for correction of classroom conversation errors. In On TESOL 76, John F. Fanselow and Ruth Crymes (Eds.), 41-53. Washington, D. C.: TESOL, Christison, Mary Ann, and Sharon Bassano. 1981. Look whos talking. S a n Francisco: Alemany Press. Clarke, Mark. 1976. Second language acquisition as a clash of consciousness. Language Learning 26 (2):377-390. Conrad, Timothy. 1982. Preference for correction in the interaction of native and non-native speakers of English. M.A. thesis, Utah State University. Coulthard, Malcolm. 1977. An introduction to discourse analysis. L o n d o n : Longman. Curran, Charles. 1968. Counseling and psychotherapy: the pursuit of values. Apple River, Illinois: Apple River Press. Curran, Charles. 1972. Counseling-learning: a whole-person model for education. New York: Grune and Stratton. Curran, Charles. 1976. Counseling-learning in second languages. Apple River, Illinois: Apple River Press. Curran, Charles. 1978. Understanding: a necessary ingredient in human belonging. Apple River, Illinois: Apple River Press. de Villiers, Peter A., and Jill G. de Villiers. 1973. A cross-sectional study of the acquisition of grammatical morphemes in child speech. Journal of Psycholinguistic Research 2 (3):267-278, Dulay, Heidi, and Marina Burt. 1974. Natural sequences in child second language acquisition. Language Learning 24 (1):37-53. Dulay, Heidi, and Marina K. Burt. 1977. Remarks on creativity in language acquisition. In Viewpoints on English as a second language, Marina Burt, Heidi Dulay, and Mary Finocchiaro (Eds.), 95-126. New York: Regents Publishing Company, Inc. Dulay, Heidi, Marina Burt, and Stephen Krashen. 1982. Language two. New York: Oxford University Press. Fanselow, John. 1977. Beyond Rashomon conceptualizing and describing the teaching act. TESOL Quarterly 11 (1):17-39. Fraser, Bruce. 1981. On apologizing. In Conversational routine, F l o r i a n Coulmas (Ed,), 259-271, The Hague: Mouton.

LANGUAGE RESEARCH AND TEACHING PRINCIPLES

645

Fraser, Bruce, and W. Nolen. 1980. The association of deference with linguistic form. International Journal of the Sociology of Language 2 7 (1):93-109. Fraser, Bruce, Ellen Rintell, and Joel Walters. 1980. An approach to conducting research on the acquisition of pragmatic competence in a second language. In Discourse analysis in second language research, Diane LarsenFreeman (Ed.), 75-91. Rowley, Massachusetts: Newbury House Publishers, Inc. Gardner, Robert C., and Wallace Lambert. 1972. Attitudes and motivation in second-language learning. Rowley, Massachusetts: Newbury House Publishers, Inc. Gardner, Robert, and Wallace Lambert. 1973. Attitudes and motivation: their role in second language acquisition. In Focus on the learner, John Oller and Jack Richards (Eds.), 235-245. Rowley, Massachusetts: Newbury House Publishers, Inc. Gary, Judith Olmstead. 1974. The effects on children of delayed oral practice in initial stages of second language learning. Ph.D. dissertation, University of California at Los Angeles. Gary, Judith Olmstead. 1975. Delayed oral practice in initial stages of second language learning. In On TESOL 75, Marina Burt and Heidi Dulay (Eds.), 89-95. Washington, D. C.: TESOL. Gaskill, William. 1980. Correction in native speakernon-native speaker conversation. In Discourse analysis in second language research, Diane LarsenFreeman (Ed.), 125-137. Rowley, Massachusetts: Newbury House Publishers, Inc. Gattegno, Caleb. 1972. Teaching foreign languages in schools: the silent way. New York: Educational Solutions. Gattegno, Caleb. 1976. The common sense of teaching foreign languages. New York: Educational Solutions. Goffman, Erving. 1967. Interaction ritual. Garden City: Anchor Books. Goffman, Erving. 1971. Relations in public. New York: Basic Books. Gordon, David, and George Lakoff. 1975. Conversational postulates. In Papers from the seventh regional meeting of the Chicago Linguistic Society. Chicago: Chicago Linguistic Society. Grice, H. Paul. 1975. Logic and conversation. In Speech acts, Peter Cole and Jerry Morgan (Eds.), 41-58. Syntax and Semantics, Volume 3. New York: Academic Press. Guiora, Alexander, Robert C.L. Brannon, and Cecilia Dull. 1972. Empathy and second language learning. Language Learning 22 (1):111-130. Guiora, Alexander, Maria Paluzny, Benjamin Beit-Hallahmi, J. C. Catford, Ralph Cooley, and Cecilia Dull. 1975. Language and person: studies in language behavior. Language Learning 25 (1):43-61. Hakuta, Kenji, and H. Cancino. 1977. Trends in second language acquisition research. Harvard Educational Review 47:294-316. Hatch, Evelyn. 1979. Apply with caution. Studies in S e c o n d L a n g u a g e Acquisition 2 (2):123-143.

646

TESOL QUARTERLY

Hendrickson, James. 1976. The effects of error correction treatments upon adequate and accurate communication in the written compositions of adult learners of English as a second language. Ph.D. dissertation, The Ohio State University. Hendrickson, James. 1977. Error analysis and selective correction in the adult ESL classroom: an experiment. Arlington, Virginia: ERIC, Center for Applied Linguistics. Hendrickson, James. 1978. Error correction in foreign language teaching: recent theory, research, and practice. Modern Language Journal 6 2 (8):387-398. Johnson, Keith, and Keith Morrow. 1981. Communication in the classroom. Essex, England: Longman Group Ltd. Johansson, Stig. 1978. Studies of error gravity. Gotheborg, Sweden: Universitatis Gothoburgensis. Kramsch, Claire. 1981. Discourse analysis and second language teaching. Language in Education: Theory and Practice 37. Washington, D. C.: Center for Applied Linguistics. Krashen, Stephen. 1978. The monitor model for second language acquisition. In Second language acquisition and foreign language teaching, R o s a r i o Gingras (Ed.), 1-26. Arlington, Virginia: Center for Applied Linguistics. Krashen, Stephen. 1981. Second language acquisition and second language learning. Oxford: Pergamon Press. Krashen, Stephen. 1982. Principles and practice in second language acquisition. Oxford: Pergamon Press. Krashen, Stephen, James Butler, Robert Birnbaum, and Judith Robertson. 1978. Two studies in language acquisition and language learning. I T L Review of Applied Linguistics 39-40:73-92. Krashen, Stephen, and Tracy Terrell. 1983. The natural approach: language acquisition in the classroom. San Francisco: Alemany Press. Larsen, Donald, and William Smalley. 1972. Becoming bilingual: a guide to language learning. New Canaan, Connecticut: Practical Anthropology. Larsen-Freeman, Diane. 1975. The acquisition of grammatical morphemes by adult ESL students. TESOL Quarterly 9 (4):409-420. Lee, Richard, Linetta McCune, and Layne Patton. 1970. Physiological responses to different modes of feedback in pronunciation testing. TESOL Quarterly 4 (2):117-122. Lozanov, Georgi. 1979. Suggestology and outlines of suggestopedy. N e w York: Gordon and Breach. Ludwig, Jeanette. 1982. Native speaker judgments of second language learners efforts at communication: a review. Modern Language Journal 66 (3):274-283. Lukmani, Yasmeen. 1972. Motivation to learn and language proficiency. Language Learning 22 (2):261-274. Manes, Joan, and Nessa Wolfson. 1981. The compliment formula. In Conversational routine, Florian Coulmas (Ed.), 115-132. The Hague: Mouton. Moskowitz, Gertrude. 1978. Caring and sharing in the foreign language class. Rowley, Massachusetts: Newbury House Publishers, Inc.

LANGUAGE RESEARCH AND TEACHING PRINCIPLES

647

Oller, John. 1981. Can affect be measured? International Review of Applied Linguistics 19 (3):227-235. Porter, John H. 1977. A cross-sectional study of morpheme acquisition in first language learners. Language Learning 27 (1):47-61. Postovsky, Valerian. 1977. Why not start speaking later? In Viewpoints on English as a second language, Marina Burt, Heidi Dulay, and Mary Finocchiaro (Eds.), 17-26. New York: Regents. Richards, Jack S. 1980. Conversation. TESOL Quarterly 14 (4):413-432. Richards, Jack C., and Ted Rodgers. 1982. Method: approach, design, and procedure. TESOL Quarterly 16 (2):153-168. Rosansky, Ellen. 1976. Methods and morphemes in second language acquisition. Language Learning 26 (2):409-425. Rubin, Joan. 1975. What the good language learner can teach us. T E S O L Quarterly 9 (1):41-51. Sacks, Harvey, Emanuel Schegloff, and Gail Jefferson. 1974. A simplest systematic for the organization of turn-taking for conversation. Language 50 (4):696-735. Saegert, Joel, Margaret Sue Scott, John Perkins, and G. Richard Tucker. 1974. A note on the relationship between English proficiency, years of language study, and medium of instruction. Language Learning 24 (1):99-104. Schmidt, Richard, and Jack C. Richards. 1980. Speech acts and second language learning. Applied Linguistics 1 (2):129-157. Schegloff, Emanuel, Gail Jefferson, and Harvey Sacks. 1977. The preference for self-correction in the organization of repair in conversation. Language 53 (2):361-382. Schumann, Francine. 1980. Diary of a language learner: a further analysis. In Research in second language acquisition, Robin Scarcella and Stephen Krashen (Eds.), 51-57. Rowley, Massachusetts: Newbury House Publishers, Inc. Schumann, John. 1976. Social distance as a factor in second language acquisition. Language Learning 26 (1):135-144. S c h u m a n n , J o h n . 1 9 7 8 . The pidginization process: a model for second language acquisition. Rowley, Massachusetts: Newbury House Publishers, Inc. Schumann, Francine, and John Schumann. 1977. Diary of a language learner: an introspective study of second language learning. In On TESOL 77, H . Douglas Brown, Carlos Yorio, and Ruth Crymes (Eds.), 241-249. Washington, D. C.: TESOL. Schwartz, Joan. 1980. The negotiation for meaning: repair in conversations between second language learners of English. In Discourse analysis in second language research, Diane Larsen-Freeman (Ed. ), 138-153. Rowley, Massachusetts: Newbury House Publishers, Inc. Snow, Catherine E. 1979. Conversations with children. In Language acquisition: studies in first language development, Paul Fletcher and Michael Garman (Eds.), 363-375. Cambridge: Cambridge University Press. Stevick, Earl. 1976. Memory, meaning and method. Rowley, Massachusetts: Newbury House Publishers, Inc.

648

TESOL QUARTERLY

Stevick, Earl. 1980. Teaching languages: a way and ways. Rowley, Massachusetts: Newbury House Publishers, Inc. Stevick, Earl. 1982. Teaching and learning languages. Cambridge: Cambridge University Press. Taylor, Barry P. 1982. In search of real reality. TESOL Quarterly 16 (1) :29-42. Taylor, Barry P. 1983. Teaching ESL: incorporating a communicative, student-centered component. TESOL Quarterly 17 (1):69-88. Taylor, Barry, and Nessa Wolfson. 1978. Breaking down the free conversation myth. TESOL Quarterly 12 (1):31-39. Terrell, Tracy. 1977. A natural approach to second language acquisition and learning. Modern Language Journal 61 (7):325-337. Terrell, Tracy. 1982. The natural approach to language teaching: an update. Modern Language Journal 66 (2):121-132. van Ek, J. A. 1975. The threshold level for modern language learning in schools. London: Longman. Wagner-Gough, Judy, and Evelyn Hatch. 1975. The importance of input data in second language acquisition studies. Language Learning 25 (2):297-308. Winitz, Harris, and James A. Reeds. 1973. Rapid acquisition of a foreign language (German) by the avoidance of speaking. International Review of Applied Linguistics 11 (4):296-317.

LANGUAGE RESEARCH AND TEACHING PRINCIPLES

649

TESOL QUARTERLY, VoI. 17, No.4, December 1983

On the Use of Composition Scoring Techniques, Objective Measures, and Objective Tests to Evaluate ESL Writing Ability
KYLE PERKINS
Southern Illinois University

The evaluation of writing ability among both L1 and L2 students has become increasingly important in recent years because the results of such evaluation are used for a variety of administrative, instructional, and research purposes. Classroom teachers, in particular, because they want to use these results to help improve, influence, refine, and shape their students attained writing ability, have specific concerns regarding the various methods of writing assessment available. Additional concerns, especially of teachers, include the issue of whether the results of any one evaluation procedure are helpful to students and how they affect students writing performance and attitudes. This article does not address the question of how best to use the results of a writing evaluation to improve student writing, but it does present a discussion of the efficacy of various direct and indirect methods of assessing writing ability. The purpose of this article is to outline the assumptions, procedures, and consequences for use of the principal direct scoring schemes and objective measures presently available for evaluating compositions, and of the objective tests currently utilized to determine (indirectly) students ability to write. It is hoped that ESL professionals (whether they are teachers, administrators, or researchers) can make use of this discussion as they try to decide how best to handle questions related to writing ability. Although some readers may question the appropriateness of using indirect measures, such as standardized tests, since those only measure recognition of surface-level syntactic phenomena, the more widely used standardized tests have been included in this survey because they are used by a number of institutions as criterion measures as well as for placement and prediction.

INTRODUCTION There are several scoring schemes in use today for evaluating 651

compositions. In addition to these direct measures of writing proficiency, there are many different objective measures and objective tests which are currently being used to assess indirectly students ability to write. When we consider the various composition scoring schemes, objective measures, and objective tests that are available, several questions immediately come to mind: What do they imply? What specific procedures do they suggest? Are these procedures and tests practical? Are they reliable and valid? Will they work for the different kinds of writing evaluation called for in administrative, instructional, and research decision-making processes? This article presents a discussion of the principal scoring schemes, objective measures, and objective tests currently used to evaluate students ability to write. For each technique and measure four issues are addressed in order to provide answers to the questions listed above. These issues include: 1) what the technique or measure is and how it can be used, 2) what its strengths are, 3) what its limitations are, and 4) an overall evaluation/recommendation and why. COMPOSITION SCORING TECHNIQUES Three composition scoring techniques will be discussed in this section of the article: holistic scoring, analytical scoring, and primary trait scoring. As an introduction, the following definitions were offered by Stiggins and Bridgeford: Holistic scoring calls for the reader to rate overall writing proficiency on a single rating scale. Analytical scoring breaks performance down into component parts (e. g., organization, wording, ideas) for rating on multiple scales. And primary trait scoring requires rating of attributes of performance unique to a particular audience and writing purpose (e. g., persuasiveness, awareness of audience) (1983:26). Holistic Scoring Of all of the composition evaluation schemes available today, holistic scoring has the highest construct validity when overall attained writing proficiency is the construct to be assessed. Holistic scoring involves one or more readers awarding a single grade based on the total impression of a composition as a whole text or discourse. Because holistic scoring evaluates a whole text rather than simply parts of a text, the grader can take any or all of the following into account: 1) whether a thesis has been clearly stated, developed, and supported, and whether an issue has been clearly raised and sufficiently resolved, 2) whether sufficient support and development (i.e., redundancy, i n

652

TESOL QUARTERLY

the communication science sense) have been provided for the reader, 3) whether the writer has tried to accommodate the reader in terms of how much (or how little) the writer thinks the reader knows about the subject (i. e., has the writer tried to accommodate the intended audience?), 4) whether there is grammatical and lexical cohesion and overall coherence, 5) whether one paragraph is hinged smoothly to preceding and succeeding paragraphs through the accomplished use of transitional words, phrases, and structure vocabulary, 6) whether new information has been converted into given information that forms the base for predicating additional new information, and 7) whether the piece makes sense. Although holistic scoring has much to recommend it as a tool for certification, placement, proficiency, and research testing, the procedure nevertheless has many drawbacks, potential and realized weaknessesreferred to collectively as threats to reliability. As previously mentioned, in scoring holistically, the grader reads the composition, forms a general impression, and assigns a mark to that composition based on some standard. That standard may either be a model composition (or series of them) to which the reader has reference, or general impressions the reader has, based on experience in reading student compositions. Such evaluation can, therefore, be highly subjective due to bias, fatigue, internal lack of consistency, previous knowledge of the student, and/or shifting standards from one paper to the next. Published research on holistic scoring in terms of reliability and concurrent validity has yielded contradictory findings. Braddock, Lloyd-Jones, and Schoer (1963) and Follman and Anderson (1967) have reported interrater reliability coefficients as high as 0.90. Similarly, Kaczmarek (1980) found that subjective teacher judgments had substantial reliability estimates in her study and strongly correlated with independent readers and with objective scores. Yet, other research has shown that professional persons, including English teachers, vary in their assessment of attained writing proficiency. The research of Remondino (1959), Diederich et al. (1961), and Diederich (1974) is most noteworthy. Diederich et al. (1961) have reported a study in which sixty professionals were asked to grade 300 papers written by college freshmen from three different schools. The readers, who represented six occupational fields and included college English teachers, social science teachers, natural science teachers, writers and educators, lawyers, and business executives, were asked to sort the 300 papers into nine groups. Diederich noted that out of the 300 essays graded, 101 received every grade from 1 to 9, 94 percent received either seven, eight, or nine different grades; and no essay received less than five different grades from fifty-three readers (1974:5).

EVALUATION OF ESL WRITING ABILITY

653

Diederich analyzed the reasons for the diversity of judgment and isolated the following predominant criteria: 1) flavor and personality (style as the revelation of a personality, individuality, originality, interest, and sincerity), 2) organization and analysis, 3) quality of ideas, 4) usage, sentence structure, punctuation, and 5) wording and spelling. An explanatory inference which explicates Diederichs finding has been offered by Hirsch (1977). Searching for reasons to explain the lack of agreement among readers, Hirsch has suggested that different weights are attached to a few traits of writing and has further noted that reliable, independent agreement in the scoring of writing samples is out of the question until widespread agreement is obtained about the qualities of good writing. Fortunately, there are techniques available to reduce the aforementioned threats to reliability (i.e., reduce the subjectivity) in scoring compositions. For example, Henning (in press) has recommended 1) a behavior-specific rating schedule for use in composition evaluation (for an excellent example of a rating schedule, see Appendix), 2) an insistence on rater competence and expertise, 3) the use of multiple independent raters, and 4) the elicitation of multiple writing samples to control for the fact that attained writing ability may vary with topic and time of day (see Arthur 1979). In a similar fashion, Jacobs et al. (1981) have suggested seven steps for more reliable composition evaluation: 1) adopt a holistic evaluation approach, 2) establish criteria to focus readers attention on significant aspects of the compositions, 3) set a common standard for judging the quality of the writing, 4) select readers from the same backgrounds, 5) train readers until they can achieve close agreement in their assessments of the same papers, 6) obtain at least two independent readings of each composition, and 7) monitor the readers periodically during the evaluation to check their consistency in applying the standards and criteria of evaluation. The previously cited techniques to reduce subjectivity in holistic scoring are necessary for the following reasons cited in Jacobs et al. (1981): 1) graders may have different standards of severity; in fact, some graders may try to use an absolute standard of quality, such as published writing, 2) graders may be inconsistent in applying the standards of the evaluation, 3) graders may react to certain elements in the evaluation or in the papers; for example, Daiker, Kerek, and Morenberg (1978) have found that graders tend to evaluate a composition by the quality of its sentence structure, and Williams (1980) has reported that a composition written in a nominal style tends to be graded higher and to be perceived as better organized, supported, and argued than a corresponding composition written in a verbal style, and 4) graders may value different aspects of a composition. 654 TESOL QUARTERLY

Cooper provides a concise summary statement about holistic evaluation: Holistic evaluation of writing is a guided procedure for sorting or ranking written pieces. The rater takes a piece of writing and either (1) matches it with another piece in a graded series of pieces or (2) scores it for the prominence of certain features important to that kind of writing or (3) assigns it a letter or number grade. The placing, scoring, or grading occurs quickly, impressionistically, after the rater has practiced the procedure with other raters. The rater does not make corrections or revisions in the paper. Holistic evaluation is usually guided by a holistic scoring guide which describes each feature and identifies high, middle, and low quality levels for each feature. Where there is commitment and time to do the work required to achieve reliability of judgment, holistic evaluation of writing remains the most valid and direct means of rank-ordering students by writing ability. Spending no more than two minutes on each paper, raters, guided by . . . holistic scoring guides . . ., can achieve a scoring reliability as high as .90 for individual writers. The scores provide a reliable rank-ordering of writers, an ordering which can then be used to make decisions about placement, special instruction, graduation, or grading (1977:3). The results of such a rank ordering can also provide students with examples of a range of quality for a particular task from previously graded papers as well as with a basis for class discussion.

Analytical Scoring An analytical scoring scheme involves the separation of the various features of a composition into components for scoring purposes. Heaton has described the analytical scoring scheme as follows: Since most teachers have little opportunity to enlist the services of two or more colleagues in marking class compositions, the analytic method is recommended for such purposes. This method depends on a marking scheme which has been carefully drawn up by the examiner or body of examiners. It consists of an attempt to separate the various features of a composition for scoring purposes. Such a procedure is ideally suited to the classroom situation: because certain features have been graded separately, each student is able to see how his particular grade has been obtained. The following is reproduced simply as one example of such an analytical scheme: in this particular case, duplicate (blank) copies of this scheme were stenciled by the teacher and attached to the end of each composition. EVALUATION OF ESL WRITING ABILITY 655

Vocabulary Mechanics Fluency Relevance

I x

TOTAL = 14

x x (Heaton 1975:136-137)

In addition to showing students how their particular grades were determined, analytical scoring has a further advantage: because an analytical scale focuses graders scoring, the procedure is supposed to ensure sufficient agreement among graders to permit a reliable score to be derived from summed multiple ratings. Analytical scores can be used for correlational research, exemption, growth measurement, prediction, placement, and program evaluation uses. And furthermore, analytical scores can serve as helpful guides to feedback on each piece of writing submitted and to formative evaluation which is used to determine the degree of mastery of a given learning task and to pinpoint the part of the task not mastered . . . The purpose is not to grade or certify the learner; it is to help the learner and the teacher focus upon the particular learning necessary for movement toward mastery (Bloom et al. 1971:61). Cooper has commented on the use of analytical scores for two types of writing evaluation: For program evaluation or for research on methods of teaching writing, an analytical scale can serve as a guide to raters choosing the better of each students paired pre- and post-essays on matched topics of the same kind of discourse (OHare 1973; Odell 1976). Where a criterion measure is required in a research study, raters can use an analytical scale to score each students writing (1977:17). Although analytical scales can be used for various purposes in writing evaluation, they are not cost-free. Developing an analytical scale is a time-consuming procedure which, according to Cooper (1977), entails a minimum of six steps. First, a large corpus of published professional and original student writing must be amassed; the features that make up the scale are derived from this corpus. 656 TESOL QUARTERLY

Second, the scale must be field-tested, after which the features may be modified. Next, descriptors must be written to describe the high, mid, and low quality levels for each feature. Fourth, the points must be anchored on a scoring line. Fifth, the graders practice using the scale on a fresh set of compositions. And finally, intra- and interrater reliability estimates are computed. Analytical scales have more serious disadvantages than simply the time and commitment required to do a good job. First, there is the problem of an immoderate standard. As with holistic scoring, some graders may try to use an absolute standard of quality, such as published professional writing. Second, the features to be analyzed are isolated from context and are scored separately. Discourse analysis and good sense tell us that a written or spoken text is more than the sum of its parts. Third, the choice of categories can be vague and certainly arbitrary because the categories themselves are determined by the graders, who base their choices on a corpus of professional and student writing. It has already been pointed out that different readers may value different aspects of a composition. The most serious drawback to analytical scales is that the scoring weight of a particular category must be adjusted for different kinds of discourse, as Cooper explains: The main problem is that they are not sensitive to the variations in purpose, speaker role, and conception of audience which can occur in pieces written in the same mode. Current discourse theory persuades us that a piece of autobiographical incident, viewed as to its function or purpose, could be either expressive, persuasive, referential, or literary, that is, concerned primarily with self, other, subject, or language, While we must acknowledge, then, that analytical scales lack certain kinds of precision, we can still demonstrate their usefulness as general or global guides for responding to a piece of writing (1977 :14).1 In sum, analytical scoring entails a list of the prominent characteristics of written discourse in a particular mode. The number of features to be evaluated is determined by the grader depending on the nature of the evaluation; in essence, graders can tailor the analytical evaluation to fit their particular needs. Each feature is then described in detail, and the high, mid, and low points are anchored and described along a scoring line. Graders learn to use the scale by studying the values and descriptors for each feature. The scale is employed on a
1 Heaton

(1975) has tried to compensate for the lack of precision with regard to proficiency levels. He has recommended that flexibility be maintained in the use of analytical scoring because the grader may need to change either the components or the weighings assigned to them depending on the proficiency level involved. For example, Heaton has suggested that the evaluation might focus on grammar and vocabulary at the elementary level, whereas at the intermediate level the focus might shift to relevance. At the advanced level, the evaluation might focus on organization and subsume mechanics and fluency into one category.

EVALUATION OF ESL WRITING ABILITY

657

batch of student papers, and reliability estimates are calculated for the raters. The results may indicate the necessity for further refinement and practice using the scale. Analytical scores are useful for a variety of evaluation purposes. With enough time and commitment on the part of the graders, reliable scores can be obtained. However, the procedure itself is time-consuming, is vulnerable to the same threats to reliability as holistic scoring, and is of questionable validityit isolates features from context and lacks sensitivity to variations in purpose, speaker role, and conception of audience. Primary Trait Scoring In primary trait scoring, a holistic score is assigned to a particular feature of writing, such as structure, tone, or vocabulary. This procedure requires that graders ascertain whether a piece of writing exhibits certain characteristicstermed primary traits which are crucial to the specific rhetorical task a writer is trying to perform. Lloyd-Jones explains: The goal of Primary Trait Scoring is to define precisely what segment of discourse will be evaluated (e.g., presenting rational persuasion between social equals in a formal situation), and to train readers to render holistic judgments accordingly. The chief steps in using the Primary Trait Scoring System are to define the universe of discourse, to define exercises which sample that universe precisely, to ensure cooperation of the writers, to devise workable scoring guides, and to use the guides (1977:37). It is evident from Lloyd-Jones article that primary trait scoring must be based on a carefully prepared theory of discourse which focuses on aims which relate to the functions of language and features which are the separate elements, devices, and mechanisms of language. The dependence on discourse theory is crucial because, as Lloyd-Jones has stated, judgments about the quality of writing areor should beprimarily related to aims (i. e., does the piece of writing fulfill its purpose?) (1977:33). The decision to use primary trait scoring requires that a grader first choose a model of discourse which identifies the finite types of discourse used in different modes of writing, and then create exercises which students can use to generate writing samples for evaluation. Lloyd-Jones has used a three-part model of discourse, which is reproduced in Figure 1. But other models are available, such as Brittons (1970) two-part model, which focuses on spectator and participant writing, and Kinneavys (1971) four-part model, which focuses on four purposes of discourse: 1) audience-directed, 2) manipulation 658 TESOL QUARTERLY

of language for its own sake, as in literature, 3) self-centered, and 4) subject-controlled. Before exercises are developed, the grader must isolate the kind of discourse to be elicited in the writing sample. The exercises are supposed to stimulate the students to write as well as they can within the narrowly defined type of discourse to be examined. As LloydJones has pointed out, any exercise must be couched within the writers range of knowledge; for example, a writing teacher would not ask a Caracas medical student to write about the Gezira Scheme for raising cotton in the Sudan. The writing exercises must be field-tested to ensure that the students understand what is required of them so that they can generate discourse in the appropriate mode. A scoring guide must then be prepared which, according to LloydJones, contains the following components: A scoring guide consists of (1) the exercise itself, (2) a statement of the primary rhetorical trait of the writing which should be elicited by the exercise (a kind of statement of the limited test objective), (3) an interpretation of the exercise indicating how each element of the stimulus is presumed to affect the respondent (a kind of hypothesis about performance), (4) an interpretation of how the situation of the exercise is related to the posited primary trait (a synthesis of #2 and #3), (5) a system for defining the shorthand which is to be used in reporting descriptions of the writing (the actual scoring guide), (6) samples of papers which have been scored (definition of the score points), and (7) discussions of why each sample paper was scored as it was (extensions of the definitions). EVALUATION OF ESL WRITING ABILITY 659

Most of the guides of persuasive or referential writing use numbers (scores) which indicate a value placed on the observed performances. Usually 4 or 5 is a highly competent performance, 1 is a serious but quite inadequate response; other symbols are used to indicate a failure to engage the exercise (1977:45). Lloyd-Jones (1977:60-66) offers a sample of a final primary trait scoring guide. It is not clear from the previous discussion exactly what traits can be evaluated and precisely what kinds of support a grader looks for while conducting primary trait scoring. Odell (1981) has offered the following insights. If graders were scoring a set of persuasive essays, for example, they would determine 1) the number of reasons the student writers gave in support of their arguments, 2) whether or not the students elaborated on these reasons, and 3) what kind(s) of authorities the students referred to. If graders were scoring a set of expressive essays, on the other hand, they would ascertain 1) whether the student successfully entered into an imaginary role specified in the writing task, and 2) whether they elaborated on that role by giving details. The principal advantage of primary trait scoring is that it provides a precise, detailed description of a students writing ability for a specific rhetorical task. Lloyd-Jones has noted that primary trait data can be used for curriculum evaluation projects, diagnosis, formative evaluation, research and correlational studies, and summative evaluation of student writing; summative evaluation is directed toward a . . . more general assessment of the degree to which the outcomes have been attained over the entire course or some substantial part of it (Bloom et al. 1971:61). Anderson (1981), demonstrating how primary trait scoring can be used to evaluate ESL compositions, has claimed that its advantages for instructors are that it easily enables them to grade specifically on one writing task, it is less time consuming than other methods, and it can be adapted to meet the specific needs of any group of students. Its advantages for the students are that it allows them to focus in on a specific task and to apply structures to a rhetorical context (1981 TESOL Convention Program: 146). Although primary trait scoring has been highly praised by LloydJones, Cooper, and Anderson, there are some serious reservations which must be addressed. Specifically, most writing teachers have become accustomed to evaluating writing samples according to a single set of criteria, but primary trait scoring is a radical departure from that practice. With primary trait scoring a grader may have to use a variety of evaluation procedures to evaluate writing samples in a given mode, and few of these procedures have been developed because discourse theory has not yet provided a complete set of 660 TESOL QUARTERLY

features which are necessary in different writing situations. Furthermore, primary trait scoring is more difficult and more time consuming than other methods: Lloyd-Jones has estimated that 60 to 80 hours of graders time is required for the preparation of each exercise, in addition to the time required to gather samples and conduct trial runs. Perhaps the most efficient use of primary trait scoring has been suggested in the last sentence of the following quote by Cooper: Primary Trait Scoring guides focus the raters attention on just those features of a piece which are relevant to the kind of discourse it is: to the special blend of audience, speaker role, purpose, and subject required by that kind of discourse and by the particular writing task. Furthermore, a unique quality of Primary Trait Scoring is that the scoring guides are constructed for a particular writing task set in a full rhetorical context. This close correspondence of rhetorically explicit task and holistic scoring guide is a new development in holistic evaluation of writing, a development almost certain to have great impact on the evaluation of writing. The correspondence permits a new precision in stating writing research hypotheses and in devising criterion measures for research and curriculum development (1977:11). THE USE OF OBJECTIVE MEASURES AND STANDARDIZED TESTS FOR ASSESSING ESL WRITING ABILITY This section of the article will discuss the use of objective measures of writing which have become quite popular in ESL research and the ubiquitous standardized tests and their application in the evaluation of ESL writing ability. Objective Measures Objective measures such as words per T-unit, clauses per T-unit, and so on, are said to be both objective and reliable because independent evaluators will exhibit high correlations with each other when quantifying objective measures of samples of written work. The T-unit was one of the first objective measures to be employed as an instrument for assessing writing. The T-unit is an index of syntactic complexity that was developed by Hunt (1965), who has defined it as a single main clause (or independent clause) plus whatever other subordinate clauses or nonclauses are attached to, or embedded within, that one main clause (Hunt 1977:93). Objective measures, including the T-unit and its derivations, have become very popular in ESL research (see, for example, Arthur 1979; Evola, Mamer, and Lentz 1980; Flahive and Snow 1980; Gaies 1976, 1980; Homburg 1980; Kaczmarek 1980; Kameen 1979; Larsen-Freeman 1978; Mullen 1980; Nas 1975; Perkins 1980; and Perkins and Homburg 1980). EVALUATION OF ESL WRITING ABILITY 661

While these objective measures may yield impressive estimates of reliability and may function as robust discriminators among proficiency levels, they are not without their critics because the validity of these objective measures for assessing attained writing proficiency remains questionable. First, by universal consent, writing quality has a necessary connection with the communication of meaning; however, objective measures, such as the mean length of T-unit, have no necessary connection with the effective communication of the writers meaning (see Hirsch 1977:179-180). Second, currently used objective measures do not quantify cohesion, coherence, organization (presenting, developing, and supporting a thesis), idiom, diction, tone, relevance, or focusall factors which contribute to good writing. For further criticism, see Ney (1966). Results from studies of objective measures vary from researcher to researcher. For example, Arthur (1979), in a study of short-term changes in EFL composition skills, found that the most significant changes in writing over the duration of a writing course were in speed and vocabulary size. No significant differences were noted in words per T-unit or error-free T-units. However, Kameen (1979), published in the same volume as Arthur, reported that 25 good and 25 poor writers differed in T-unit length, clause length, and incidence of dynamic passive constructions. And Perkins (1980), in a study of ten objective measures of writing proficiency, found that only those measures which took the absence of errors into account discriminated among holistic evaluations by experienced ESL teachers. In a previous study of objective measures, Perkins and Homburg (1980), using conservative statistical procedures, reported that total errors and errors per T-unit discriminated among three different holistic evaluations. Flahive and Snow (1980) employed discriminant analyses to determine which of a set of objective measures had significant robustness in terms of differentiating among different proficiency levels of writing; length of T-unit and clause per T-unit ratio were the only two variables found to be significant. Objective measures have been employed in at least one battery of instruments to measure growth in student writing. Metzger (1978) suggested pre- and post-analyses of writing by using words per T-unit, a count of deviances from standard English, and a measure of writing anxiety. In sum, the use of objective measures is impractical, tedious, and time consuming for classroom use; while objective measures may be of interest to researchers, they, seemingly, are of little value in assessing the underlying constructs of writing because the intent to communicate is neither assessed nor measured by them.

662

TESOL QUARTERLY

Standardized Tests Multiple-choice, objective standardized tests of writing skills are indirect measures, at best, because they do not require the examinee to perform the composing process. Opinions of standardized tests vary from one researcher to another. Michael and Shaeffer, who favor objective tests, have stated the following: A short multiple choice examination requiring knowledge of principles about grammar, usage, choice of words and phrases, and syntax such as the TSWE (Test of Standard Written English) or a subtest of the CSUC-EPT (California State University and Colleges English Placement Test) reflecting similar competencies provides a more accurate prediction of success in a basic English composition course emphasizing writing skills than does an essay examination requiring the actual demonstration of writing skills (1979:132). Diederich also favors objective tests: My final word of advice on such tests [objective tests] is not to despise them. Objective test items can easily, quickly and reliably test a students knowledge of the rules and conventions of English. In writing, if a student is not sure that he knows how to use a certain construction, he can change his sentence to avoid using it, but in an objective test, you can give him a sentence with that construction in it, and he has to decide whether it is correct or incorrect. Arguments over whether answering objective items is the same thing as actual writing or speaking are futile. For that matter, writing one essay is not the same thing as writing another essay, even on the same day; we have seen that even the most carefully determined grades on such essays rarely correlate higher than 0.70. Then the astonishing thing is that scores on a good objective test of English usage often correlate about 0.70 with averages of the two essay grades. It does not matter that they do not really measure the same thing. If students who are good at one tend to be good at the other, and vice versa, then it is a good indicator of proficiency in written English. Call it an editing test if you, like, but I can promise you that students who do well on it tend to be good writers (1974:79-80). For further discussion of the correlation between direct and indirect test scores in L1 composition, see Davis, Scriven, and Thomas (1981); Hogan and Mishler (1980); Moss, Cole, and Khampalikit (1982); and Stiggins (1982). But other researchers, such as Braddock, Lloyd-Jones, and Schoer (1963) and Breland and Gaynor (1979), have criticized objective tests for their lack of validity and the impressions about the low valuation of writing which they convey. The most serious charge against multiple choice tests is their lack of validity. Not only do they not require the examinee to perform the behavior being measuredhe does not do actual writingbut these tests EVALUATION OF ESL WRITING ABILITY 663

also make no attempt to measure the larger elements of composition even indirectly (Braddock, Lloyd-Jones, and Schoer 1963:42). Indirect measures lack face validity and credibility among members of the English profession and educators generally, and they tend to deliver a message to students that writing is not important (Breland and Gaynor 1979:127). The above criticisms notwithstanding, standardized tests are used for a variety of purposes in ESL writing programs. Examples are Sequential Tests of Educational Progress (STEP) English Expression subtest (Educational Testing Service 1956-63); California Achievement Test (CAT) (Tiegs and Clark 1933-72); Stanford Test of Academic Skills (Gardner et al. 1972-74); Cooperative English Test (CET) Effectiveness and Mechanics subtests (Derrick, Harris, and Walker 1940-60); Test of Ability to Subordinate (TAS) (Davidson 1978); Test of Standard Written English (TWSE) (Educational Testing Service 1974); Michigan Test of English Language Proficiency (MTELP) ( U p s h u r et al. 1961-66). Gray and Slaughter (1980) have reported the successful use of some of these tests to place students into three levels of a Basic Writing program, using the following cutoff points for placement purposes:

Kunz (1980) has found that TAS and cloze scores correlate significantly with the MTELP and placement scores. The TAS is a 50-item sentence-combining test which is said to measure specific grammatical points and is recommended for diagnosis and placement purposes. Kunz has reported that at the City University of New York the MTELP and holistic scores are used to place ESL students into different proficiency levels, using the following cutoff points:

664

TESOL QUARTERLY

There has been considerable research conducted with both the TAS and TSWE at Southern Illinois University at Carbondale. Perkins and Konneker (1981) have reported the results of a project which employed five sections of expository writing for foreign students. Two experimental sections received systematic practice in sentence combining; the control sections received none. The two experimental sections and one control section showed significant gain in pre- and post- T A S scores, but no overall improvement in attained writing proficiency. In a validity study, Perkins (1982) found that the TAS had predictive validity for eventual (15 weeks) attained writing proficiency, using a holistic marking scheme as the criterion measure; however, the T A S did not seem to exhibit any concurrent validity, using holistic and analytical schemes as criterion measures. In this case, predictive validity refers to a tests ability to predict learners future writing ability. More specifically, the subjects who received the highest scores on the TAS at the beginning of a 15-week semester tended to receive the highest holistic evaluations on compositions written at the end of the term. Concurrent validity refers to a tests ability to tap the same constructs which other similar tests purport to measure. In this particular study, the TAS which was given at the end of the term did not account for a significant amount of variance in the holistic and analytical scores from compositions which were written during the same testing period. The TSWE is a thirty-minute 50-item multiple-choice test which evaluates a students ability to recognize standard written English. The purpose of TSWE scores is to help place native English speakers into appropriate freshman composition courses. It has been administered to native speaker populations as a part of the Educational Testing Services Scholastic Aptitude Test (SAT). Two types of questions are included in the testthose on usage and those on sentence correction. Students scoring 60 (converted score) are usually exempt from freshman composition; those scoring from 35 to 59, mainstream composition; those scoring below 35, remedial composition. Perkins (1980) found that the TSWE would not discriminate between three different holistic evaluations (e.g., pass, low pass, and fail) of ESL compositions. An apparent implication of that study was that tests designed for native speakers should be used cautiously with L2 populations. Can standardized, objective tests be used for placement purposes in large-scale settings? The answer is yes, if such tests exhibit high magnitudes of concurrent validity with set criteria of placement essay productive writing samples (i.e., students who get high marks on the objective tests also get high marks on the composition placement examination and vice versa). These conditions have apparently been met in the Basic Writing and Intensive ESL programs at the City EVALUATION OF ESL WRITING ABILITY 665

University of New York (see Gray and Slaughter 1980 and Kunz 1980). But it must be remembered that objective tests lack face validity (i.e., the extent to which a test looks like a test) and construct validity (i.e., the extent to which a test measures specific traits in accordance with language learning and behavior theory), at minimum, because no performance of the composing process is elicited from the examinees. Even the TAS does not entail extensive text construction. CONCLUSION ESL composition evaluation consists of two tasks: commenting on student papers and assigning grades. I have not addressed the issue of commentary here because a thorough treatment of a rationale and methodology for commentary would go well beyond the scope of this article. However, this article has been devoted to a discussion of the principal scoring schemes, objective measures, and objective tests currently used in assessing writing ability. In addition to the previous discussion of each technique or measures strengths, weaknesses, and recommended uses, I want to make two additional points: 1 ) no test or scoring procedure is suitable for all purposes, and 2) even with guidelines and set criteria, the analytical and holistic scoring schemes can produce unreliable and invalid test information; for further discussion on this second point, see Jacobs et al. (1981:24-28). In sum, no test or composition scoring procedure is perfect. Perhaps a fitting coda to this article is the following quote from Irmscher: Evaluation obviously implies values, but many teachers evaluate without defining them or just feel frustrated because they cant quantify the value they hold. Without clearly defined values, it is impossible to make consistent judgments and discriminations. And it is better yet if we can verbalize them so that commenting and grading do not seem personal without reference to objective criteria . . . Lacking a set of values [teachers] tend to deal only with particular flaws . . , Not knowing what else to do, teachers proofread instead of reading critically. Thus, error-free writing becomes synonymous in the minds of students with good writing, but of course it isnt necessarily (1979:142-143).

666

TESOL QUARTERLY

ACKNOWLEDGMENTS
This is an expanded version of a paper presented at the 1982 TESOL Convention in Honolulu. Special thanks are due to Barry Taylor for helpful comments and suggestions which, hopefully, were properly incorporated into the final draft.

THE AUTHOR
Kyle Perkins is Associate Professor of Linguistics at Southern Illinois University at Carbondale.

REFERENCES
Anderson, Karen. 1981. Primary trait scoring: new approach to evaluating ESL compositions. Demonstration presented at the 15th Annual TESOL Convention, Detroit, Michigan, March 3-8, 1981. Arthur, Bradford. 1979. Short-term changes in ESL composition. In O n TESOL 79, Carlos Yorio, Kyle Perkins, and Jacquelyn Schachter (Eds.), 330-342. Washington, D. C.: TESOL. Bloom, Benjamin S., J. Thomas Hastings, and George F. Madaus. 1971. Handbook on formative and summative evaluation of student learning. New York: McGraw-Hill. Braddock, Richard, Richard Lloyd-Jones, and Lowell Schoer. 1963. Research in written composition. Urbana, Illinois: National Council of Teachers of English. Breland, Hunter M., and Judith L. Gaynor. 1979. A comparison of direct and indirect assessments of writing skill. Journal of Educational Measurement 16 (2):119-128. Britton, James. 1970. Language and learning. Coral Gables, Florida: University of Miami Press. Cooper, Charles R. 1977. Holistic evaluation of writing. In Evaluating writing: describing, measuring, judging, Charles R. Cooper and Lee Odell (Eds.), 3-31. Urbana, Illinois: National Council of Teachers of English. Daiker, Donald, Andrew Kerek, and Max Morenberg. 1978. Sentence combining and syntactic maturity in freshman English. College Composition and Communication 19 (1):36-41. Davidson, David M. 1978. Test of ability to subordinate. New York: Language Innovations, Inc. Davis, Barbara, Michael Scriven, and Susan Thomas. 1981. The evaluation of composition in instruction. Pt. Reyes, California: Edgepress. Derrick, Clarence, David P. Harris, and Biron Walker (Revisors). 1940-60. Cooperative English tests. Princeton, New Jersey: Cooperative Tests and Services, Educational Testing Service. Diederich, Paul. 1974. Measuring growth in English. Urbana, Illinois: National Council of Teachers of English.

EVALUATION OF ESL WRITING ABILITY

667

Diederich, Paul, John French, and Sydell Carlton. 1961. Factors in judgments of writing ability. Educational Testing Service Research Bulletin R B - 6 1 - 1 5 . Princeton, New Jersey: Educational Testing Service. Educational Testing Service. 1956-63. Sequential tests of educational progress. Princeton, New Jersey: Cooperative Test Division, Educational Testing Service. Educational Testing Service. 1974. Test of standard written English. Princeton, New Jersey: Educational Testing Service. Evola, Jill, Ellen Mamer, and Becky Lentz. 1980. Discrete point versus global scoring for cohesive devices. In Research in language testing, John W. Oller, Jr. and Kyle Perkins (Eds.), 177-181. Rowley, Massachusetts: Newbury House. Flahive, Douglas, and Becky Snow. 1980. Measures of syntactic complexity in evaluating ESL compositions. In Research in language testing, John W. Oller, Jr. and Kyle Perkins (Eds.), 171-176. Rowley, Massachusetts: Newbury House. Follman, John C., and James A. Anderson. 1967. An investigation of the reliability of five procedures for grading English themes. Research in the Teaching of English 1 (2):190-200. Gaies, Stephen. 1976. Sentence-combining: a technique for assessing proficiency in a second language. Paper presented at the Conference on Perspectives in Language, University of Louisville, May 6-8, 1976. Gaies, Stephen. 1980. T-unit analysis in second language research: applications, problems, and limitations. TESOL Quarterly 14 (2):85-98. Gardner, Eric F., Robert Callis, Jack C. Merwin, and Richard Madden. 197274. Stanford test of academic skills. New York: Harcourt Brace Jovanovich. Gray, Barbara Quint, and Virginia B. Slaughter. 1980. Writing. In Teaching basic skills in college, Alice Stewart Trillin and Associates (Eds.), 12-90. San Francisco: Jossey-Bass Publishers. Heaton, J. B. 1975. Writing English language tests. London: Longman Group, Ltd. Henning, Grant H. In press. Language test development. Rowley, Massachusetts: Newbury House. Hirsch, E. D., Jr. 1977. T h e p h i l o s o p h y o f c o m p o s i t i o n . C h i c a g o : T h e University of Chicago Press. Hogan, Thomas P., and Carol Mishler. 1980. Relationships between essay tests and objective tests of language skills for elementary school students. Journal of Educational Measurement 17:219-227. Homburg, Taco J. 1980. A syntactic complexity measure of attained ESL writing proficiency. M.A. thesis, Southern Illinois University. Hunt, Kellogg W. 1965. Grammatical structures written at three grade-levels. Urbana, Illinois: National Council of Teachers of English. Hunt, Kellogg W. 1977. Early blooming and late blooming syntactic structures. In Evaluating writing: describing, measuring, judging, Charles R. Cooper and Lee Odell (Eds.), 91-104. Urbana, Illinois: National Council of Teachers of English. Irmscher, William F. 1979. Teaching expository writing. New York: Holt, Rinehart and Winston.

668

TESOL QUARTERLY

Jacobs, Holly L., Stephen A. Zingraf, Deanna R. Wormuth, V. Faye Hartfiel, and Jane B. Hughey. 1981. Testing ESL composition: a practical approach. Rowley, Massachusetts: Newbury House Publishers, Inc. Kaczmarek, Celeste M. 1980. Scoring and rating essay talks. In Research in language testing, John W. Oller, Jr. and Kyle Perkins (Eds.), 151-159. Rowley, Massachusetts: Newbury House Publishers, Inc. Kameen, Patrick. 1979. Syntactic skill and ESL writing quality. In On TESOL 79, Carlos Yorio, Kyle Perkins, and Jacquelyn Schachter (Eds.), 343-350. Washington, D. C.: TESOL. Kinneavy, James. 1971. A theory of discourse. Englewood Cliffs, New Jersey: Prentice-Hall. Kunz, Linda Ann. 1980. English as a second language. In Teaching basic skills in college, Alice Stewart Trillin and Associates (Eds.), 145-207. San Francisco: Jossey-Bass Publishers. Larsen-Freeman, Diane. 1978. An ESL index of development. Paper presented at the 12th Annual TESOL Convention, Mexico City, April 4-9, 1978. Lloyd-Jones, Richard. 1977. Primary trait scoring. In Evaluating writing: describing, measuring, judging, Charles R. Cooper and Lee Odell (Eds.), 33-66. Urbana, Illinois: National Council of Teachers of English. Metzger, Elizabeth. 1978. A scheme for measuring growth in college writing. Journal of Basic Writing 1 (4):71-81. Michael, William B., and Phyllis Shaeffer. 1979. A comparison of the validity of the Test of Standard Written English (TSWE) and of the California State U n i v e r s i t y a n d C o l l e g e s E n g l i s h P l a c e m e n t T e s t ( C S U C - E P T ) in the prediction of grades in a basic English Composition course and of overall freshman-year grade point average. Educational and Psychological Measurement 39 (1):131-145. Moss, Pamela A., Nancy S. Cole, and Choosak Khampalikit. 1982. A comparison of procedures to assess written language skills in grades 4, 7, and 10. Journal of Educational Measurement 19 (1):37-48. Mullen, Karen. 1980. Evaluating writing proficiency in ESL. In Research in language testing, John W. Oller, Jr: and Kyle Perkins (Eds.), 160-170. Rowley, Massachusetts: Newbury House. Nas, G. 1975. Determining the communicative value of written discourse produced by L2 learners. Utrecht, The Netherlands: Institute of Applied Linguistics, Ney, James W. 1966. Review of Grammatical structures written at three grade levels, Kellogg Hunt. Language Learning 16 (3 and 4):230-235. Odell, Lee. 1976. The classroom teacher as researcher. English Journal 6 5 (1):106-111. Odell, Lee. 1981. Teachers of composition and needed research in discourse theory. In The writing teachers sourcebook, Gary Tate and Edward P. J. Corbett (Eds.), 53-61. New York: Oxford University Press. OHare, Frank. 1973. Sentence-combining: improving student writing without formal grammar instruction. NCTE Research Report No. 15. U r b a n a , Illinois: National Council of Teachers of English. Perkins, Kyle. 1980. Using objective methods of attained writing proficiency to discriminate among holistic evaluations. TESOL Quarterly 14 (1):61-69.

EVALUATION OF ESL WRITING ABILITY

669

Perkins, Kyle, and Taco Homburg. 1980. Three different statistical analyses of objective measures of attained writing proficiency. In Occasional Papers on Linguistics No. 6, Proceedings of the Third International Conference on Frontiers in Language Proficiency and Dominance Testing, R a y m o n d Silverstein (Ed.), 306-317. Carbondale, Illinois: Southern Illinois University, Department of Linguistics. Perkins, Kyle, and Beverly H. Konneker. 1981. Testing the effectiveness of sentence-combining in ESL composition. Paper presented at the 15th Annual TESOL Convention, Detroit, Michigan, March 3-8, 1981. Perkins, Kyle. 1982. The test of ability to subordinate: predictive and concurrent validity for attained ESL composition. In Proceedings of the Fourth International Language Testing Symposium of the Interuniversitare Sprachtestgruppe held at the University of Essex, 14-17 September, 1981. Occasional Papers 26, Terry Culhane, Christine Klein Braley, and Douglas K. Stevenson (Eds.), 104-112. Colchester: University of Essex, Department of Language and Linguistics. Remondino, C. 1959. A factorial analysis of the evaluation of scholastic compositions in the mother tongue. British Journal of Educational Psychology 29 (1):242-251. Stiggins, Richard J. 1982. A comparison of direct and indirect writing assessment methods. Research in the Teaching of English 16 (2):101-114. Stiggins, Richard J., and Nancy J. Bridgeford. 1983. An analysis of published tests of writing proficiency. Educational Measurement: Issues and Practices 2 (1):6-19, 26. Tiegs, Ernest W., and Willis W. Clark. 1933-72. California achievement tests: language. Del Monte Research Park, Monterey, California: CTB/ McGraw-Hill. Upshur, John A., et al. 1961-66. Michigan test of English language proficiency (Forms A-E). Ann Arbor, Michigan: University of Michigan, Testing and Certification Division, English Language Institute. Williams, Joseph M. 1980. Non-linguistic linguistics and the teaching of style. Language and Style XIII (3):24-40. APPENDIX HOLISTIC CRITERIA City University of New York Freshman Wills Assessment Program Evaluation Scale for Writing Assessment Test 6: The essay is completely organized and the ideas are expressed in appropriate language. A sense of pattern or development is present from beginning to end. The writer supports assertions with explanation or illustrations. Sentences reflect a command of syntax within the ordinary range of standard written English. Grammar, punctuation, and spelling are generally correct.

670

TESOL QUARTERLY

4-5: The writer introduces some point or idea and demonstrates an awareness that development or illustration is called for. The essay presents a discernible pattern or organization, even if there are occasional digressions. The essay demonstrates sufficient command of vocabulary to convey, without serious distortion or excessive simplification, the range of the writers ideas. Sentences reflect a sufficient command of syntax to ensure reasonable clarity of expression. The writer generally avoids both the monotony of rudimentary syntax and the incoherence created by tangled syntax. The writer demonstrates through punctuation an understanding of the boundaries of the sentence. The writer spells the common words of the language with a reasonable degree of accuracy. Exceptions can be made for the so-called spelling demons which frequently trouble even an advanced writer. The writer shows the ability to use regularly, but not necessarily faultlessly, the common forms of agreement and of grammatical inflection in standard written English. 2-3: An idea or point is suggested, but is underdeveloped or presented in a purely repetitious way. The pattern of the essay is somewhat random and relationships between sentences and paragraphs are rarely signaled. The essay is restricted to a very narrow range of language, so that the vocabulary chosen frequently does not serve the needs of the writer. The syntax of the essay is not sufficiently stable to ensure reasonable clarity of expression. The syntax is rudimentary or tangled. The writer frequently commits errors of punctuation which obscure sentence boundaries. The writer spells the common words of the language with only intermittent accuracy. The essay reveals recurrent grammatical problems; if there are only occasional problems, this may be due to the extremely narrow range of syntactical choices the writer has used. 1: The essay suffers from general incoherence and has no discernible pattern of organization. It displays a high frequency of error in the regular features of standard written English. Lapses in punctuation, spelling and grammar often frustrate the reader. Or: The essay is so brief that any reasonably accurate judgment of the writers competence is impossible. (Gray and Slaughter 1980:25-26)

EVALUATION OF ESL WRITING ABILITY

671

REVIEWS
The TESOL Quarterly welcomes evaluative reviews of publications of relevance to TESOL professionals. In addition to textbooks and reference materials, these include computer and video software, testing instruments, and other forms of non-print materials.

Edited by CHARLENE J. SATO University of Hawaii at Manoa

Language and Culture in Conflict: Problem-Posing in the ESL Classroom Nina Wallerstein. Reading, Massachusetts: Addison-Wesley Publishing Company, 1983. Pp. 209. This resource book for teachers of adult ESL classes is an outgrowth of author Nina Wallersteins experience in community education. As such, it offers a class-tested model for those who wish to adapt Paulo Freires problem-posing education (Freire 1973) to their own classrooms. Part One of Wallersteins two-part book defines problemposing education and describes appropriate problem-posing teaching techniques. Part Two compiles eight teaching units applying the techniques introduced in Part One. While Freires writings (e.g., 1970a, 1970b) have been criticized (Boston 1972, Donohue 1972, Elias 1974, Friedenberg 1971, Gleeson 1974, Jerez and Hernndez-Pico, undated) for being stylistically obtuse and clouded by specialized vocabulary, Wallersteins presentation in Part One is remarkably clear, direct, and simple. By providing specific and concrete suggestions throughout this section, she clarifies the sometimes elusive transition from problem-posing theory to practice. Freires problem-posing education proposes that true education must be rooted in issues of direct relevance to learners. The teacher identifies these issues by listening to the students. Through a method of interaction called dialogue, this approach involves learners first in reflection on those critical issues and then leads them to action, with the resulting effect of improving their lives. Wallerstein has explored and added to the dialogue stage (Freire 1970b) of problem-posing education by identifying a five-step teaching process which is intended to promote the development of critical thinking through dialogue extending into action. The teachers questions described in Part One (20-24) guide the learning group to 1) describe the content and feelings contained in a code (a representation; e.g., a dialogue, story, picture, 673

or song relating to an issue in the lives of the students); 2) define the problem represented by the code; 3) determine the presence of the problem in the students own lives; 4) relate the problem to the larger historical, cultural, social, and political context; and 5) identify actions the students themselves can take to resolve the problem. Part Two of the book contains eight units, of eight or nine lessons each, on the following themes: autobiography, family, culture and conflict, neighborhood, immigration, health, work, and money. Each lesson includes an initial code in the form of a dialogue, story, or picture, and some combination of the following instructional approaches: 1) a list of suggested questions for moving through the five steps of dialogue, 2) grammar exercises, 3) a vocabulary list, 4) group conversation exercises, 5) writing exercises, and 6) suggestions for additional activities. The questions Wallerstein proposes for the teachers use in stimulating dialogue may seem overly simple. A transcript of an actual class participating in one lesson of a unit, however, could have demonstrated the excitement, energy, and communicative fluidity that these questions can generate in a classroom. As it is, the reader is left to imagine this interaction. This reviewer anticipates that, in practice, the most powerful questions at the fifth step of dialogue, the stage of action, will be those that arise from the discussion of neighborhood concerns. Since the stage of action is the one in which North American educators are least practiced, teachers might want to concentrate on the neighborhood unit for some concrete examples of how to move into that stage. A problem-posing approach to language learning necessitates teaching the vocabulary of emotions and actions as well as natural conversation patterns. Situational curricula, according to Wallerstein, teach survival coping skills but do not deal with feelings, conflicts, and possibilities for changing situations (39-42). Wallerstein attempts to fulfill these requirements by identifying the language functions to be addressed in each unit. Similar applications of the notional-functional syllabus have already appeared in many North American ESL texts. Wallersteins particular contribution lies in her integration of problemposing education with this current development in language teaching. However, this reviewer does question Wallersteins inclusion of the notional-functional concept in a chart (26) that compares problemposing with other curricular approaches. Problem-posing is a comprehensive approach to curriculum design. The notional-functional syllabus, on the other hand, is only a method of content definition which can be integrated into any number of curriculum designs, including problem-posing. Therefore, a comparison of problem-posing and a notional-functional syllabus is not entirely appropriate and is less

674

TESOL QUARTERLY

useful than would be a discussion of the utilization of the notionalfunctional syllabus within a problem-posing curriculum design. Wallersteins experiential basis for this book lends it a strong measure of realism. She is realistic about her students as language learners, evidenced by her assertion that the use of problem-posing does not obviate concurrent uses of more systematic language learning methodologies. She is also sensitive to her students and highlights areas where special consideration may be warranted (e. g., immigration status). Her generous provision of examples to illustrate concepts demonstrates the feasibility of conducting problem-posing ESL, as well as Adult Basic Education and General Education Diploma, classes. Although this book is a powerful model, it is only one model of how problem-posing can be implemented; other problem-posing designs are also possible. For example, another model might provide only pictorial codes and have the students produce accompanying dialogues or stories which, in turn, would determine the linguistic content to be learned. Furthermore, it should be noted that Wallerstein has not intended to supply a complete, fully developed text that is transferable to any given ESL classroom. The unit themes are sufficiently universal to furnish a basis for initial dialogue. The codes, though, may well have to be reconstructed to fit the needs and cultures of particular classes. In any case, in a problem-posing classroom, the students would soon be suggesting their own themes for study, In order to expand and adapt Wallersteins model effectively, while at the same time maintaining the problem-posing posture, teachers should provide themselves with a broad theoretical foundation in problem-posing education. At a minimum, they should read Freires (1973) essay, Education as the Practice of Freedom. With this background, the model which Wallerstein introduces can be helpful not only to teachers of adult ESL classes but also to the secondary school ESL and foreign language teacher. Teachers at this level have, in the past, been deterred from implementing problem-posing because of the heterogeneity of their classes and the difficulty of moving into the action stage. To address these problems, Wallerstein proposes using the classroom situation as the common experience from which to generate dialogue and using the school setting as the initial setting for action. With this core of common experience and setting for action, and with Wallersteins model as a guide, progress in adapting problemposing to a variety of levels and subject areas can be anticipated.

REVIEWS

675

REFERENCES Boston, Bruce O. 1972. Paulo Freire: notes of a loving critic. In Paulo Freire: a revolutionary dilemma for the adult educator, Stanley M. Grabowski (Ed.), 83-92. Occasional Papers Number 32. Syracuse: Syracuse University Publications in Continuing Education and ERIC Clearinghouse on Adult Education. Donohue, John W. 1972. Paulo Freire: philosopher of adult education. America 127(6): 167-170. Elias, John L. 1974. A comparison and critical evaluation of the social and educational thought of Paulo Freire and Ivan Illich: with a particular emphasis upon the religious inspiration of their thought. Ed. D. dissertation, Temple University. Freire, Paulo. 1970a. Cultural action for freedom. Monograph Series Number 1. Cambridge, Massachusetts: Harvard Educational Review and Center for the Study of Development and Social Change. Freire, Paulo. 1970b. Pedagogy of the oppressed. New York: Herder and Herder. Freire, Paulo. 1973. Education as the practice of freedom. In Education for critical consciousness, Paulo Freire, 1-84. New York: The Seabury Press. Friedenberg, Edgar A. 1971. Review of Pedagogy of the oppressed, P a o l o (sic) Freire. Comparative Education Review 15(3):378-380. Gleeson, Denis. 1974. Theory and practice in the sociology of Paulo Freire. Universities Quarterly 28(3):362-371. Jerez, Czar, and Juan Hernndez-Pico. Undated. Cultural action for freedom. In Paulo Freire, 29-51. The LADOC Keyhole Series Number 1. Washington, D. C.: Division for Latin America-USCC. LINDA M. CRAWFORD-LANGE Jackson .lunior High School Champlin, Minnesota

Essential Idioms in EnglishA New Revised Edition Robert J. Dixon. New York: Regents Publishing Company, 1983. Pp. xii + 196. True to its title, Essential Idioms in English is a collection of very useful idiomatic expressions. As stated in the Foreword, the idioms for study should have practical value and be well within the students grasp. Such expressions as to carry coals to Newcastle . . . . w h i l e v e r y colorful, do not help the student achieve the goal of gaining fluency. An unusual aspect of the text is a listing of Spanish, French, and German equivalents of the idioms covered. Unfortunately, there is little else positive about Essential Idioms in

676

TESOL QUARTERLY

English. When the new 1983 edition is compared with the revised 1971 edition (the original was published in 1951), it is readily apparent that very little has been changed aside from the layout. The new cover is attractive, and the print is large. Old-fashioned illustrations (a school bus, a flat tire, a dress shop) have been replaced with New Yorker magazine cartoons, which are attractive, but which appear to have a purely cosmetic purpose. Each chapter begins with a list of twelve basically unrelated idioms, each followed by a short definition. Two or three sentences containing the idioms follow. There is one cartoon per lesson, with the caption containing only one of the twelve idioms. Nowhere in the text is there any structured exploitation of these cartoons, nor is there any discussion of the cultural aspects of the situations depicted. Since 1951 there has been considerable research in the areas of second language acquisition, sociolinguistics, and communicative language teaching, and the notion of contextualization in language teaching has almost become a rule. Nevertheless, the presentation of idioms in this text is completely uncontextualized. This is, of course, not surprising, considering that the material was essentially written over thirty years ago. As the Preface of the 1971 edition states, Finally, it should be borne in mind that this is a drill book designed to teach students how to use idioms. The Foreword of the 1983 edition at least omits this reference to drills, a word with more negative than positive connotations in modern language teaching. The fact remains, however, that the new version, like its predecessors, is essentially a drill book. In each lesson students are given two exercises: the first is a multiplechoice or substitution exercise, and the second is a list of questions containing the twelve idioms. There is no attempt to have the students use the idioms to communicate; in fact, an exercise in writing dialogues or role playing would be quite difficult because of the lack of a relationship among the idioms in each list. The sentences following each idiom leave it to the student (and the teacher) to determine how each expression is used grammatically. We can induce from Hal goes in for tennis while his wife goes in for painting (90) that to go in for is followed by either a common noun or a gerund. But from the examples We are looking forward to Christmas and . . . she has nothing to look forward to except . . . (53), we may not realize that to here is a preposition (not the infinitive to) which is always followed by a gerund or other nominal form. We are shown how the two-word verb bring up is separable (bring him up, bring the matter up, bring her children up), but we are not shown any sentences containing, for example, bring up the matter. Granted that an idiom text is not a grammar text, but for students to use (and not merely recognize) the idioms, their knowledge of the grammatical usage of REVIEWS 677

each expression must exceed simply knowing whether or not an idiom is separable. It might appear unfair to judge an old text by new standards, but Regents Publishing Company has opened itself to criticism by releasing this text. The Foreword states that Essential Idioms in English is . . . the first comprehensive text to attempt to teach idioms by means of extensive practice exercises. The 1983 edition is by no means the first comprehensive idioms text. And more importantly, the practice exercises are less than extensive. The addition of wonderful (but unutilized) cartoons and the replacement of out-dated expressions with a few new ones (e. g., to be into something) do not make this a new idioms textbook. This review of Essential Idioms in English is a criticism not solely of the author, but also of a publishing company which has chosen to publish old and questionable material and present it as new in 1983. HELEN K. FRAGIADAKIS University of California Extension, Berkeley

678

TESOL QUARTERLY

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 cent act the authors for more details.
Edited by ANN FATHMAN

Caada College

New Methodologies for Studying the Relationsbip of Bilingualism and Cognitive Fexibility
KENJI HAKUTA Yale University

The turn of the century saw the introduction of psychometric tests of intelligence to the United States, mostly due to the efforts of H. Goddard and L. Terman. Both of these psychologists were hereditarians, who believed that the genetic composition of individuals was the primary determinant of performance on these tests. A major thrust of work in those dark days was directed toward showing that individuals from certain racial groups (primarily Southern and Eastern Europeans) were of inferior intelligence; together with the assumption that this was due to their genes, certain social policies such as the restriction of immigration were effected (see Gould 1981). There was, however, a major stumbling block in reaching these conclusions from the data that existed (mostly results of massive IQ testing of World War I recruits): many of the testees were not native speakers of English, yet they were tested in English. Thus, whether these individuals were suffering from a language handicap in testing became a major issue. Hereditarians, such as Termans student Kimball Young, argued that there was no language handicap. Those in favor of environmental determinants of IQ stressed the language handicap. The social context in which this early research on bilingualism and intelligence was conducted is radically different from the context of the more recent studies of the matter, basically modeled after Peal and Lamberts (1962) well-known study. These different contexts are primarily responsible for the differences in the focus of the research and the methodologies adopted. From our current perspective, the methodology of the earlier research was wildly deficient, failing to control for a variety of obvious variables such as socio-economic status (SES). But the earlier researchers assumed that low SES was the result, not the cause, of inferior performance on IQ tests, so this questionable methodology passed the inspection lines of science (a review of this history will appear in Hakuta, forthcoming). We contemporary researchers are not, however, so objective either. I think 679

it would not be too far from the truth to say that most researchers engaged in the problems of bilingualism today have a favorable attitude toward it. Thus, the Peal and Lambert findings that bilingualism seems to have a positive effect on performance on intelligence tests (we will hereafter call it cognitioe ability) are consonant with our beliefs. But there are actually some profound methodological difficulties in the standard design, some of which were pointed out quite early by John Macnamara (1966). More recently, MacNab (1979) has forcefully criticized the inferences drawn from the studies, given their methodology (also Hakuta and Diaz [in press] develop an independent but similar line of argument). The difficulties need to be addressed, even if the results agree with our biases. So what is this methodology that is so questionable, and what are the problems? The methodology is simple. Like the early line of research, a group of bilingual and a group of monolingual are compared. However, the bilingual group is defined according to various criteria as those who are equally proficient in both languages. To obtain the results, the two groups are simply equated on SES and a few other relevant variables. Usually, the bilingual outperform the monolingual (for recent reviews, see Cummins 1976, Diaz 1983). What are the difficulties? For one thing, these studies are between-group comparisons, and one should always be concerned about whether the differences that are found are indeed because of the supposed treatment variable, which is bilingualism, or whether they are due to some other unsuspected factor that had not been controlled for. And since there are in fact a potentially infinite number of dimensions along which two groups can differ, if indeed a difference is found, it is not always possible to know exactly how to interpret it. If our work could be performed in the ideal experimental laboratory, we would randomly assign subjects to either a bilingual group or a monolingual group, but that is not the way bilingualism distributes itself in the real world. A second concern is the problem of cause and effect. With a few exceptions, these studies observed children at only one point in time, rather than longitudinally. So, the inference on the direction of causality that was being drawn (i.e., that bilingualism causes cognitive flexibility) was unsupported. It could just as easily be the other way around, that children who are more cognitively flexible become bilingual. Indeed, the results from the longitudinal studies that are available are ambiguous (see review by MacNab 1979) . A third concern is one of experimenter bias, a well-documented problem in behavioral research. Essentially, it is not too difficult for subjects in research studies to perceive the intentions and biases of the experimenter, and to act accordingly. Thus the experimenter, even unknowingly, can influence the results in his or her desired direction. This problem can be attenuated to a large degree by the use of a blind procedure, where the experimenter is kept ignorant of the group to which the subject belongs. Such has not been the case, apparently, in previous research. These problems can all be overcome, or at least minimized. For one thing, the problem of comparability can be solved by looking within a group of 680 TESOL QUARTERLY

bilingual who vary in their degree of bilingualism. The cause-effect problem can be illuminated by studying subjects longitudinally. And the experimenter bias problem is minimized if blind procedures are used in a within-group comparison. Our study (funded by the National Institute of Education, in collaboration with Aida Comulada, New Haven Public Schools) conforms to these three characteristics. We are studying several cohorts of low SES Puerto Rican elementary school children, roughly 300 in all, in the bilingual program in New Haven, over a period of two to three years. We have measures of their Spanish and English vocabularies (validated against other language measures), and several indicators of cognitive ability, including the Ravens Progressive Matrices Test, and a test of judgment of grammaticality in their dominant language, Spanish. Children are repeatedly tested. The results are still best considered preliminary (reported in Hakuta and Diaz, in press) since we are just completing the final year of data collection. However, they are encouraging. Using multivariate statistical analyses, we have shown that degree of bilingualism is positively related to performance on a variety of cognitive measures. For example, the correlation between our measure of degree of bilingualism and the Ravens test, controlling for extraneous variables, ranges between .20 and .40. Furthermore, there are preliminary indications (although anyone familiar with the problems of analyzing data should sense the warranted caution) that the direction of causality best supported is from bilingualism to cognitive ability. REFERENCES
Cummins, James. 1976. The influence of bilingualism on cognitive growth: a synthesis of research findings and explanatory hypotheses. Working Papers on Bilingualism 9:1-43. Diaz, Rafael. 1983. Thought and two languages: the impact of bilingualism on cognitive development. Review of Research in Education 10:23-54. Gould, S. J. 1981. The mismeasure of man. New York: Norton. Hakuta, Kenji. Forthcoming. Bilingualism in the individual and in society. New York: Basic Books. Hakuta, Kenji, and Rafael Diaz. In press. The relationship between degree of bilingualism and cognitive ability: some longitudinal data. In Childrens language, Volume 5, K. E. Nelson (Ed.). Hillsdale, N. J.: Erlbaum. Macnamara, John. 1966. Bilingualism and primary education. Edinburgh: University Press. MacNab, G. L. 1979. Cognition and bilingualism: a reanalysis of studies. Linguistics 17:231-255. Peal, E., and W. Lambert. 1962. The relation of bilingualism to intelligence. Psychological Monographs 76 (546):1-23.

Authors Address: Department of Psychology, Yale University, Box 11A Yale Station New Haven, CT 06520-7447.

BRIEF REPORTS AND SUMMARIES

681

Unfamiliar Instructional Practice and Its Relationship to Student Emotional Distress in the ESL Classroom
SHARRON BASSANO Santa Cruz Adult School It is generally recognized that students feelings about learning affect their ability to learn. One of the three strongest influencing factors on second language acquisition is classroom anxiety (Gardner, C1ment, Gliksman 1976). This study focuses on student emotional distress in the ESL classroom. The results suggest that this distress is often caused by the unfamiliar instructional practices which these students encounter in the United States. To explore student in-class emotional negativity and its relationship to instructional factors, a survey was conducted of 72 adult learners of ESL and their 15 instructors. They were asked to respond to 42 items related to 1) language skill development priorities, 2) learning/teaching activity preference, 3) classroom protocol, and 4) classroom environment. The students were also asked to describe their feelings during their ESL classtime in the United States and to state whether they preferred their classes in their native country or here, and why. Instructors were asked to give brief descriptions of their teaching styles, both in the survey and in follow-up interviews. Of the 72 learners surveyed, 23 rated their in-class feelings as mostly negative. The highest percentage of self-reported distress was found in students who had studied English for more than five years in their native country and for more than two years in the United States. Negative feelings were also increasingly more prevalent as the students became older. These findings corroborate the results of the Gardner et al. study in Canada (Gardner, C1ment, Gliksman 1976). Sex seemed to have little significance as a predictor of distress, with 34% of the men and 30% of the women reporting feelings of distress in their English classes. There were, however, differences noted among different nationalities; Asian students, for example, reported more affective problems than did other groups. A composite picture of the teachers was drawn from their descriptions and interviews. The teachers tended to characterize themselves as innovative and working in non-traditional classroom settings. They reported that they used an egalitarian, informal approach where students had a say in the choice of the content and process of the class. Passivity on the part of the students was generally discouraged. The teachers had rather low, but perhaps realistic, expectations related to their students realization of native-like fluency. Even though 68% of the students rated themselves in such a way as to appear to be adapting fairly well to their American teachers presentations, 32% did not. These 23 students appeared to have high personal goals related to all language skills, including pronunciation and grammar. They seemed to want a more traditional model of leadership and authority from their instructors in addition to more order and formality in the class.

682

TESOL QUARTERLY

The final analysis of the data indicates that the students did, indeed, have expectations and preferences that differed from those of the instructors in several ways. The areas of greatest disparity were related to the degree of formality preferred in the classroom, the necessary degree of perfection of skills, the importance of composition writing in the language learning process, and the desirability of independent group work during class hours. The data from this study have provided valuable information which will have several implications for pedagogical change within the university and adult programs where the study was done. This type of survey may also serve as a valuable prototype for other programs that wish to better accommodate their teaching programs to their students learning styles and goals. (M.A. Thesis, University of California at Santa Cruz, 1983) REFERENCES
Gardner, Robert, R, C1ment, and L. Gliksman. 1976. Second-1anguage learning: a social psychological perspective. Canadian Modern Language Review 32 (3):198-213. Authors Address: 436 9th Avenue, Santa Cruz, CA 95062

Agreement and Disagreement: A Study of Speech Acts in Discourse and ESL/EFL Materials
ELOISE PEARSON Sophia University, Tokyo

In Notional Syllabuses, Wilkins (1976) attempted to outline a theory of the communicative content of language which could be used as a basis for elaborating a language teaching syllabus. Since that time, materials based on the notional/functional approach have been steadily increasing both in number and in popularity among teachers and materials writers alike. Although many of these materials purport to teach English as it is spoken by native speakers of the language, this approach has frequently been criticized because of the absence of an empirical data base and the reliance on native speaker intuition in the implementation of the linguistic realizations of the language functions presented to the learner (see Brumfit 1980, 1981, Paulston 1981, Widdowson 1979, 1981, Stratton 1977). A desire to determine the validity of the claims made by the materials writers in the face of such criticisms prompted the empirical research reported here. For this purpose, the function agreement/disagreement as it naturally occurs in native speaker informal conversation was chosen. Data were collected by surreptitiously recording both face-to-face and non-faceto-face natural conversational interactions resulting in 950 minutes of recorded corpus. This was then transcribed and examined for instances of agreement/ disagreement. The data revealed that this function can optionally occur as a response move made by a second speaker to an initiation move made by the previous

BRIEF REPORTS AND SUMMARIES

683

speaker. In addition, the initiation move must contain some assessment of a referent so that a speaker produces agreement/disagreement as a response move. This move occurs as a second assessment of an initiation assessment move made by a prior speaker. The preferred second assessment of a referent is agreement, which occurred 74% of the time in the data; it is considered more polite than disagreement, which occurred only 26% of the time. Furthermore, it was found that the data could be classified into three types of agreement and three types of disagreement, ranging from most polite and most frequently occurring to least polite and least frequently occurring. A comparison of the results of the study was made with two ESL/EFL notional/functional textbooks which purport to teach English as it is spoken by native speakers in conversational encounters. The comparison showed that the linguistic realizations of this speech act given in the two textbooks did not concur with the conversational data collected for this study. The results of this research project indicate 1) that more empirical research is needed into what native English speakers really say as well as what they do with the language, and 2) that native speaker intuition alone is inadequate for the preparation of materials which purport to teach English as it is spoken by native speakers. (M.A. Thesis, University of Hawaii, 1983) REFERENCES
Brumfit, Christopher. 1980. Problems and principles in English teaching. Oxford: Pergamon Press. Brumfit, Christopher. 1981. Notional syllabuses revisited: a response. Applied Linguistics 2 (1):90-92. Paulston, Christina B. 1981. Notional syllabuses revisited: some comments. Applied Linguistics 2 (1):93-95. Stratton, Florence. 1977. Putting the communicative syllabus in its place. TESOL Quarterly 2 (2):131-141. Widdowson, Henry G. 1979. Explorations in applied linguistics. Oxford: Oxford University Press. Widdowson, Henry G. 1981. English for specific purposes: criteria for course design. In English for academic and technical purposes, Larry Selinker, Elaine Tarone, Victor Hanzeli (Eds.), 1-11. Rowley, Massachusetts: Newbury House. Wilkins, D. A. 1976. Notional syllabuses. Oxford: Oxford University Press. Authors Address: Sugacho 8 banchi, Shinjuku-ku, Tokyo 160, Japan

684

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 any articles or remarks published here in The Forum or elsewhere in the Quarterly.

Comments on Patrich Carrell's Cohesion Is Not Coherence

A Reader Reacts. . .
MOHSEN GHADESSY National University of Singapore

I am writing to respond to Patricia Carrells article, Cohesion Is Not Coherence, which appeared in a recent issue of the TESOL Quarterly (Vol. 16, No. 4, December 1982). I would like to argue that Halliday and Hasan (H and H), in their book Cohesion in English (1976), have justifiably established a discourse category and have demonstrated the significance of text cohesion for the preparation of instructional materials for EFL/ESL students, especially at the secondary and tertiary levels. I believe that Ms. Carrell and the other researchers referred to in support of her claims have misinterpreted the concept of cohesion as represented by H and H and have added to it attributes which it never had. I also believe that Ms. Carrells claim that H and H have equated cohesion with coherence is false and is based mainly on a consideration of discourse coherence, which must be distinguished from text cohesion. The schema-theoretical views of text processing will no doubt provide answers as to how people understand texts (i. e., how discourse coherence is achieved); they do not, however, say anything about how a string of sentences is joined together (i. e., how texture, textuality, or cohesion is achieved). If the same text is interpreted differently by different readers/listeners, then surely cohesion is not at fault. The underlying propositions and the relationships between them remain the same. The different interpretations are due to peoples interaction with the text. The first criticism by Feathers (1981), and supported by Carrell (482), states that H and Hs notion of cohesion theory operates on the superficial surface structure of a text in establishing the cohesive devices. This seems to be in direct contradiction to H and Hs emphasis throughout their book that this notion is a semantic one, 685

that it refers to relations of meaning that exist within the text, and that define it as a text (Halliday and Has an 1976:4). Ms. Carrells major criticism (483), however, is based on Morgan and Sellners views (1980) that in the example given by H and H (Wash and core six cooking apples. Put them into a fireproof dish), them in the second sentence refers to the linguistic expression six cooking apples, which they claim is H and Hs view, and not to the six cooking apples themselves. Nothing can be further from the truth, as indicated by the following quotation from H and H: They (them and six cooking apples) refer to the same thing. The two items are identical in reference, or COREFERENTIAL (Halliday and Hasan 1976:3; their emphasis). Despite this, Ms. Carrell concludes that coreferentiality is not what H and H have in mind (483). A deliberate omission on Ms. Carrells part seems to be the very important distinction that H and H make between exophoric and endophoric reference. H and H claim that only endophoric reference is a cohesive device, and therefore they are not concerned with exophoric reference. Exophoric reference contributes to CREATION of text, in that it links the language with the context of situation; but it does not contribute to the INTEGRATION of one passage with another so that the two together form part of the SAME text. Hence it does not contribute directly to cohesion as we have defined it (Halliday and Hasan 1976:37; their emphasis). However, Ms. Carrells argument is mainly based on consideration of exophoric reference. Finally, I do not see any reason why what is called texture, textuality, or cohesion should be equated with coherence. Nowhere in their book do H and H indicate that it is to be interpreted in this way, and the word coherence is not even listed in the index. Widdowson, in his book Teaching Language as Communication (1978), clearly shows the difference between the two concepts, implying that H and H have been justified in establishing cohesion as a discourse category. We know what H and H mean by cohesion. Can we ask Ms. Carrell to provide a definition for coherence?

REFERENCES
Feathers, Karen, 1981. Text unity: a semantic perspective on mapping cohesion and coherence. Manuscript, Indiana University.

686

TESOL QUARTERLY

Halliday, M. A. K., and Ruqaiya Hasan. 1976. Cohesion in English. London: Longman. Morgan, Jerry L., and Manfred B. Sellner. 1980. Discourse and linguistic theory. In Theoretical issues in reading comprehension, Rand J. Spiro, Bertram C. Bruce, and William F. Brewer (Eds.). Hillsdale, New Jersey: Lawrence Erlbaum Associates. Widdowson, Henry G. 1978. Teaching language as communication. Oxford, England: Oxford University Press.

The Author Responds. . .

PATRICIA L. CARRELL Southern Illinois University at Carbondale

Mr. Ghadessys first contention is that Ms. Carrell and other researchers referred to in support of her claims have misinterpreted the concept of cohesion as represented by H and H and have added to it attributes which it never had. If I and the other researchers I referred to have misinterpreted H and Hs concept of cohesion, we have done so based on what H and H say about cohesion and about how it relates to the broader concept of textuality, as I feel I demonstrated in my original article. Mr. Ghadessys second contention concerns Ms. Carrells claim that H and H have equated cohesion with coherence. Nowhere in my article do I claim that H and H have equated cohesion with coherence. I was careful to explain that H and H consider cohesion to be only one of two aspects of texture, the other being register. Cohesion, however, and not register, is H and Hs central focus. I do criticize H and Hs view of cohesion as a measure of, an index of, a factor in, a linguistic property contributing to, and a linguistic resource used to create what they term texture. Further, I do claim that H and Hs concept of texture, based on what they say about it, seems to be what I and others (Morgan and Sellner 1980:178) mean by coherence.
The concept of TEXTURE is entirely appropriate to express the property of being a text. A text has texture, and that is what distinguishes it from being something that is not a text. It derives the texture from the fact that it functions as a unity with respect to its environment (Halliday and Hasan 1976:2).

When, at the end of his comments, Mr. Ghadessy asks for my definition of coherence, I refer him to H and Hs definition of

THE FORUM

687

texturethe property of being a text and of functioning as a unity with respect to its environment. As I tried to make clear in the article, I took H and Hs definition of texture at face value and equated it with my (and others) notion of coherence. If H and H had something less or something different in mind with their term texture, that is not at all clear from what they say about it. I, of course, agree with Mr. Ghadessy that (discourse) coherence must be distinguished from text cohesion. What I criticize in the article is H and Hs notion that it is the cohesive devices of a language that create or contribute to coherence, or as they call it, texture.1 A third criticism of Mr. Ghadessys concerns Feathers (1981) and my contention that H and Hs notion of cohesion operates on the superficial surface structure of a text (Carrell 1982:482). While in Mr. Ghadessys view this may seem to be in direct contradiction to H and Hs emphasis throughout the book that this notion is a semantic one and that . . . it refers to relations of meaning that exist within the text, and that define it as a text (Halliday and Hasan 1976:4), in fact, there is no contradiction. The semantic notions that are captured and represented by cohesion theory are, in fact, just those which are reflected in the superficial surface structure of a text. As the quote from H and H says, just those semantic propositions which exist within the text (1976:4) are represented via a cohesion analysis. No underlying proposition which is not reflected in the surface structure of the text is so represented. Consider again the mini-text I used in the article: The picnic was ruined. No one remembered to bring a corkscrew. The underlying semantic relationship of causality which exists between the second and the first sentence is not reflected in cohesion theory because it is not expressed in the surface structure of the text. In
1 As

is clear from this discussion, as well as from the original article, I am equating the term coherence with the terms textuality and texture. I explicitly do not add cohesion to that equation, since I consider cohesion to be an effect of coherence. H and H consider cohesion to be a cause of coherence or texture/textuality. Unfortunately, Mr. Ghadessy, unintentionally I hope, contributes to the impression of an equation of cohesion with textuality and texture when he says things such as , . . how texture, textuality, or cohesion is achieved, and . what is called texture, textuality, or cohesion. It is Mr. Ghadessy, and neither I nor H and H, who is equating cohesion with textuality and texture. There is indeed some confusion in the literature about such terms as texture, textuality, cohesion, and coherence. However, in every case with which I am familiar, cohesion is the most restrictive term, normally limited to surface linguistic phenomena, Textuality is generally the most comprehensive term, apparently comparable to H and Hs texture. For example, Beaugrande and Dressier use textuality as the most general term. A TEXT will be defined as a COMMUNICATIVE OCCURRENCE which meets seven standards of TEXTUALITY (1981:3). The seven standards of textuality include cohesion and coherence as separate and interstandards, in addition to intentionality, acceptability, informativity, situationality, textuality. My schema-theoretical use of the term coherence includes many of Beaugrande and Dresders five other, user-centered, interactive standards of textuality (e.g.,, the scripts, plans, goals, schemata of intentionality and acceptability; exophora of situationality; the interactions of text-presented knowledge and stored world-knowledge of intertextuality).

688

TESOL QUARTERLY

fact, it is precisely these kinds of relationships among propositions (e.g., causality), with no overt linguistic expression, that many researchers besides myself consider to be in the domain of coherence and not cohesion. See, for example, Mr. Ghadessys own reference, Widdowson (1978), as well as Beaugrande (1980), and Beaugrande and Dressier (1981 ).2 Mr. Ghadessy next comments on the cooking apples example and the relationship of reference. A careful reading of H and H shows that Morgan and Sellner (1980) and I have correctly represented what H and H say about the cohesive relation in this example. H and H are quite clear in claiming that the texture is provided by the cohesive RELATION that exists between them and six cooking apples (Halliday and Hasan 1976:2). The italicization of the linguistic forms, plus what H and H go on to say about them, shows clearly that the cohesive relation exists between the word them and the words six cooking apples. That is, the cohesion relation is a relation between words in a text. Note H and Hs use of the term item in the continuation of the above quote: It is important to make this point, because we shall be constantly focusing attention on the items, such as them, which typically refer back to something that has gone before; but the cohesion is effected not by the presence of the referring item alone but by the presence of both the referring item and the item that it refers to (Halliday and Hasan 1976:2-3). Thus, the cohesion relation is clearly claimed by H and H to be one existing between linguistic forms within the text, and not between a linguistic form and something outside the text. Nor can the cohesion relation of reference be one of coreference, despite the partial quote offered by Mr. Ghadessy. If Mr. Ghadessy had included the complete quote from H and H, we would see that H and H use the term coreferential to talk about the meaning of the cohesion relation, not the relation itself: What is the MEANING of the cohesive relation between them and six cooking apples? The meaning is that they refer to the same thing. The two items are identical in reference, or COREFERENTIAL (Halliday and Hasan 1976:3).
2 Widdowson

speaks of cohesion as the overt, linguistically-signalled relationship between propositions (1978:31) and coherence as the recognition of an underlying illocutionary value (1978:30) linking the propositions, Widdowsons examples of coherence are very much like my picnic-corkscrew example. Beaugrande and Dressier also define cohesion as concerning the ways in which the components of the SURFACE TEXT, i.e. the actual words we hear or see, are mutually connected within a sequence (1981:3). They define coherence as concerning the ways in which the components of the TEXTUAL WORLD, i.e. the configuration of CONCEPTS and RELATIONS which underlie surface text, are mutually accessible and relevant (1981:4).

THE FORUM

689

A S I originally indicated, H and H give us no reason to believe that such a secondary, derived notion as the meaning of a cohesive relation, as opposed to the cohesive relation itself, contributes to texture.3 Mr. Ghadessys next comment concerns what he claims is a deliberate omission on my part of H and Hs distinction between exophoric and endophoric reference, and H and Hs claim that only endophoric reference is a cohesive device. The omission was certainly not deliberate, and, in fact, I am grateful to Mr. Ghadessy for raising this issue now, for it serves to further support the basic criticisms I made in the original article. Furthermore, the quote Mr. Ghadessy supplies from H and H (1976:37) supports the two immediately preceding arguments (the first about cohesion and surface structure, and the second about reference and coreference). First, since cohesion can involve only overt, superficial, surface grammatical relations within a text, exophoric reference must, by definition, be excluded as a cohesive device. Second, since coreferentiality is a relation which by definition involves reference to something beyond the linguistic forms within a text (the linguistic forms refer to the same thing, but that same thing is exophoric to the text), clearly H and H cannot have intended coreferentiality as a cohesive relation. And finally on this issue, as to Mr. Ghadessys claim that my argument is mainly based on consideration of exophoric reference, I would point out that my argument is built on the very same examples used by H and H or, at least in the case of lexical cohesion, on examples very much like H and Hs. If my argument depends on exophoric reference, it is because H and Hs own examples of reference cannot be explained without appeal to extratextual matters. To reiterate my basic point from the original article, cohesion may be an interesting linguistic phenomenon to study and to teach to EFL/ESL students; however, we must be clear on the relationship between cohesion and coherence. Cohesion does not create coherence, or texture. Cohesion can be a meaningful concept only if we assume that coherence is a logically prior concept; we cannot speak sensibly of linguistic cohesion unless we can assume psychological coherence. Overt cohesion depends on an assumption of prior psychological coherence. Indeed, cohesion may be viewed as making overt or explicit those meanings which are present covertly because of a texts coherence (see Widdowson 1978:27-31). Cohesion supports coherence (Beaugrande and Dressier 1981:101); cohesion is grafted onto, makes explicit, is parasitic on, presupposes, is an effect of (not a cause
3 This

confusion by H and H between meaning relations (e g., cohesive relations.) and the meaning of such meaning (cohesive) relations is claimed by Morgan and SelIner to reflect H and Hs quite general confusion on matters of meaning (1980:180). Morgan and Sellner cite additional examples from H and H for this claim,

690

TESOL QUARTERLY

of) coherence. Wash and core six cooking apples. Put them into a fireproof dish (Halliday and Hasan 1976:2). Although this text is claimed by H and H to contain an overt cohesive tie, interpreting it as such depends on a logically prior assumption of coherence. If we werent able to construct a textual world in which a continuity of senses (Beaugrande and Dressier 1981:84) dictated our interpreting the second sentence as being coherently relatable to the first one, it would make no sense to speak of a linguistic cohesive relation. As I pointed out in the original article: What leads to the conclusion that them, in fact, is intended to refer to the apples and not, say, to the authors children, or the pages of the cookbook, or anything else for that matter? It is not knowledge of language that leads to this conclusion. It is our background knowledge of cooking and of the authors purpose, as well as our ability to reason, and the assumption that the recipe is coherent. Without this latter assumption, there would be no way of knowing what them is intended to refer to (Carrell 1982:483).

REFERENCES
Beaugrande, Robert de. 1980. Text, discourse, and process: toward a multidisciplinary science of texts. Norwood, New Jersey: Ablex Publishing Corporation. Beaugrande, Robert de, and Wolfgang Dressier. 1981. Introduction to text linguistics. London: Longman. Carrell, Patricia L. 1982. Cohesion is not coherence. TESOL Quarterly 1 6 (4):479-488. Feathers, Karen. 1981. Text unity: a semantic perspective on mapping cohesion and coherence. Manuscript, Indiana University. Halliday, M. A. K., and Ruqaiya Hasan. 1976. Cohesion in English. London: Longman. Morgan, Jerry L., and Manfred B. Sellner. 1980. Discourse and linguistic theory. In Theoretical issues in reading comprehension, Rand J. Spiro, Bertram C. Bruce, and William F. Brewer (Eds.). Hillsdale, New Jersey: Lawrence Erlbaum Associates. Widdowson, Henry G. 1978. Teaching language as communication. L o n d o n : Oxford University Press.

THE FORUM

691