Traditions of Writing Research
Traditions of Writing Research reflects the different styles of work offered at the Writing Research Across Borders conference. Organized by Charles Bazerman, one of the pre-eminent scholars in writing studies, the conference brought together an unprecedented gathering of writing researchers. Representing the best of the works presented, this collection focuses solely on writing research, in its lifespan scope bringing together writing researchers interested in early childhood through adult writing practices. It brings together differing research traditions, and offers a broad international scope, with contributor-presenters including top international researchers in the field. The volume’s opening section presents writing research agendas from different regions and research groups. The next section addresses the national, political, and historical contexts that shape educational institutions and the writing initiatives developed there. The following sections represent a wide range of research approaches for investigating writing processes and practices in primary, secondary, and higher education. The volume ends with theoretical and methodological reflections. This exemplary collection, like the conference that it grew out of, will bring new perspectives to the rich dialogue of contemporary research on writing and advance understanding of this complex and important human activity.
Traditions of Writing Research
Edited by Charles Bazerman Robert Krut Karen Lunsford Susan McLeod Suzie Null Paul Rogers Amanda Stansell
First published 2010 by Routledge 270 Madison Ave, New York, NY 10016 Simultaneously published in the UK by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN Routledge is an imprint of the Taylor & Francis Group, an informa business This edition published in the Taylor & Francis e-Library, 2009. To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of eBooks please go to www.eBookstore.tandf.co.uk. © 2010 Taylor & Francis All rights reserved. No part of this book may be reprinted or reproduced or utilized in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. Library of Congress Cataloging in Publication Data Traditions of writing research / edited by Charles Bazerman ... [et al.]. p. cm. Papers presented at the 2008 WRAB conference. 1. Rhetoric–Study and teaching–Congresses. 2. Rhetoric–Research– Congresses. 3. Composition (Language arts)–Study and teaching– Research–Congresses. 4. Written communication–Research–Congresses. I. Bazerman, Charles. P53.27T73 2009 2009015687 808–dc22 ISBN 0-203-89232-1 Master e-book ISBN
ISBN10: 0-415-99337-7 (hbk) ISBN10: 0-415-99338-5 (pbk) ISBN10: 0-203-89232-1 (ebk) ISBN13: 978-0-415-99337-1 (hbk) ISBN13: 978-0-415-99338-8 (pbk) ISBN13: 978-0-203-89232-9 (ebk)
Contents
Preface
ix
Part I
Approaches in various regions
1
1 Modern “writingology” in China
3
C h en Hui j un
2 The French didactics approach to writing, from elementary school to university
17
I . D elcambre and Y . R euter
3 What factors influence the improvement of academic writing practices? A study of reform of undergraduate writing in Norwegian higher education
31
O l g a D yst h e
4 Mapping genre research in Brazil: an exploratory study
44
A ntonia D ilamar A ra ú j o
5 The teaching and learning of writing in Portugal: the case of a research group
58
L u í sa Á l v ares P ereira , C oncei ç ã o A leixo , I n ê s C ardoso , and L uciana Gra ç a
6 Spanish research on writing instruction for students with and without learning disabilities Jes ú s - N icasio Garc í a , A na - M ar í a de C aso - F uertes , R a q uel F idal g o - R edondo , O l g a A rias - Gund í n , and M ark T orrance
71
vi Contents Part II
Writing education in political and historical contexts
83
7 Writing, from Stalinism to democracy: literacy education and politics in Poland, 1945–1999
85
C e z ar M . O rnatowski
8 A pilot investigation: a longitudinal study of student writing in a post-totalitarian state
97
Gil Harootunian
9 The continuum illiterate–literate and the contrast between different ethnicities
111
M aria S í l v ia C intra
10 Strategies, policies and research on reading and writing in Colombian universities
122
B lanca Y anet h Gon z ále z P in z ó n
Part III
Research on primary and secondary school practice
133
11 Young children revising their own texts in school settings
135
M irta C astedo and E milia F erreiro
12 Written representations of nominal morphology by Chinese and Moroccan children learning a Romance language
151
L iliana T olc h insky and N aym é S alas
13 Relationships between idea generation and transcription: how the act of writing shapes what children write
166
Jo h n R . Hayes and Vir g inia W . B ernin g er
14 Academic writing in Spanish compulsory education: improvements after didactic intervention on sixth graders’ expository texts
181
T eodoro Á l v are z A n g ulo and I sabel Garc í a P are j o
15 Caught in the middle: improving writing in the middle and upper primary years Val F aulkner , Judit h R i v alland , and Janet Hunter
198
Contents vii 16 Teachers as mediators of instructional texts
212
S u z ie Y . N ull
17 Pushing the boundaries of writing: the consequentiality of visualizing voice in bilingual youth radio
224
D ebora h R omero and D ana W alker
18 Classroom teachers as authors of the professional article: National Writing Project influence on teachers who publish
237
A nne W h itney
PART IV
Research on higher education practice
249
19 The international WAC/WID mapping project: objectives, methods, and early results
251
C h ris T h aiss
20 Rhetorical features of student science writing in introductory university oceanography
265
Gre g ory J . K elly , C h arles B a z erman , A udra S kukauskaite , and W illiam P rot h ero
21 Reading and writing in the social sciences in Argentine universities
283
P aula C arlino
22 Preparing students to write: a case study of the role played by student questions in their quest to understand how to write an assignment in economics
297
B arbara W ake
23 Can archived TV interviews with social sciences scholars enhance the quality of students’ academic writing?
309
T erry I n g lese
24 Social academic writing: exploring academic literacies in text-based computer conferencing W arren M . L iew and A rnet h a F . B all
325
viii Contents 25 Between peer review and peer production: genre, wikis, and the politics of digital code in academe
339
D oreen S tarke - M eyerrin g
PART V
Theories and methodologies for understanding writing and writing processes
351
26 Writing in multiple contexts: Vygotskian CHAT meets the phenomenology of genre
353
D a v id R . R ussell
27 The contributions of North American longitudinal studies of writing in higher education to our understanding of writing development
365
P aul R o g ers
28 Statistical modeling of writing processes
378
D aniel P errin and M arc W ildi
29 Writers’ eye movements
394
M ark T orrance and Å sa W en g elin
30 Text analysis as theory-laden methodology
406
N ancy N elson and S tep h anie Grote - Garcia
31 On textual silences, large and small
419
T h omas Huckin
Index
432
Preface
The work of writing researchers today crosses many geographic and disciplinary borders. Researchers who view writing as a complex human activity in Asia, Europe, South America, Africa, Australia, and North America are increasingly working together and drawing upon each other’s work in carrying out their own research programs. Writing researchers in anthropology, psychology, linguistics, rhetoric and composition, sociology, science studies, cultural studies, and education are extending the breadth and depth of their research. While, appropriately, a great deal of this research has focused on the learning and teaching of writing in both L1 and L2 school settings, research on writing continues to expand into many new and emerging areas of practice, including the acquisition of print literacy prior to schooling; writing across the curriculum, in the disciplines, and in the professions; and writing development across the lifespan. Additionally, new political exigencies, educational pressures, research methods, and technological tools have stimulated writing research and deepened our scientific understanding of how writers write and what writing does. For example, in the study of writing and cognition, the use of eye-tracking software in combination with the monitoring of keyboard strokes and handwriting has yielded new insights into composing processes, which have complemented, refined, and built upon the early work on cognition and writing, which began in the 1970s. It would appear that writing research is growing out of its adolescent phase, and is gaining an equal footing with research on reading. Rather than a smattering of isolated studies we now see the emergence of global traditions of writing research. While the interest and activity in writing is global, the responses are local. The educational institutions around the world that potentially provide writing instruction are themselves varied in policies, structures, ideologies, and practices. Approaches also vary according to the level and type of educational setting. Even more, the traditions of research developed to comprehend these settings and practices are necessarily varied, as they reflect responses to specific educational challenges and the histories of different academic communities. The current great interest and energy directed toward developing writing pedagogy, practices, and programs as well as research include a
x Preface mutual recognition of the value of exchanging knowledge and experience among those of different regions, educational traditions, and research communities. Networks of writing research, programs, and practice that have been gradually building in North America have also been developing in Europe, Latin America, and Asia Pacific. Moreover, across the globe, particularly vibrant writing research communities have been emerging in the last two decades. These communities have been generating an increasing number of opportunities and vehicles for sharing knowledge. Two international handbooks on writing research have appeared recently, and international conferences (such as the International Writing Across the Curriculum (IWAC), the International Symposium on Studies on Genre (SIGET), the International Writing Centers Association (IWCA), the European Asssociation of Learning and Instruction’s Writing Special Interest Group (EARLI SigWriting), the European Association for the Teaching of Academic Writing (EATAW), Red de Lectura y Escritura en Educación Superior (REDLEES), the Conference on College Composition and Communication (CCC), and the Canadian Association for the Study of Discourse and Writing (CASDW)) have gained increasing global participation. The Writing Research conference series sponsored by the University of California, Santa Barbara has followed a similar trajectory. What began as a small regional conference in 2002 expanded to a national research conference with a smattering of international participation in 2005. In 2008, it expanded again to a global conference that aimed at being inclusive of all regions, all educational and writing developmental levels, and all research and theory traditions. Surpassing the organizers’ initial expectations, the 2008 meeting hosted over 650 participants from 33 countries from all continents. Attendees included writing researchers representing the full spectrum of writing development, from the communication of small children to the lifelong learning strategies of retirees. The 2008 Writing Research Across Borders (WRAB) conference brought together and provided opportunities for interchange among people from widely varying perspectives that reflected the multifaceted practices that comprise writing, as well as the multifaceted political realities that shape educational institutions. Participants defined their own research trajectories in diverse ways, with some focusing on national and regional histories of writing pedagogy and disciplinarity (as in Brazil, France, and China), and others focusing on methodological and theoretical approaches (such as cultural-historical activity theory, computer parsing of linguistic databases, cognitive theories of writing, and ethnographic observations and interviews). In addition to characterizing different research traditions, the over 450 presentations often included concrete examples of classroom materials that represented different writing curricula and activities based on those traditions, as well as ways of reflecting on and evaluating the classroom. Thus, participants encountered three days’ worth of sharp contrasts, often learning about research and pedagogies they had no prior inkling of, and becoming aware of intellectual frameworks that challenged
Preface xi previously held assumptions. Out of this intellectual ferment came deeper appreciations of the intertwined complexities of writing, writing research, and literacy pedagogy. Although no conference attendee could imbibe all the richness the three days offered, nor could any conference represent all writing research under way, we are all becoming aware that there is far more to the world of writing research and educational practice than any individual had been previously aware. Comprehensive montages of global research on writing are just being sketched out now, and their completion lies in the future. Likewise, we have just begun the archeological uncovering of what lies behind each of the traditions we are witnessing. Still, even though we have little certainty of how to synthesize the global range of knowledge and experience into a more comprehensive account of writing, it is time for us to look about with open minds and methodological flexibility to consider what each tradition is producing so that we can carry our common interests forward. This volume attempts to be a wide-ranging sampler of the best writing research currently under way in the world, at least as it was represented at the 2008 WRAB conference. The editors hope that the volume has the same engaging and challenging effect on readers as the conference did on attendees. The volume’s contents reflect the different styles of work offered at the conference. The opening section presents six writing research agendas from different regions and research groups. In the next section, four chapters address the national, political, and historical contexts that shape educational institutions and the writing initiatives developed there. The two sections that follow represent a wide range of research approaches for investigating writing processes and practices in primary, secondary, and higher education. The volume ends with several theoretical and methodological reflections. The editors hope this volume, like the conference that it grew out of, will bring new perspectives to the rich dialogue of contemporary research on writing and advance our understanding of this complex and important human activity. Charles Bazerman, Robert Krut, Karen Lunsford, Susan McLeod, Suzie Null, Paul Rogers, Amanda Stansell
Part I
Approaches in various regions
1 Modern “writingology” in China1 Chen Huijun D epartment of Foreign Languages, China University of Geosciences, Beijing, China
Introduction There are four popular English translations for the term “Xie3 Zuo4 Xue2”2 in China: “writingology,” “theories on writing,” “writing studies,” and “writing research.” “Writingology” is chosen here as it best matches the concept of its Chinese equivalent, which refers to a branch of social sciences that studies the laws lying behind the act, the art, the process, and the product of writing. According to the Xinhua Chinese Dictionary, writing is a human-specific activity, the narrow sense of which refers to writing texts specifically—i.e., penning or forming letters or words to record, transmit ideas or to express emotions or feelings, including writing school compositions—and the broad sense also covers translating and compiling activities and creating artistic products, such as music, drawings, and movies. However, “writingology” only studies writing in its narrow sense, and the disciplinary architecture covers both (specific) studies and (abstract) theories on literary works, rhetoric, and school compositions. Two opinions exist concerning the division of the history of Chinese “writingology.” Some researchers support a two-stage development, namely, “ancient writingology” and “modern writingology”; the “5.4 Movement” (1919) being the dividing line. Others support a three-stage division, with the 5.4 Movement separating “ancient writingology” from “traditional writingology” and the founding of the China Writing Society in 1980 separating “traditional writingology” from “modern writingology.” This chapter follows the latter opinion. Chinese “ancient writingology” originated from literary writing. Despite the constant turbulence from shifts of political power and frequent wars in ancient China, literature flourished. There were many great literary and philosophical masters, including Confucians who produced the earliest theories on writing. Confucius’ (Confucius: 551–479 bc) remarks on writing poetry played an important role in guiding the development of Chinese “writingology.” However, later feudal governments chose officials or offered scholarships based on applicants’ performance at exams at which examinees were required to improvise “eight-legged” (Baku) texts of a fixed format and a limited number of words in ancient Chinese rather
4 Chen Huijun than contemporary daily Chinese. As time went by, such texts became so archaic and difficult for later generations to understand that education was (and even now is still) necessary for interpretation, which greatly affected the spread of knowledge and information. To change this situation, some great writers (e.g., Hu Shi and Lu Xun) initiated the “Vernacular Movement” a few years prior to the 5.4 Movement in 1919, proposing to replace ancient Chinese with contemporary daily Chinese in writing. It was not only a reform in linguistic expression, but also a reform in the content and conventions of writing. The reformation received great resistance from those who argued that writing in ancient Chinese was real scholarly writing. Despite this, the reformers struggled and pushed forward the reformation during the 5.4 Movement. The 5.4 Movement was started by students in Peking on May 4, 1919 and later extended to other parts of the country. In the Movement, out of patriotism, students fought against the government, opposing the offering of land and territories to other countries. Many famous writers, including those reformers in the Vernacular Movement, were also actively involved in the 5.4 Movement and supported the Movement via writing articles to criticize the government. For propaganda purposes, they wrote in vernacular contemporary Chinese so as to be understood by common people, and thus they became very popular and influential. In this way, archaic ancient Chinese was mostly replaced by vernacular (contemporary daily) Chinese. As a result, theories on archaic ancient Chinese writing could no longer provide satisfactory explanations for phenomena in contemporary writing, and accordingly new theories were needed to guide the practice and teaching of writing. Having no ready theories, leaders of the Vernacular Movement introduced theories on grammar, stylistics, and rhetoric from Western countries (the United Kingdom, the United States, France, Germany, and Italy) and literary theories from the Soviet Union. Combining these theories with Chinese writing practice, Chinese writing scholars gradually built a new system of knowledge on writing. Therefore, the 5.4 Movement in 1919 is generally considered the dividing line of Chinese “ancient writingology” and Chinese “traditional writingology” (or the dividing line of “ancient writingology” and “modern writingology” to some researchers). In “traditional writingology,” literary studies still remained in the spotlight, and rhetoric studies began to draw increasing attention. Although both ancient writing scholars and traditional writing researchers have produced numerous theories on writing, few of them are aware of the disciplinary construction of “writingology.” In the late 1970s, modern Chinese writing scholars created the concept of “writingology” and proposed constructing “writingology” as an independent discipline. There after, they founded the first professional organization of writing in China—the China Society of Writing—in 1980 and issued the first professional journal, Writing, in 1981. Ever since, Chinese “writingology” has entered a new, organized era. Although traditional literary studies remain
Modern “writingology” in China 5 powerful, rhetoric studies have made great progress. Therefore, the founding of the China Society of Writing is widely considered an epoch-making event that marked the beginning of “modern writingology.” A careful search among publications in English shows that no studies have touched upon the introduction of Chinese “modern writingology.” This chapter focuses on Chinese “modern writingology,” aiming to make known recent progress in Chinese theoretical writing studies and to show what contributions Chinese scholars have made to writing studies in the world.
Changes in comparison with “traditional writingology” Since the 1980s, great changes have taken place in Chinese “writingology” thanks to the practice of the open-door policy and the freedom policy issued in late 1978, two years after the Cultural Revolution. Like everything else in China, writing studies drew to a halt during the disastrous ten-year Cultural Revolution, which was started by Chairman Mao. The executives of his policy went to extremes and exercised severe control over the mass media in all cultural fields (broadcasts, movies, plays, books, articles, etc.). They did not allow people to express different opinions and sent opponents to jail or farms to receive “re-education” (political brain- washing). The Cultural Revolution turned out to be a disaster for the whole nation. In the years that followed, the influence was still pervasive. The top governor of the nation Hua Guofeng, the immediate successor of Chairman Mao, insisted that all Chinese people should absolutely abide by whatever policies Chairman Mao had issued and should follow whatever Chairman Mao had said. This situation ended in December 1978, when the third plenary meeting of the eleventh Central Committee of the Communist Party (CCCP) demolished the extreme restrictions, issued the open- door policy, and put forward the proposal for freedom of academic debate. This has made researchers and scholars active and has brought profound changes in China in all fields, including Chinese writing studies. The founding of the first professional organization and the issue of the first professional journal Before 1980s, no organizations and journals were committed to professional communication or to research on writing in China; no conferences or forums were dedicated to writing or composition studies. The open- door policy and proposal for freedom of debate put forward at the eleventh CCCP were inspiring to writing scholars; they became active in research and grew comfortable challenging different opinions. In 1978–1980, the debate over the promise of writing arose because of the widespread negative views in some colleges and universities: “There is little science in writing,” “Writing research is not promising,” and “Writing instructors do not have a bright future.” These views had a great
6 Chen Huijun impact on people’s attitude toward writing studies and the teaching of writing in universities. Many researchers and instructors lost confidence in the future of their careers, and became less motivated in their work. To make matters worse, this attitude influenced many students, who had already become fed up with the difficulty of writing and were bored with those tedious practices irrelevant to their daily life. To prevent the further spread and influence of these negative thoughts, writing scholars who still had faith in writing organized seminars and forums for discussions. They argued for the existence of scientific laws in writing and also for the necessity of exploring the laws. They believed that the importance of writing in life made it certain that writing studies and the teaching of writing would have a great future. One of the rewarding results of the arguments was that some scholars eventually realized that the lack of confidence in writing might be due to insufficient communication and education about writing. Hence, they felt it necessary to set up a professional organization for disciplinary construction and communication about writing. After two years’ effort, they finally founded the China Writing Society in 1980 and officially released the journal Writing in 1981. Ever since then, writing studies in China have entered an organized era and have been developing fast. The shift in research and teaching about writing In the late 1970s, still under the influence of Soviet theories, textbooks in universities continued stressing theoretical knowledge, which did not help to improve students’ abilities to write compositions. Therefore, many people lost faith in writing classes in colleges and universities. Even Ye Shengtao, the most respected and prestigious educator and writer of the time, complained that writing courses in universities were similar to those in secondary schools; those courses on writing were just make-up lessons that students missed during the Cultural Revolution; courses on writing should eventually be canceled. His fame and position made the view spread widely and quickly, and immediately it aroused a heated debate in 1984–1986.3 Many scholars criticized this view from different perspectives, arguing that the teaching of writing in universities ought to be completely different from the composition classes in high schools in regard to the theoretical level, the goal, the tasks, and the requirements, and therefore courses on writing should not be canceled in colleges and universities. Encouragingly, Writing (5/1984) published a preface with an inspiring note from the President of the National Political Consultative Conference Deng Yingchao, which said, “Recover the prosperity of the writing discipline to serve the Construction of Four Modernizations.” This settled the debate and saved writing classes in universities. Eventually, the writing world found that the over-emphasis on theoretical knowledge was the direct cause of the confusion as it did little help in improving students’ writing skills. Therefore,
Modern “writingology” in China 7 many writing scholars supported the shift from stressing theoretical knowledge to stressing training abilities and skills. This shift can be clearly traced in textbooks edited by Lu (1983, 1984). The 1983 version followed the traditional “eight-block” framework, which consists of “introduction,” “material,” “theme,” “structure,” “expression,” “language,” “style,” and “revision,” whereas the 1984 version shifts to “selecting material,” “deciding on a theme,” “structuring the layout,” “writing techniques,” “expressing ideas,” “making clauses,” and “revising,” which displays a combination of knowledge and training of skills. After Lu, more scholars proposed theoretical research on the abilities and skills of writing (e.g., Li, 1993) and some influential theories appeared concerning the training of writing abilities, such as “Writing Ability Theory” (Lin, 1985), “Systemic Theory” (Du, 1988a), and “Three-level Training System” (Gao & Liu, 1989). Dynamic, writer-oriented studies have replaced and then have been combined with static, text-oriented research; multiple patterns have replaced the homogeneous static textual-analysis pattern in research into writing Influenced by Soviet theories on knowledge of writing imported in the 1950s, research and teaching on writing in the 1970s followed the traditional, static text-oriented “eight-block” framework. In 1990–1997, a debate arose on whether research on writing should be writer-oriented or text-oriented. Many scholars believed that the basis of research about writing was the writing behavior of the “writer,” and therefore, theories on writing should cover the whole composing process, and writing studies should center on the “writer,” because the ultimate goal of composing was to express the thoughts of human beings. Thereafter, many researchers shifted to focusing on writer-oriented dynamic writing processes and reconsidering the process and laws of writing from dynamic and writer-oriented perspectives. Summary of Composition Methods (Li, 1982) first displayed this change by replacing the traditional “eight blocks” with more dynamic writer-oriented content, namely “introduction, composing process, the theme, spreading the theme—choosing materials, arranging materials, writing paragraphs, writing phrases and sentences, and revising.” As research has progressed, some scholars have come to realize that both overdue attention to the “writer” and ignorance of writing conventions might lead to overemphasis on subjective research and ignorance of objective issues. Researchers have begun to pay more attention to scientific analysis of texts while focusing on the “writer.” They believe that only in this way can they find out the causes of success or failure of a text and shape related theories. Therefore, many scholars advocate combining dynamic writer-oriented research with static textual analysis. Consequently, multiple patterns have replaced the homogeneous textual analysis that dominated writing studies for decades: the “process–
8 Chen Huijun elements” pattern, e.g., Summary of Composition Methods (Li, 1982); the “general–details” pattern, e.g., Basics of Writing Theory (Liu, 1985); the “intelligence–skills” pattern, e.g., Summary of Basic Writing Theories (Lin, 1985); the “text–writer–summary” pattern, e.g., Modern Theories of Writing (Zhu, 1986a); the “essence–process–techniques–styles” pattern, e.g., New Manuscript on Modern Writing (Pei, 1987); the “basic theories– stylistic theories” pattern, e.g., Advanced Course of Writing Theories (Zhou, 1989); and the “nature–process–behavior–teaching” pattern, e.g., Introduction to Advanced “Writingology” (Ma, 2002). Writing studies has shifted from stressing literary research to stressing practical studies and then to stressing parallel development of both Literary studies have a long history since ancient China. Research into practical writing used to be neglected. The routine for research into practical writing used to follow the “format+examples” pattern, and before the 1980s, practical research used to lag far behind literary studies. With the rapid development of society and the economy, practical writing is rising in popularity and thus is becoming a hot research topic. From 1994 to 2007, nine “International Seminars on Practical Writing” took place in China, and publication of practical studies increased sharply (e.g., Li, 1988; Shao & Ye, 1998; Wang, 1985; Ning, 1999; Yu, 1992; Kuang, 1986; Si, 1984). Apart from those on basic theories, countless publications appear on branch theories of practical writing, such as administrative literacy (e.g., Wang, 1985; Zhou, 1994; Miao, 1993), scientific literacy (e.g., Si, 1984; Jiang, 1986), legislative literacy (e.g., Ning, 1999), financial literacy (e.g., Yu, 1992; Li, 1987), and news literacy (e.g., Kuang, 1986; Hong, 1986). Essential theoretical issues, such as the aim, essence, tasks, functions, and means of practical writing have been made clear. Cooperation has started among scholars from the mainland, Hong Kong, Macao, and Taiwan since the first international seminar on practical writing in 1994. Series of textbooks have been published. An influential one is Xian4 Dai4 Ying1 Yong4 Wen2 (Modern Applied Writing) (Yu, 1996). The book is co-authored by scholars from the mainland, Hong Kong, and Taiwan. This bilingual (Chinese and English) textbook focuses on subgenres of applied writing. Applied texts are divided into nine sub-genres: “correspondence,” “administrative documents,” “ceremonial documents,” “documents for modern enterprises,” “international trading documents,” “contracts and agreements,” “advertisements,” “international legislative documents,” and “reports.” The book contains a detailed discussion on the characteristics, functions, production, forms, language, requirements of each sub-genre and classification of emerging sub-genres, such as e-documents, emails, and e-transmission of charts and texts. The discussion of each genre is followed by various exercises.
Modern “writingology” in China 9 Writing studies have shifted from focusing on traditional paper-and-pen literacy to focusing on both traditional and modern digital literacy Since the 1990s, the practice of a market economy has had an enormous impact on writing studies. Many people began caring about making money from writing. Studies about practical writing that can directly serve economic and social needs have become extremely popular. Relevant businesses, for example advertisement companies, writing services, and writing training centers, have been making good money. In contrast, high-level fundamental research can hardly make any profit. This led to a debate during 1994–2000 about the value of research on writing in a market- economy society. Serious discussions reached the common ground that current writing studies should be in alignment with modern social needs; researchers and writers should care about both the economic benefits and social functions of writing. In particular, scholars should further develop theories on practical and digital literacy to satisfy modern social needs. Many publications on digital literacy have appeared, including Dian4 Nao3 Xie3 Zuo4: Fang1 Fa3 Yu3 Ji4 Qiao3 (Writing via Computers: Methods and Techniques) (Jin, 1999). According to Jin, computer literacy is different from traditional (pen– paper) literacy in that computer writing is a writer-characteristic activity that records, exchanges, and transfers information with a computer as a medium, with the keyboard a language-input device, with the monitor as the interface, with the software or hard-disk as the carrier. His framework of “computer writingology” consists of “thinking pattern,” “Chinese interface,” “input methods,” “searching material,” “expressing means,” “structure,” “linguistic features,” “stylistic features,” “aesthetics,” “drafting and revising,” “saving and printing,” “practical techniques,” and “common styles.” Additionally, Jin reveals three important characteristics of computer literacy. The first is nonlinearity, i.e., hyperlinks enable the writer to jump easily from one point to another, from one text to another, and even from one medium to another, and thus he breaks the old sacred laws of linear narrative. The second is networkability, referring to the forming process of an interactive network consisting of the writer, media, texts, and other factors while the writer goes back and forth among different media or texts, which greatly expands the free space of thinking. The third is auto nomy, which refers to the fact that high technology provides the writer with many choices of media or forms to express himself or herself (e.g., words, sounds, flashes, cartoons, videos, pictures, and movies), and the writer can exercise control over the production and reproduction capacity of these means while producing a piece of writing on computer.
10 Chen Huijun
Current status of Chinese “modern writingology” Open-ended theoretical frameworks have been established at different levels With the development of “Chinese modern writingology,” researchers are no longer satisfied with just focusing on specific issues, such as “selecting materials,” “choosing the theme,” “structuring the text,” “beginning a composition,” “arranging materials,” and “ending a composition” (Zhang, 1952; Wu, 1954; He, 1958; Hu, 1960; Peking University, 1964). Instead, they have started constructing a framework for Chinese “modern writingology,”4 e.g., the disciplinary theoretical system (e.g., Liu, 1985; Pei, 1987; Zhu, 1986a; Zhou, 1989), the training system (e.g., Wu et al., 1985; Li, 1988; Wei, 1998; Pan, 2000; Ma, 1999), the teaching system (Li, 1994; Dong, 2000), and the assessment system (Zhu, 1986b). Writing studies are now developing toward the parallel development of studies on details and research on macro-strategic issues (Du, 2004; Lin, 1991; Jin, 1996; Chen, 1995; Dong, 2000). Currently, an open-ended general theoretical framework has been accomplished. Using Maxims as the philosophical ground, the framework consists of three levels: basic theories, branch theories, and theories on writing training. At the upper level, basic theories cover the definition, essence, scope, fundamental laws, and constitution of writing (e.g., processing theories, ability theories, and product theories); theories at this level orient the other two levels. Located at the middle level are branch theories (e.g., literary writing theories, applied writing theories, news writing theories, economics writing theories, and administrative writing theories), interdisciplinary writing theories (e.g., psychology of writing, thinking of writing, and esthetics of writing), and historical research (e.g., history of writing, history of writing research, history of teaching of writing); theories at this level show the direction of development. The lower level covers theories on training (e.g., theories on training of writing styles and assessment of training), which aims to provide solutions to specific problems in actual writing activities. This general framework is open to the integration of theories from other disciplines, for example philosophy, psychology, esthetics, sociology, and linguistics, which leads to the occurrence of branches of “writingology,” such as “literary writingology” (e.g., Sun, 1987), “applied writingology” (e.g., Yu, 1996), “scientific writingology” (e.g., Si, 1984), “news writingology” (e.g., Kuang, 1986), thinking of writing (e.g., Jin, 1989), esthetics of writing (e.g., Lin, 1991), “computer writingology” (Jin, 1999), and psychology of writing (e.g., Lin, 1985). Certainly, with the fast development of the sciences, writing scholars will continue integrating into their writing studies new knowledge from emerging disciplines, which will lead to more emerging branch theories of Chinese “writingology.” Apart from the general framework, specific frameworks have also formed. For example, at the level of basic theories, Jin (1996) proposes
Modern “writingology” in China 11 that the framework consists of a vertical system and a horizontal system. The vertical framework shows the interactive relationship among four links (object, apperceiving, conceiving, and text), and the horizontal framework shows the interactive relationship of four elements: object, writer, carrier (text), and recipient (audience). “Object” is the source and base of writing; “writer” is the dominator in writing; “carrier” is the result or product of writing; “recipient” is the recipient of the product. Therefore, this framework takes “writer” as the research center, aiming at characterizing both the links and the entire dynamic running process. Therefore, this framework stresses both dynamic research on the writing process and static textual analysis. Further, Lin (2002) proposes that, horizontally, the four basic elements of writing, namely, “writer,” “object,” “carrier” (the text) and “recipient” (audience), are transformed into the writer’s mind and turn into four forms of ideology: personal ideology, thematic ideology, ideology about audience, and ideology about product; the writer sums up these four forms of ideology via perception, digests them at the conceiving stage and unifies them at the expressing phase and finally produces a product. Vertically, there exists a ternary transformational law: the writer first internalizes the object (internalization), and then apperceives the internalized object (apperception), and finally externalizes the ideology (externalization). In a writing process, the writer operates and exercises control over four constituent links: collecting, apperceiving, expressing, and assessing. The approval for enrolling MA students marks the official acknowledgment of “writingology” as a discipline It is inspiring to Chinese writing scholars that in 2006 Wuhan University and Fudan University received approval to enroll MA students of writing as a secondary discipline under the primary discipline of Chinese language and literature. In 2007, these two universities began accepting students majoring in rhetoric in Wuhan University and creative (literary) writing in Fudan University. Wuhan University sets two directions: basic theories of “writingology” and “styles of writing.” The former focuses on historical studies about ancient, traditional, and modern theories on rhetoric; further exploration about the essence and laws of writing; and methodology of teaching and research on basic “writingology.” The latter studies the evolution of some common genres, the features of those genres that are urgently needed in society, and also methodology of teaching and research on each genre. Fudan University sets three directions: university “writing ology,” research and practice of fiction (novel) writing, and research and practice of prose writing. The approval for acceptance of MA students marks the official acknowledgment of “writingology” as a discipline.
12 Chen Huijun Schools of research have been formed 5 With the rapid development of “writingology,” several influential schools of research on writing have formed in China. The first consists of those who focus on textual analysis, especially on structural analysis of texts, aiming to discover attributes that lead to the success of an article; scholars following this tradition include Zhang Zhigong, Zhang Shoukang, Cheng Funing, Zeng Xiangqin, Zheng Shouyin, Chen Yali, Wu Bowei, Yang Yinhu, and Zhang Enpu. The second is those who focus on philosophy of writing; influential scholars include Tang Daixing, Zhang Weide, and Zhao Yu. The third is those who focus on the principles of writing (e.g., the dynamic process of writing), aiming to uncover the laws and characteristics of the whole process and the constituents of human writing activity that include the writer, the object, the carrier (text), the audience, the essence, perception, thinking, expressing, and revising; famous scholars include Pei Xiansheng, Zhu Boshi, Lin Kefu, Jin Changmin, Liu Xiqing, Yan Chunjun, She Zuochen, Zhang Deyi, and Li Jinglong. The fourth is those who study the training of writing, aiming to establish an effective training system to improve basic writing skills and abilities; scholars following this tradition include Lin Kefu, Wu Sijing, Gao Yuan, Li Baijian, and Du Benchen. The fifth is those who study psychology or cognition (particularly wisdom or intelligence), focusing on the constitutional and developmental laws of the writer’s intelligence or wisdom of writing; the most important scholars in this tradition include Ma Zhengping, Liu Xiaogang, Jin Daoxing, Qiu Anchang, Liu Yu, Wang Kejian, Zhou Decang, and Gao Nan. The sixth is researchers who focus on applied or practical writing; scholars in this tradition include Zhou Jichang, Hong Weilei, and Zhang Chuanzhen. The seventh is those who focus on historical studies; influential scholars are Wang Zhibin, Zhao Yu, and Pan Xinhe. The eighth is those who pay special attention to studies on how cultures affect writing activities; well-known scholars include Ma Zhengpin, Ren Suihu, and Tao Jiawei.
Conclusion: direction of development Jin (1996) proposes three trends of Chinese “modern writingology.” The first is practicality. To be practical means to meet social needs. With the rapid development of a market economy, practicality becomes the primary concern in writing research and education. People are more concerned about performance in writing modern practical genres, for example résumés, legislative documents, cross-cultural (business) letters (emails), agreements and contracts, public speeches, professional papers. Research and education about modern practical literacy becomes one of the major and urgent tasks for writing scholars. Also, since practical literacy on the mainland is different from that in Hong Kong, Macao, and Taiwan in terms of categorization, format, linguistic features, and cultural
Modern “writingology” in China 13 connotation, it might be necessary to unify the form and usage of Chinese characters as well as the format and style of writing, so as to meet the needs of all Chinese people across the globe. The second tendency is that “modern writingology” should keep abreast of the development of the times. This information age keeps pushing forward the progress of computer literacy. The use of the computer in writing has brought about changes in traditional modes of thinking, means of writing, ways of expressing, features of language, and formal patterns of writing. Traditional theories and frameworks of research need renewing as they can no longer provide good explanations for and correct guidance to actual writing practice. In addition, the importance of computer literacy is stressed more than ever before, and the fast development of the times has completely changed the idea of illiteracy. Those who could not read, write, and calculate used to be considered functionally illiterate, whereas since the 1980s, those who cannot drive a car, read in English, and write via computers are considered functionally illiterate. Therefore, writing research and education should satisfy the desire of both schoolchildren and people from all walks of life for learning how to write well via computers so as to enable them to keep abreast of the progress of modern society. The third is the dialectic “separate–merge” tendency, which is considered the general direction of the discipline. In this modern age, theories in all fields develop fast. Researchers have tried borrowing theories across disciplines in scientific studies. Multidisciplinary studies have blurred the division of separate disciplines. Psychology and linguistics are separate disciplines, but have merged into each other and formed psycholinguistics; economics and math are also considered two separate disciplines, but at present math–-economics and economics–math have also been developed. Likewise, many modern new branches have been continuously appearing in “writingology,” such as info-writingology, psycho-writingology, literary writingology, esthetic writingology, and rhetoric writingology. The appearance of these subdisciplines shows the “separate–merge” direction of “writingology” which results from continuous interdisciplinary studies.
Notes 1 My thanks go to the Visiting Fulbright Scholar Program for providing me funding and offering me the fortunate chance to work with Professor Charles Bazerman at UCSB. I am grateful to Charles Bazerman and Karen Lunsford, who encouraged me to write about this topic and have spent a lot of time reading all the drafts, making detailed comments and offering insightful suggestions. They made my stay at UCSB an incredibly rewarding experience for my career. 2 This chapter uses Arabic numerals to indicate tones of Chinese characters: “1” indicates the first tone (the level tone), e.g., “zhi1”; “2” indicates the second tone (the rising tone), e.g., “shi2”; “3” indicates the third tone, e.g., “xie3”; “4” indicates the fourth tone (the falling tone), e.g., “zuo4”; romanized Chinese without an Arabic numeral represents the slight tone, e.g., “de.” 3 The journal Writing dedicated a special column in the fourth and the fifth issues of 1984 to discussions about this topic.
14 Chen Huijun 4 Transitional works include those by Shangxi Teachers’ College (1973) and Hebei University (1979). 5 Part of the information comes from Du (2004).
References Chen, G. A. (1995). Xian4 dai4 xie3 zuo3 xue2 yuan2 li3 (Principles of modern writingology). Changsha: Central South University Press. Dong, X. Y. (2000). Xian4 dai4 xie3 zuo3 xue2 jiao4 cheng2 (Course of modern writingology). Beijing: High Education Press. Du, B. Ch. (1988a). Zhong1 xue2 sheng1 zuo4 wen2 xun4 lian4 ti3 xi4 he2 fang1 fa3 (System and methods of composition training in secondary schools). Changchun: Northeast Normal University Press. Du, B. Ch. (1988b). Da4 xue2 xie3 zuo4 ji1 chu3 (Basic college writing). Changchun: Jilin Education Press. Du, F. L. (2004). Zhong1 guo2 xie3 zuo4 xue2 li3 lun4 yan2 jiu1 yu3 fa1 zhan3 (Research and development of writing theories in China). Beijing: Central Compilation and Translation Press. Hong, T. G. (1986). Xian4 dai4 xin1 wen2 xie3 zuo4 ji4 qiao3 (Techniques of modern news writing). Beijing: China News Press. Hu, W. Sh. (1960). Xie3 zuo4 ji1 chu3 zhi1 shi (Basic knowledge on writing). Shanghai: Fudan University Press. Gao, N. (1991). Xie3 zuo4 si1 wei2 san1 ti2 (Three issues on writing thinking). In Ma Zhengping (ed.), Current progress of writing research in China. Hong Kong: Hong Kong New Century Press. Gao, Y., & Liu, F. F. (1989). Xie3 zuo4 xun4 lian4 ti3 xi4 san1 ji2 tiao4 (Three- jump training system of writing). Beijing: Guangmin Daily Press. Hebei University. (1979). Xie3 zuo4 ji1 chu3 zhi1 shi2 (Basic knowledge on writing). Shijiazhuang: Hebei Renmin Press. He, J. H. (1958). Xie3 zuo4 ji1 chu3 zhi1 shi2 jiang3 hua4 (Speech on basic knowledge on writing). Shanghai: New Knowledge Press. Jiang, R. S. (1986). Da4 xue2 ke1 ji4 xie3 zuo4 (Scientific writing in universities). Beijing: Guangming Press. Jin, Ch. M. (1996). Xian4 dai4 xie3 zuo4 xue2 ji1 ben3 yuan3 li3 (Basic principles of modern writingology). Tianjin: Tianjin People’s Press. Jin, Ch. M. (1989). Xie3 zuo4 yun4 si1 xue2 yin3 lun4 (Introduction to writing thinking theories). Xi’an: Shangxi Renmin Education Press. Jin, Ch. M., & Lin, K. F. (1991) Gao1 deng3 shi1 fan4 xie3 zuo4 jiao4 cheng2 (A course book of writing for advanced normal schools). Fuzhou: Fujian Education Press. Jin, Zh. B. (1999). Dian4 nao3 xie3 zuo4: fang1 fa3 yu3 ji4 qiao3 (Computer writing: Methods and techniques). Dalian: Liaoning Normal University Press. Kuang, Y. M. (1986). Xin1 wen2 xie3 zuo4 jiao4 cheng2 (Course on news writing). Guangzhou: Guangdong Education Press. Li, B. Ch. (1993). Chuang4 zuo4 ji4 qiao3 xue2 (Theories on techniques of writing). Huhhot: Inner Mongolia Education Press. Li, B. J. (1994). Kuai4 le4 zuo4 wen2: jiao4 yu3 xue2 (Happy writing: Teaching and learning). Beijing: Yu Wen Press. Li, J. L. (1982). Zuo4 wen2 fa3 gai4 yao4 (Summary of composition methods). Shengyang: Liaoning People’s Press.
Modern “writingology” in China 15 Li, K. Y. (1988). Gao1 deng3 ying1 yong4 xie3 zuo4 jiao4 cheng2 (Advanced course of applied writing). Beijing: Yuwen Press. Li, X. Y. (1987). Ying1 yong4 xie3 zuo4 xun4 lian4 jiao4 cheng2 (Course on training of applied writing). Dalian: Northeast Financial University Press. Li, Zh. L. (1988). Pu3 tong1 xie3 zuo4 xue2 jiao4 cheng2 (Course on writing for general purposes). Lanzhou: Lanzhou University Press. Lin, K. F. (1985). Ji1 chu3 xie3 zuo4 li3 lun4 (Basic theories of writing). Fuzhou: Fujian Renmin Press. Lin, K. F. (1991). Gao1 deng3 shi1 fan4 xie3 zuo4 jiao4 cheng2 (Advanced course on writing for teachers training programs). Fuzhou: Fujian Education Press. Lin, K. F. (2002). Xian1 dai4 xie3 zuo4 xue2: kai1 tuo4 yu3 geng1 yun2 (Modern writingology: Deploitation and tillage). Nanjing: Normal University Press. Liu, X. Q. (1985). Ji1 chu3 xie3 zuo4 xue3 (Basic writingology). Beijing: Central Broadcasting TV University Press. Ma, X. D. (1999). Xie3 zuo4 xun4 lian4 lun4 (Theories on writing training). Mudanjiang: Heilongjiang Karean Press. Ma, Zh. P. (1995). Xie3 zuo4 zhi4 hui4 lun4 (Wisdom of writing). Chongqing: Southwest Normal University Press. Ma, Zh. P. (2002). Gao1 deng3 xie3 zuo4 xue2 yin3 lun4 (An introduction to advanced writingology). Beijing: Renmin University Press. Miao, F. L. (1993). Zhong1 guo2 dang1 dai4 gong1 wen2 xie3 zuo4 (China contemporary administrative writing). Weihai: Shangdong University Press. Ning, Zh. Y. (1999). Si1 fa3 wen2 shu1 xue2 (Legislative writing theories). Beijing: China University of Politics and Law Press. Pan, X. H. (2000). Gao1 deng3 shi1 fan4 xie3 zuo4 san1 neng2 jiao4 cheng2 (Advanced course for teachers’ training programs on three basic writing abilities). Beijing: Renmin Education Press. Pei, X. Sh. (1986). Xian4 dai4 xie3 zuo4 xue2 liu4 ren2 tan2 (Six people’s talks on modern writing research). Chinese Guide, 4(1986). Pei, X. Sh. (1987). Xie3 zuo4 xue2 xin1 gao3 (New manuscripts on research of writing). Nanjing: Jiangsu Education Press. Peking University, Chinese Section. (1964). Xie3 zuo4 zhi1 shi2 (Knowledge on writing). Beijing: Peking University Press. Qiu, Sh. H. (1986). Si1 fa3 wen2 shu1 tong1 lun4 (General theories on legislative writing). Beijing: Qunzhong Press. Shanxi Teachers’ College. (1973). Xie3 zuo4 zhi1 shi2 (Knowledge on writing). Taiyuan: Shanxi Teachers’ College Press. Shao, B. J., & Ye, X. P. (1998). Ying1 yong4 xie3 zuo4 zue2 (Applied writingology). Lanzhou: Gansu Education Press. Si, Y. H. (1984). Ke1 ji4 xie3 zuo4 jian3 ming2 jiao4 cheng2 (Course on scientific writing). Hefei: Anhui Education Press. Sun, Sh. Zh. (1987). Wen2 xue2 chuang4 zuo4 lun4 (Literary writing). Shenyang: Chunfeng Literary Press. Tao, J. W. (1998). Xie3 zuo4 yu3 wen2 hua4 (Writing and culture). Shanghai: Shanghai Foreign Languages Education Press. Wang, J. Sh. (1985). Shi2 yong4 gong1 wen2 xie3 zuo4 jiao4 cheng2 (Course on practical administrative writing). Beijing: Beijing University Press. Wang, Zh. B. (2002). Er4 shi2 shi4 ji4 zhong1 guo2 xie3 zuo4 li3 lun4 shi3 (History of Chinese writing theories in the twentieth century). Nanjing: Nanjing University Press.
16 Chen Huijun Wei, T. J. (1998). Tan4 xun2 xie3 zuo4 de ao4 mi4 (Exploring secrets in writing). Nanjing: Nanjing Normal University. Wu, B. X. (1954). Xie3 zuo4 he2 yue4 du2 de ji1 ben3 wen4 ti2 (Basic issues in reading and writing). Shanghai: Oriental Book Publishing House. Wu, B. W., et al. (1985). Ji1 chu3 xie3 zuo4 jiao4 cheng2 (Course of basic writing). Taiyuan: Shanxi Renmin Press. Yang, Y. H. (1983). Wen2 zhang1 xie3 zuo4 er4 shi2 wu3 jiang3 (Twenty-five lectures on writing). Changchun: Jilin Renmin Press. Yin, J. Sh. (1984). Zhi4 li4 fa1 zhan3 yu3 xian4 dai4 xie3 zuo4 xue2 ke1 (Development of intelligence vs. modern writing discipline). Writing, 4. Yu, Ch. K. (1996). Xian4 dai4 ying1 yong4 wen2 (Modern applied writing). Shanghai: Fudan University Press. Yu, G. R. (1992). Jing1 ji4 xie3 zuo4 xue2 (Economics writingology). Beijing: China Economics Press. Zhang, Zh. G. (1952). Xie3 zuo4 fang1 fa3: cong2 kai1 tou2 dao4 jie2 wei3 (Writing methods: From beginning to the end). Shanghai: Kaiming Book Publishing House. Zhou, D. C. (1998) Xie3 zuo4 xun4 lian4 jiao4 cheng2 (Writing training course). Shanghai: Oriental Press Center. Zhou, J. Ch. (1989) Xie3 zuo4 xue2 gao1 ji2 jiao4 cheng2 (Advanced course of writing theories). Wuhan: Wuhan University Press. Zhou, S. J. (1994). Zhong1 guo2 xian4 dai4 gong1 wen2 xie3 zuog. Beijing: Knowledge Press. Zhu, B. Sh. (1986a). Xian4 dai4 xie3 zuo4 xue2 (Modern theories of writing). Beijing: People’s Daily Press. Zhu, B. Sh. (1986b). Xie3 zuo4 yu3 lun4 wen2 ping2 gai3 (Writing and composition assessment). Beijing: High Education Press.
2 The French didactics approach to writing, from elementary school to university I. Delcambre and Y. Reuter Université Charles-de-Gaulle-Lille3
“Didactics” in France refers to research disciplines that analyze content (knowledge and know-how) as the object of teaching and learning, related to school disciplines (Reuter, 2007a). But, when we use the term didactics, we are not referring to teaching/training disciplines. For example, the didactics of French is concerned with neither the same objects nor the same practices as “French” as a subject taught by teachers and learnt by learners, or which trainee teachers attempt to acquire at teacher-training colleges. In contrast to what Lee S. Shulman (1986) called “pedagogical content knowledge,” which is a particular type of professional knowledge linking certain aspects of content knowledge with questions of the teachability of that knowledge, didactics pursues a theory of this pedagogical content knowledge by empirical analysis of data collected in the classroom, construction of a body of specific concepts, and reflections on methodological problems (Lahanier-Reuter & Roditi, 2007; Perrin-Glorian & Reuter, 2006). The target of this research is thus descriptive or epistemological, rather than pedagogical or oriented toward teacher education, even though praxeological questions are both the point of departure and the horizon of expectations of didactics. In this chapter, we will present research being developed on French didactics in the specific domain of writing. The presentation will be based in the research that Yves Reuter has carried out for the past 30 years on teaching and learning writing, as well as the current research agenda directed by Isabelle Delcambre concerning university-level writing.1 The didactic approach to writing developed in France from the 1980s onward, essentially in relation to the discipline “French,” as it is explicitly responsible for the teaching of writing in primary and secondary schools. It was dominated in this period by cognitive psychology, psycholinguistics, and genetic psychology (Barré-de Miniac, 1995), but it progressively built itself as a specific research field. Today it focuses on interrogating the theoretical paradigms at its center, and begins to carry out a theoretical dialogue with those forms of sociology that focus on the description of variations in human behavior rather than on establishing their universal features (Lahire, 2007). The second part of this chapter will present an example of articulations between the didactics of writing and relevant psychological and sociological paradigms.
18 I. Delcambre and Y. Reuter Research about university writing developed much later, near the end of the 1990s, by French didacticians, following the broad movement of sociological research about universities. This research, beginning early in the 1990s, studied the “massification” and the transformation of the college-going public, their daily life and material conditions, their different relationships with the rhythm and timing of college studies, their study habits, and the mechanisms by which universities select students and reproduce inequalities (Rey, 2005). There is thus nothing in common between French didactics research on university writing, quite recent, and the much more established movement of composition theory, linked to the institutionalization of teaching writing in higher education in the United States (Donahue, 2008), a phenomenon that did not exist in France.
From the didactics of French to university literacies The coherence of the didactics of writing, as a subdiscipline of the didactics of French, comes from the analysis and description of teaching and learning writing at primary and secondary school, where “French” as a discipline contains, along with grammar, spelling, reading, literature, and so on, a learning object which is called “writing.” Universities are not the “natural territories” of the didactics of French: it is difficult to designate French as a discipline at university, given that French university professors are more likely to speak of Literature or Letters, Linguistics, Language Study, or Communications. The concepts and methods that have been developed in the didactics of French, however, are of interest in the study of university writing. Moreover, as many studies in didactics at school have shown, the teaching and learning of writing is not limited to French classes. Didacticians of French describe and analyze, often in collaboration with other didacticians, writing in disciplines other than French—for example, in science or mathematics.2 It should also be noted that didacticians from other disciplines appropriate the tools and methods elaborated in the didactics of French or in the field of language study in order to describe and analyze language practices (written or oral) in the classroom (science, mathematics, and so on). These relations between didacticians of different disciplines, when they investigate the role of language practices in learning and teaching, are nearly the same that can be found in secondary schools between teachers of different disciplines when they think about how to improve the students’ writing abilities (Reuter, in press). For all of the above reasons, the absence of French and French writing as a university discipline, the relative transversal nature of language practices (which must be, however, questioned), the contribution of French didactics to research on university writing, we propose to use the term university literacies to designate the field of research which analyzes written language practices at university.
The French didactics approach to writing 19
Research in the field of the didactics of writing The development of the didactics of French experienced a veritable explosion from 1985 with the theorization of educational activities and educational writing practices. The didactics of writing was mainly developed in the context of research on the teaching of French at primary school and the initial years of secondary school, and, to a lesser degree, in the final years of secondary school. These didactic studies were not only influenced by research in cognitive psychology on the processes involved in writing, but also by research in social psychology on representations associated with writing and writing tasks, which eventually led to the conception of the notion of how a student writer relates to writing, which was developed, for example, by Christine Barré-de Miniac (2000). Other studies on the history of the teaching of French and writing in genetic psychology or genetic criticism also played an important role in influencing the orientation of initial didactic studies. Rather than reviewing the history of this field, however, we present a didactic definition of writing, which is the foundation for the work THEODILE has been doing and is the outcome of theoretical and empirical research carried out by Y. Reuter in the past 30 years. We will try to show, first, how, in order to create a model for writing (as for reading; Reuter, 2003), didactics must include two antagonistic concepts, the first unifying writing at an abstract level, the second highlighting its contextualization and variations. We will then try to clarify the notion of contextualization in this framework, as well as its consequences for the specific difficulties that learners encounter in singular practical situations. Finally, we will develop the value of the notion of tension in order to clarify the ways in which writing practices work and the problems their use encounters, a notion which is important in theorizing texts and writing. Creating writing models in a didactic perspective It is possible to position the numerous writing models developed in different fields (Reuter, 1996) on a continuum between two extremes. The first tends toward association with the field of philosophy or certain strands of anthropology or psychology, and presents writing as singular, based on three principles: its unique nature, its autonomization (from other activities), and its abstraction-decontextualization (from the subject-writer’s history, from situations, from institutions). This conception, which we are of course oversimplifying to an extreme, is largely focused on seeking out invariants, and on assimilating writing into what is supposed to be most specific (e.g., encoding, with help from linguistic units; constitution of lasting traces; or the possible autonomization of the conditions for the situation of the utterance). This consequently renders secondary the other dimensions of writing and its possible variations in modes such as marginalization, superficiality, or “window dressing.”
20 I. Delcambre and Y. Reuter The other end of the continuum of conceptions about writing, which we are oversimplifying as much as the first, is based more on sociology, ethnology, or history. It is less concerned with writing (singular) than with practices (plural), writing practices or practices including writing. In this framework, writing is not reduced to the moment of transcribing (the writing itself) but structurally integrates a before and an after that determine the conditions of possibility, the functions, the uses, the meanings, and the values. Writing can only be understood as inscribed in the history of the subject and articulated with the writer’s overall practices. In addition, writing is defined by its modes of actualization, of which different components carry meaning within social spaces (whether of the writer him/herself, the situation, the procedures used, the categories of writing . . .). The concepts associated with this end of the continuum are thus grounded in principles like diversity, contextualization, and the interaction between activities within practices and actualization. These conceptions thus seek out the relevant axes of variation and hold that whatever is designated by the term “writing” cannot be reduced to that which is specific to it. In other words, the variations restructure, in a key way, the set of components. In a didactic framework, researchers need both ends of the spectrum because each one illuminates and can help the field to develop modes of teachability and learnability. The first conception is that writing (in the singular) tends to function congruent to the representations of teaching and learning according to which competence is built independently of the actual use or doing of an activity: writers learn before doing, prerequisites are necessary, activities can be broken down into component parts that can each be taught and practiced specifically. The issue of “usefulness” or function is left to a future time and attention stays focused on the analysis of forms. In the same way, writing—when articulated based on a neutralized vision of language (understood as shared capital)—is taken on in successive stages that should go from the most simple to the most complex elements, all of these elements themselves categorized in abstract, decontextualized, neutral ways (as, for example, in text types). This conception seems, to us, to align in perfect harmony with the demands of the school form, in the historico- sociological sense of the term (Vincent, 1994), enabling the establishment of a progressive, common curriculum. The second conception constructs competence as depending on the activity actually occurring: we learn by doing and in order to do, constructing, in this doing, the necessary dimensions of the activity, which is structured by its uses. In this framework, practices and their meanings have a central role, and formal reflections are subordinate to them. Writing—as articulated in a social vision of language practices (as a space of conflict and negotiation)—is taken on from the start in diversified and complex forms of realization, characterized more by their relationship to contexts, to institutions, to social activities which frame the genres and discourses. This second conception, less anchored in
The French didactics approach to writing 21 school than the first, can only be maintained as a pedagogical approach insofar as it responds to important school preoccupations, such as by offering a palette of exercise forms and teaching and learning practices, based on subject areas, or modes of pedagogical work, or difficulties encountered. Analyzing contextualization in a didactic framework The second consequent thesis clarifies how didacticians should think about contextualization in a didactic framework. It seems to us that, in such a perspective, contextualization can be described through three interconnected systems: the school system (in the sense that Vincent, 1994 presents the school form), the pedagogical system, and the disciplinary system. Writing practices are thus rendered functional, either as tools or as objects, essentially in relation to teaching, learning, and evaluation, and are shaped in a way that makes them as easily managed as possible in school spaces (determining temporality, spatiality, tools, media, length, genres, and so on). These writing practices are, in a complementary way, structured by a kind of pedagogy, what Marcel Lesne (1977/1994) calls a pedagogical work mode, which will weigh heavily on their frequency, their form, their modes of connection, and their modes of evaluation. In addition, these writing practices are inscribed in/with disciplines (and at particular points in a disciplinary curriculum), which makes them more or less central disciplinary objects, or disciplinary tools in relation to different practices (analyzing, commenting, observing, experiencing) and with different genres specific to those disciplines. This contextualization confers, in any case, specificity to different school practices, differentiating them from extra- scholastic practices that are shaped by other systems in relation to the socio-institutional spheres in which they are used (Reuter, 1996). The importance we are assigning to contextualization, as well as this way of constructing it, has significant consequences in terms of both apprehending students’ difficulties and intervening in them. In fact, in this perspective, difficulties cannot be considered writing difficulties “in general,” but problems in carrying out specific practices, in order to do something with a particular context. This is how Michel Brossard (1998/2002) was able to show that many difficulties are born of a lack of cognitive clarity or, in another shape, difficulty overcoming the opacity of the context of formal learning situations for many students. In the same way, Bernard Lahire (1993) or Bernard Charlot, Elisabeth Bautier, and Jean-Yves Rochex (1992) have shown that a good number of students’ problems, in particular students from lower-class backgrounds, can be attributed to their different relationships to language or to learning, in contrast with those that are more established by classic pedagogico-didactic functioning (decomposition, attention paid to form rather than meaning, analytic distance, progressivity, and so on). Thus Jean-Paul Bernié and his Bordeaux colleagues (Bernié, 2002) were able to develop the hypothesis that certain
22 I. Delcambre and Y. Reuter difficulties are caused by students’ difficulty moving from everyday, current spaces into disciplinary modes of thinking, acting, and discourse. In a way particularly interesting for a didactic perspective, considering these problems as strongly related to a context and to specific practices allows didacticians, we believe, to safeguard against fatalism by affording real power to educational intervention in terms of seeing these difficulties coming or in remediating them, because it is within a teacher’s power to transform, meaningfully, contexts and practices and to introduce learners to them. The value of the notion of tension in a didactic perspective The third and final thesis we propose here—based on many of our research projects about texts, reading, and writing, from kindergarten to university—seeks to clarify the value of the notion of tension. This is thus one more step, in a didactic perspective, toward formalizing the notion of practices (in this case, writing practices) and toward specifying the difficulties that come with teaching them and learning them. By tension we mean that language activities and practices, textual objects, relationships to these objects, and the means available for teaching (understood here in the broad sense of situations, exercises, prescriptions) that seek to establish them are all structured by antagonistic (opposite) ends of the spectrum. Writing, its teaching and learning, can thus be conceived of as the management of tensions in variant forms according to situations and genres of writing. This necessitates key theoretico-methodological shifts in order to describe activities and objects and formalize the principle tensions that structure them. For example, in writing there are the tensions between phonographemes and logographemes in written codes, and tensions between transcription of pre-existing thought and construction of thought in the rendering of meaning into text. In writing summaries, there is a tension between fidelity to the original text and autonomy of the new text. In representation of writing, there are tensions between protecting one’s intimate self and showing that self (Dabène, 1987).3 Reuter has thus been able to show how many difficulties in descriptive writing (Reuter, 1998) or narrative writing (Reuter, 2007b) in school settings were due to tensions between the manner of imagining these language behaviors in different disciplines or even in the same discipline at different points: description in mathematics, for example, seeks economy of expression as compared to description in French, and the “story,” in French elementary school, can be satisfactorily constructed in a simple chronological way but must, later in schooling, be complementarily structured by other types of logical relationship. He has also been able to show (Reuter, 2004) how many problems in writing university-level texts (MA-level theses or PhD dissertations), what he calls research writing in evolution, were due to a structural tension in the writer’s enunciative positions, between researcher (who must, in this role, produce new knowledge for his or her
The French didactics approach to writing 23 community) and person-in-formation (who must, in this role, show that he or she has adopted the norms and knowledge of his or her community as well as demonstrating his or her status as learner, ready to be evaluated as such). In this perspective, difficulty can come from a source other than a “lack” (in the learner) or different relationships with school practices. The difficulties are—at least partially—programmed by the very structure of the activities and objects in question, as deviations, hyperaccentuations, or imbalances. Using this approach, didacticians can analyze problems in teaching, when teachers, for example, hesitate to construct multiple dimensions of writing at the same time or redistribute them throughout a school trajectory. They can also identify difficulties in learning arising from an unwanted transfer of competences from one discipline to another (describing in mathematics like in French) or an obscuring of acquired competence when a student goes from one discipline to another. To summarize, we have offered two models of writing (singular activity or plural practices), the three spaces of contextualization for writing (school as institution, pedagogy, and disciplines), and the concept of tension. In the later section of this chapter, we will make a wide presentation of French studies about writing at university as examples of the methodological and theoretical problems didactics has to face.
French research in the area of university literacies The stakes of the new field of research that we propose to call university literacies are to explore what the sociological studies mentioned in the introduction generally leave aside, that is writing practices, and the link between these practices and disciplinary or epistemological frames in which these practices are exercised. When sociologists focus on study practices, they focus largely on reading practices (Fraisse, 1993; Lahire, 1998), but little on the writing practices. And the sociologists have not focused on the disciplines (Alava & Romainville, 2001). Thus, the aim of didactic studies in the field of university literacies is to designate the disciplines as a framework for analyzing student practices and representations and to tackle writing practices, that is to say the practices involved in discursive production which punctuate the program and which can be seen as steps in the comprehension of the implicit codes and modes of disciplinary reflection. With regard to French universities, the aim would be to better fill the program contained within the use of the plural form of “university literacies,” which, as Mary Lea (2008) points out regarding academic literacies, encompasses both reading and writing, seen as both social and cultural practices, situated in specific contexts and not as individual cognitive activities. However, the field of French research on academic literacies is still heterogeneous and reflects a number of contrasts which can be linked to either university pedagogical practices or to trends in research in the
24 I. Delcambre and Y. Reuter didactics of writing. There is, in fact, quite a large difference between two ways of approaching writing at university. On the one hand, we have approaches based on “production techniques,” conceptions which dominated in the 1970s, that envisage a transversal view of the teaching of writing, without referring to individual disciplines—teaching writing transversally is seen as a prerequisite to academic study or as a means of bringing the student up to the required level, based on mastery of language and communication skills (the linguistic deficiency model). These techniques are mainly viewed in terms of compensating or remedying. They are mainly aimed at students in the first year. This model is always present when there is a desire to find solutions to tackle drop-out in the first year of university (as in the case of the recent Réussir en Licence (“success as an undergraduate”) plan). On the other hand, since the end of the 1990s, researchers in the field of French didactics have been working on analyzing and describing difficulties students face when they write, be these difficulties related to academic, professional, or research writing. For example, Marie-Christine Pollet and Françoise Boch (2002) introduce a collection of papers presented at the first conference on “Writing in Higher Education” in Brussels by underlining that interest in the written production of students is currently taking over from the numerous studies carried out on reading. They further argue that it is necessary to open the field to all forms of higher education and not only university education (which gives rise to numerous papers on writing in professional training). Isabelle Laborde-Milaa, Françoise Boch, and Yves Reuter (2004) emphasize the emergence of a separate focus on research writing, a form of theoretical or theorizing writing with which advanced students are confronted. This form of writing is both the pinnacle of the students’ studies and “structures the identity of the institutions and its actors” (p. 5). It is thus the object of “anxiety-inducing or misleading representations,” which should be taken into account when dealing with training students for research and understanding the heuristic functions of writing. These theoretical discourse-focused analyses examine the articulation between a subject who is the writer (how this person relates to writing, how he/she interprets the writing context) and the characteristics of the disciplinary genre. Epistemological analysis of the relationship between writing and disciplines plays a central role in this approach (Daunay, 2008). In order to present a panorama of research in this area, we can mainly refer to these two publications, without excluding reference to other works. Discourse genres Most of the studies focus on textual analysis of the writing genres produced in higher education (including teacher training): academic genres
The French didactics approach to writing 25 required by the institution (dissertations, text commentaries, theoretical texts, reports written after internships, case studies, abstracts, back-cover texts), or genres linked to research (proposing an oral presentation). Sometimes they describe different parts of the text (introductions, annexes, research questions), or personal writing practices linked to work (taking notes, rewriting). Most of the time, these descriptions are linked to specific questions: for example, problems related to referencing and referring to sources (Boch & Grossmann, 2001) or reformulating the discourse of someone else (Delcambre, 2001); the traces of the position of the writer, in other words the image that the writer constructs of him/herself as an enunciator, which in turn shows a variability in the degree of mastery of the implicit norms (Delcambre & Laborde-Milla, 2002; Laborde-Milla, 2004). These analyses of university genres are grounded in broad linguistics research questions about discursive heterogeneity but insist on linking the formal aspects of text to questions of writerly identity constructed in the writing. Certain research projects use the notion of writerly images (Delcambre & Reuter, 2002a), which allows scholars to construct, in the form of a system, the dimensions of the subject-writer, those of the text as an object of analysis, and those of the interpretative activity of the reader- receptor. This approach opens interesting perspectives in analyzing writing during training, in that it allows the articulation of an analysis of tasks and written production, the interpretation of difficulties faced by the writer, and the interpretation of the support given by the lecturers. The attention devoted to the description of texts students have to write is strongly linked to the construction of specific knowledge on research writings which researchers in Grenoble are undertaking as part of a separate ANR program which is running in parallel to our program.4 Different components of a research article, mainly in the field of language study, have thus already been described: the theoretical framework (Boch, Rinck, & Grossmann, in press), the modes used to refer to others (Grossmann, 2002), the enunciative position adopted (Rinck, Boch, & Grossmann, 2007), and the construction of an “author figure” (Rinck, 2006). This precise description of the writing genre being researched, along with its variations, allows researchers, during the analysis of the students’ productions, to avoid falling into the trap of analyzing according to their own representations, and to question the manner in which the students’ writing constitutes a type of research writing. It also allows researchers to investigate how students’ writing can contribute to an evolution in research writing. Relationships to writing Other studies explore the manner in which students relate to writing in different contexts (exams, dissertations, dossiers). These have been described using different methodologies—solicitation of memories related to writing (Daunay & Reuter, 2002), semi-structured interviews (Delcambre &
26 I. Delcambre and Y. Reuter Reuter, 2002b), questionnaires (Reuter, Ruellan, Genes, & Picard, 2000), or even analysis of written production (Laborde-Milaa, 2002). The concept of “relationship to” has been constructed by education researchers and draws on sociological and psychoanalytic perspectives about learning (Charlot, Bautier, & Rochex, 1992). It allows researchers to bring together a whole series of components of the act of learning, bringing them together with the question of the subject: relationships to content, to the activity, to the situation, to others, to knowledge in general, to language, to oneself, and to one’s image of oneself as a learner. This concept, in didactics, indicates the relation (cognitive but also socio-psycho-affective) in which the learner engages with the content and which conditions in part their learning: a relationship to content that does not correspond with the one school envisions can increase the difficulty of a student’s access to the subject being taught. (Reuter, 2007a, p. 191) Barré-de Miniac (2000) describes the relationship to writing as simultaneously the subject’s investment in the writing (the affective interest in writing), the opinions and attitudes about this (that result in social representations of writing, its functions, its difficulties), and the conceptions of writing and learning to write (most notably in the opposition between writing as transcription of a pre-elaborated content and writing as the construction of thought). In terms of research about university writing, the aim of a significant number of studies is to specify what students understand about academic or research writing and the problems that it poses for students during their studies (Reuter, 1998; Reuter, 2004): the obstacle posed by the assumed transparency of what is written, the desire to avoid stylistic writing, the objectivity that research writing supposes, one’s self-image. These issues highlight a fundamental tension between the researcher and the person being trained (cf. above). Thus, the researcher-in-training must find the right balance between an overly low or overly high position, for example in the presentation of theories or of theorists in one’s field: how can a student avoid giving the impression that he or she is evaluating expert researchers when presenting the value or limits of their work for the project in which he or she is involved? How can the student show understanding of the experts without offering oversimplified explanations that imply the reader does not know anything? How can the student find the right balance between too many references (which can make the text almost unreadable or uninteresting) and an insufficient number of references (which gives the impression that the student has not done enough reading or has done it haphazardly)? And so on (Reuter, 2004, pp. 22–26). The notion of tension is essential in order to account for several obstacles that the student must overcome in learning to write at university.
The French didactics approach to writing 27 In the same vein, Anne-Marie Jovenet (2004) has analyzed, from a psychoanalytical perspective, these tensions based on questions related to the relationship between the subject-writer and the research object in question and the conflicts between the desire to write something about oneself and the unconscious barrier raised by the need to respect university norms. Links between writing and disciplines The last collection of studies on university writing I will present attempts to establish links between difficulties faced by the students and the specificities of individual disciplines. Until now, these links have mainly been studied in terms of language practices or semiotic practices, related to description in the first case and to tables in the second case. In a collective publication, Reuter (1998) explores the different forms and the different functions that descriptions can take, according to the discipline in question (linguistics, medicine, psychology, sociology). Reuter specifies, based on a descriptive model, variations in the forms and the functions of description, according to the type of writing and the disciplines (at school or at university). In her particular contribution to this study, Delcambre (1998) shows how the discipline to which the trainee teachers belong (arts, biology, or mathematics) orients their descriptive writing in their professional dissertations: we thus see descriptions that attempt to objectify in scientific dissertations (using quantitative, statistical tools), compared with realistic descriptions in the dissertations of arts students (creating portraits). For her part, Dominique Lahanier-Reuter (2003a, 2003b, 2003c) has examined writing in the teaching of statistics, particularly by comparing writing tables, graphs, and “linear writing” in two different educational contexts: teaching in traditional degree courses in science or mathematics and teaching in the field of education sciences. She shows the important variations between these two contexts in terms of writing norms and writing and reading practices. Carrying forward the work of these studies, our ANR research program aims to explore, with relation to five disciplines of the social sciences and humanities, the disciplinary aspects of university writing. Reflecting upon the epistemological dimensions of university disciplines, their reference frames, and theoretical models of writing is one way in which we hope to contribute to an understanding of the difficulties students face and the different manners in which we can accompany them in their writing at the university. Between sociologists’ attention to university reading practices and didacticians’ attention to writing in primary and secondary school, there is thus an opportunity for this study to build upon the initial didactical studies, by examining the relative diversity or homogeneity of university writing practices (which depend on the course of study) and the relative importance of the discipline with regard to writing practices used in the construction of knowledge.
28 I. Delcambre and Y. Reuter
Notes 1 A three-year project, funded by the French National Research Agency (ANR) called “University Writing: Inventory, Practices, Models” (ANR-06-APPR-019). 2 The French team THEODILE has undertaken many studies about these relations between writing and school disciplines. 3 The concept of “representation” in France comes from the field of social psychology. It is used in didactics for designating modes of knowing that are different from scientific concepts: knowledge drawn from individuals’ experiences. In teaching situations, this knowledge often surfaces as preconceived ideas that actually obstruct learning. Representations can concern objects, processes, or activities, and thus researchers can use the term “representations” of writing to name the ideas that people construct about writing processes, writing’s functions, its objectives, and so on. 4 SCIENTEXT “Un corpus et des outils pour étudier le positionnement et le raisonnement de l’auteur dans les écrits scientifiques.”
References Alava, S., & Romainville, M. (2001). Les pratiques d’étude entre socialisation et cognition. Revue Française de Pédagogie, 136, 159–180. Barré-de Miniac, C. (1995). La didactique de l’écriture: Nouveaux éclairages pluridisciplinaires et état de la recherche. Revue Française de Pédagogie, 113, 93–133. Barré-de Miniac, C. (2000). Le rapport à l’écriture. Aspects théoriques et didactiques. Villeneuve d’Ascq: Presses universitaires du Septentrion. Bernié, J.-P. (2002). L’approche des pratiques langagières à travers la notion de “communauté discursive”: Un apport à la didactique comparée?. Revue française de pédagogie, 141, Vers une didactique comparée, 77–88. Boch, F., & Grossmann, F. (2001). De l’usage des citations dans le discours théorique: Des constats aux propositions didactiques. Lidil, 24, Pratiques de l’écrit et modes d’accès au savoir dans l’enseignement supérieur, 91–112. Boch, F., Rinck, F., & Grossmann, F. (in press). Le cadrage théorique dans l’article scientifique, un lieu propice à la circulation des discours. In L. Rozier & M.-A. Paveau (Eds.), Actes du Colloque Cit-dit, Québec, September 2007. Brossard, M. (1998/2002). Approches socio-historiques de situations d’apprentissage de l’écrit. In M. Brossard & J. Fijalkow (Eds.), Apprendre à l’école: Perspectives piagétiennes et vygotskiennes. Bordeaux: Presses Universitaires de Bordeaux, 37–50. Charlot, B., Bautier, E., & Rochex, J.-Y. (1992). Ecole et savoir dans les banlieues et ailleurs. Paris, France: Armand Colin. Dabène, M. (1987). L’adulte et l’écriture. Contribution à une didactique de l’écrit en langue maternelle. Brussels: De Boeck-Wesmael. Daunay, B. (2008). The evolution of the French field of Didactique de l’écrit: Theorizing the teaching practices of writing in the disciplines. In Y. Reuter & C. Donahue (Eds.), Disciplines, language, activities, cultures: Perspectives on teaching and learning in higher education from France and the United States. Amsterdam: Kluwer Academic Publishers. Daunay, B., & Reuter, Y. (2002). Le rapport à l’écriture. Etude exploratoire au travers de souvenirs sollicités. Cahiers THEODILE, 2, 3–38. Delcambre, I. (1998). Parler d’une classe (la décrire?) dans un mémoire professionnel. Pratiques, 99, La description, 89–103.
The French didactics approach to writing 29 Delcambre, I. (2001). Formes diverses d’articulation entre discours d’autrui et discours propre. Analyses de commentaires de textes théoriques. Lidil, 24, Pratiques de l’écrit et modes d’accès au savoir dans l’enseignement supérieur, 135–166. Delcambre, I., & Laborde-Milaa, I. (2002). Diversité des modes d’investissement du scripteur dans l’introduction du mémoire professionnel. Enjeux, 53, L’écrit dans l’enseignement supérieur, I, 11–22. Delcambre, I., & Reuter, Y. (2002a). Images du scripteur et rapports à l’écriture. Pratiques, 113–114, Images du scripteur et rapports à l’écriture, 7–28. Delcambre, I., & Reuter, Y. (2002b). Le rapport à l’écriture d’étudiants en licence et maîtrise: Première approche. Spirale, 29, Lire-écrire dans le supérieur, 7–28. Donahue, T. (2008). Ecrire à l’université. Analyze comparée en France et aux Etats-Unis. Villeneuve d’Ascq: Presses universitaires du Septentrion. Fraisse, E. (Ed.). (1993). Les étudiants et la lecture. Paris: PUF. Grossmann, F. (2002). Les modes de référence à autrui: L’exemple de la revue “Langages.” Faits de langue, 19, 255–262. Jovenet, A.-M. (2004). Tensions et écriture de recherche. Pratiques, 121–122, Les écrits universitaires, 143–156. Laborde-Milaa, I. (2002). Le passé-simple dans des rapports de stage: Tentatives d’explication. Le Français aujourd’hui, 139, Les verbes, de la phrase aux discours, 97–107. Laborde-Milaa, I. (2004). Auto-reformulation et investissement du scripteur: Abstracts et quatrièmes de couvertures de mémoires de maîtrise. Pratiques, 121–122, Les écrits universitaires, 183–198. Laborde-Milaa, I., Boch, F., & Reuter, Y. (2004). Présentation. Pratiques, 121–122, Les écrits universitaires, 3–8. Lahanier-Reuter, D. (2003a). Tableaux et parcours de lecture. Annales de didactique et sciences cognitives, vol. 8, 69–83. Lahanier-Reuter, D. (2003b). Tableaux apparaissant dans des textes à visée formatrice. Spirale 32, L’organisation visuelle des tableaux, 33–46. Lahanier-Reuter, D. (2003c). Différents types de tableaux dans l’enseignement des statistiques. Spirale, 32, L’organisation visuelle des tableaux, 143–154. Lahanier-Reuter, D., & Roditi, E. (Eds.). (2007). Questions de temporalité, Les méthodes de recherche en didactiques 2. Villeneuve d’Ascq: Presses universitaires du Septentrion. Lahire, B. (1993). Culture écrite et inégalités scolaires. Sociologie de l’ “échec scolaire” à l’école primaire. Lyon: Presses Universitaires de Lyon. Lahire, B. (1998). Matrices de socialisation et lectures étudiantes. Bulletin des bibliothèques de France, 43(5), 58–61. Lahire, B. (2007). La sociologie, la didactique et leurs domaines scientifiques. Education & Didactique, 1(1), 73–82. Lea, M. (2008). Academic literacies in theory and practice. In B. V. Street & N. H. Hornberger (Eds.), Encyclopedia of language and education, vol. 2: Literacy (2nd ed., pp. 227–238). Dordrecht: Kluwer. Lesne, M. (1977/1994). Travail pédagogique et formation d’adultes. Paris: L’Harmattan. Perrin-Glorian, M.-J., & Reuter, Y. (Eds.). (2006). Les méthodes de recherche en didactiques. Villeneuve d’Ascq: Presses universitaires du Septentrion. Pollet, M.-C., & Boch, F. (2002). Introduction. Enjeux, 53, L’écrit dans l’enseignement supérieur, I, 5–8.
30 I. Delcambre and Y. Reuter Reuter, Y. (1996). Enseigner et apprendre à écrire. Construire une didactique de l’écriture. Paris: ESF. Reuter, Y. (1998). De quelques obstacles à l’écriture de recherche. Lidil, 17, Pratiques de l’écrit et des modes d’accès au savoir dans l’enseignement supérieur, 11–23. Reuter, Y. (Ed.). (1998). La description. Théories, recherches, formation, enseignement. Villeneuve d’Ascq: Presses universitaires du Septentrion. Reuter, Y. (2003). La construction de la lecture en didactique. In E. Eggert (Ed.), Mobiles et mouvements pédagogiques. Un choix d’itinéraires offerts à Jacques Weiss. Neuchâtel: IRDP, 37–43. Reuter, Y. (2004). Analyser les problèmes de l’écriture de recherche en formation. Pratiques, 121–122, Les écrits universitaires, 9–27. Reuter, Y. (Ed.). (2007a). Dictionnaire des concepts fondamentaux des didactiques. Brussels: De Boeck. Reuter, Y. (Ed.). (2007b). Pratiques, 133–134, Récits et disciplines scolaires. Reuter, Y. (in press). Interroger l’appareillage conceptuel de la didactique du français. La Lettre de l’AIRDF. Reuter, Y., Ruellan, F., Genes, S. & Picard, C. (2002). A propos de la fonction cognitive de l’écriture en formation. In D. G. Brassard (Ed.), Pratiques de l’écrit et modes d’accès au savoir dans l’enseignement supérieur. Villeneuve d’Ascq: CEGES, Université Charles de Gaulle, Lille 3, 45–52. Rey, O. (2005). L’enseignement supérieur sous le regard des chercheurs. Retrieved August 20, 2008, from www.inrp.fr/vst/Dossiers/Ens_Sup/sommaire.htm. Rinck, F. (2006). Ecrire au nom de la science et de sa discipline: La figure de l’auteur dans l’article en sciences humaines. Sciences de la Société, 67, 95–112. Rinck, F., Boch, F., & Grossmann, F. (2007). Quelques lieux de variation du positionnement énonciatif dans l’article de recherche. In P. Lambert, A. Millet, M. Rispail, & C. Trimaille (Eds.), Variations au cœur et aux marges de la sociolinguistique. Mélanges offerts à Jacqueline Billiez. Paris: L’Harmattan. Shulman, L. S. (1986). Those who understand: Knowledge growth in teaching. Educational Researcher, 15, 4–14. Vincent, G. (Ed.). (1994). L’éducation prisonnière de la forme scolaire? Scolarisation et socialisation dans les sociétés industrielles, Lyon: Presses Universitaires de Lyon.
3 What factors influence the improvement of academic writing practices? A study of reform of undergraduate writing in Norwegian higher education Olga Dysthe University of Bergen, Norway
Introduction This chapter offers a glimpse into the practices of writing in higher education in a European country and describes the influence of macro-level educational policies on writing in undergraduate studies. The context is a major educational reform in Norway called the Quality Reform implemented in the higher education sector from 2002. The Quality Reform affected the structure, financing, and leadership of higher education as well as the pedagogy. The reform is closely connected with the “Bologna process,” the most important effort ever made to align higher education systems in all European countries.1 Periods of reform offer a particularly interesting time setting for research because cultural traditions and established practices are then being challenged, lines of division among faculty become visible, and underlying views of teaching and learning come to the fore, even in academic cultures where this is not usually on the agenda. Norwegian universities, like most continental European universities, previously required very little undergraduate student writing and mainly assessed students on traditional sit-down exams. The Quality Reform did not explicitly deal with this situation, but nevertheless one of the pedagogical consequences of the reform was that virtually all courses now include compulsory student writing assignments and teacher feedback. The empirical basis for this chapter consists of two nationwide survey studies of higher education in Norway in which I have been involved and two case studies, one in history and one in law. The most comprehensive of the surveys is the research evaluation study of the Quality Reform while the second is a study of portfolio practices after the reform. The following overarching questions are discussed, drawing on sociocultural theory perspectives: (1) To what extent has the Quality Reform resulted in change in writing and feedback practices? (2) How can these changes be explained in
32 O. Dysthe relation to other aspects in the reform like changes in assessment (i.e., portfolios)? (3) What other factors have contributed to the change? In order to provide a context for the readers I will focus primarily on the reform as this is essential to understand the significance of the findings and the discussion.
The context Norway is a country with only 4.5 million inhabitants. It has six state universities, five scientific colleges, 25 state colleges and 26 private colleges. The majority of students are in the state system. The Quality Reform resulted in considerable financial and structural changes. The new study structure represented a radical break with Norwegian higher education traditions. The common structure proposed for all European countries, three- year Bachelor’s + two-year Master’s + three-year PhD, was implemented in Norway. This meant a reduction in study time from four to three years for Bachelor’s students, a dramatic loss of time given that the reform ambition was to improve the overall study quality. A new European credit point system (ECTS)2 was introduced and our grading system changed from a very detailed numerical scale to a letter scale (ABCDEF).3 All courses were modularized, for most courses this means one-third or half a semester (10 or 15 ECTS). Because a major purpose of the Quality Reform was to make studies more effective in the sense that students finished within the standard time frame, three pedagogical issues were particularly highlighted in the reform documents: (1) more use of student active teaching methods, (2) closer follow-up of each student and regular feedback, (3) closer connection between teaching and assessment, and more variation in assessment (portfolio assessment was mentioned in the reform documents as an example). While most top-down educational reforms have very little impact on classroom pedagogy (Cuban, 1990, 1993, 2002), this reform resulted in substantial pedagogical changes (Dysthe, Raaheim, & Lima, 2006a). In particular, the three clearly formulated pedagogical expectations from the Ministry of Education combined with the changes in the study structure altered some very basic conditions for academic writing in Norway.
Theoretical perspectives From a sociocultural theory perspective, writing, feedback, and learning practices are deeply embedded in academic, institutional, and disciplinary traditions. Changes in educational practices, including writing practices, can only be understood by studying the tacit and explicit macro-level expectations and decisions and analyze the complexity of structural and cultural characteristics of our higher education at different levels. This is the underlying perspective of this chapter, but since the relationships are very complex, I just offer some tentative answers. By “macro level” I first
What factors improve academic writing? 33 of all mean the political policy level. This is particularly important in Norway because higher education has always been under more rigid national regulation than in the United States. “Meso level” is the domain where bottom-up meets top-down; in our case, the local educational institution with its departments and programs. The macro level of academic communities of practice included strong academic traditions, some of which are international and some more local. As in most of Europe, universities were traditionally rather elitist and students came well trained in writing from secondary school. Thus, explicit teaching of writing was not deemed necessary. As universities were opened to a broader range of students, many lacked the necessary writing competence. At the same time, the Quality Reform in Norway has increased the demands on students’ writing and the logical follow-up would be to introduce explicit teaching of writing, but the traditional mindset is still so strong that positions for writing teachers are very difficult to fund.
Materials and method The present study makes use of three data sources. The most important is a major evaluation study of the Quality Reform, which consisted of a national survey, eight institutional case studies, and a series of student surveys. The purpose of the national survey was to get quantitative measures of the consequences of the Quality Reform as experienced by the teachers. The survey consisted of 62 questions and was sent out to a sample of professors, associate professors, and lecturers in all higher education institutions in Norway. The number of respondents were 2,060 and the response rate 60.3 percent. As a member of the research team, I interviewed leaders, teachers, and students in the eight institutions, four universities, three university colleges and one scientific college.4 The second data source is a nationwide survey study of portfolio practices in higher education based on a randomized selection from all public universities (5) and university colleges (22). The respondents were professors responsible for portfolio-assessed courses. An earlier version was conducted in one major university and three university colleges, and was published in Assessing Writing (Dysthe, Engelsen, & Lima, 2007). The results from these two surveys are very similar and this reinforces the validity of the findings. The portfolio survey data are very relevant to my topic for three reasons. First of all, the portfolios in all disciplines consisted of written texts. Second, the portfolio texts were assessed and thus taken seriously by students. Third, portfolios were introduced in the reform documents and got a lot of attention from students as well as teachers as something new. The third data source is two case studies from the University of Bergen, one in the Humanities Faculty and one in the Law Faculty. These give more specific insight into two university sites after the Quality Reform. All these studies have either been published or are forthcoming separate
34 O. Dysthe a rticles, but the present study is the first effort to draw together this comprehensive research data.
Findings The Quality Reform resulted in more and compulsory undergraduate student writing The evaluation study of the Quality Reform clearly documented three major pedagogical changes: (1) more diversified assessment practices, (2) more compulsory student writing, and (3) more feedback and closer follow-up of students (Dysthe et al., 2006a; Michelsen & Åmodt, 2006). These three areas of change are clearly interrelated, as will be discussed below. Interestingly, increasing the amount of student writing had not been stated as a specific goal in the reform documents, but this turned out to be one of the clearest findings. We do not have data to document the exact increase in student writing after the Quality Reform, but the survey data combined with the interviews give a fairly clear picture of increased compulsory student writing. Of all the respondents in the survey, 71 percent report great or consider able changes in assessment. There is no significant difference between universities and university colleges in this respect. A greater number of smaller written assignments are reported by 32 percent of these. Portfolio assessment is reported by 37 percent as the biggest change (Table 3.1). This means that of the 71 percent who have changed assessment practices (i.e., small assignment combined with tests or portfolios, or projects combined with or instead of final exams), a total of 81 percent have instigated changes involving more compulsory student writing. The data from our case studies show that there has been a quite substantial change in all the departments included in our study in the direction of compulsory student papers, and that there are no significant differences between universities and university colleges. There is a close connection between changes in assessment and increase in writing. From Table 3.1 we see that 36 percent of those who changed assessment introduced portfolios, and 13 percent projects, both of which involved writing during the course and not just for the final exam. Altogether 32 percent have introduced continuous assessment in the form of a series of smaller assignments that were graded separately and added up to a final mark, but this figure includes short tests, not just written assignments. Many of the interviewed teachers saw the increase in student writing as a way of achieving more “student active learning,” one of the key pedagogical concepts and goals of the Quality Reform. Compulsory writing assignments meant, according to the teachers who were responsible for the courses that all students engaged actively in the content of the subject instead of just listening to teachers and fellow students. Thus writing could
What factors improve academic writing? 35 Table 3.1 Use of portfolios and projects after the reform To those who answered great or medium changes University University Total in assessment: Which of the following has colleges changed most? (%) More smaller types of assignments/tests during the course of study
24
36
32
Portfolio assessment combined with a big final exam
28
16
21
Portfolio assessment instead of a big final exam
19
13
15
Projects assessment combined with a big final exam
9
10
9
Projects assessment instead of a big final exam
4
4
4
Fewer big tests
5
9
7
Other
10
12
12
Total
100
100
100
Source: Dysthe et al., 2007.
be said to contribute to “student activity” and the two became more or less synonymous. Regular writing also contributed to another goal explicitly stated in the reform documents, namely that students needed to spread their workload more evenly during the semester instead of concentrating their hard work before the final exams, as several studies of student study behavior had confirmed. It thus seems obvious that the political documents of the Quality Reform about teaching and assessment (portfolios) were a very important factor in the change toward more writing. More feedback to students One of stated aims in the Quality Reform was closer follow-up of students, and a clear finding in the evaluation study was a considerable increase in feedback. We lack comparative data before the reform, but the self-report data in Table 3.2 is nevertheless informative. Table 3.2 shows that of those in the national evaluation survey who reported changes in their teaching, approximately 70 percent gave more feedback to the students than before the Quality Reform, and 60 percent provide more supervision. The student surveys show the same tendency, and Table 3.3 shows students’ answers to the question of what kind of feedback they got.5 Table 3.3 shows that written feedback on individual assignments was most common, but almost half of the students also got oral feedback and/ or feedback on group assignments. In another of the reform evaluation studies we found more detailed quantitative evidence of the frequency of
36 O. Dysthe Table 3.2 Changes in the time spent on teaching activities To those (69%) who answered great or medium changes in their teaching activities: Changes in time spent on different teaching activites (percentages) University colleges
University
Total
Clearly About Clearly Clearly About Clearly Clearly more the same less more the same less more Written feedback 77 to students
21
2
68
31
2
71
Supervision
79
19
3
60
19
3
66
Exam-related work
53
39
8
64
32
4
61
Seminars
24
68
9
47
50
4
40
Lectures
10
55
35
16
73
12
14
Source: Dysthe et al., 2006b.
Table 3.3 “Did you last semester get. . . .?” All Male students Female students (%) (%) (%) Oral feedback on individual assignments
48
48
48
Written feedback on individual assignments
63
66
61
Oral feedback on group assignments
48
48
47
Written feedback on group assignments
45
49
43
Source: Ugreninov & Vaage, 2006.
feedback (Aamodt, Hovdhaugen, & Opheim, 2006). We find for instance that 52 percent of the students take feedback into account “to a high degree” and 39 percent “to some degree,” while only 4 percent disregard feedback. In the portfolio study we found that about one-quarter of the teachers involved their students in peer feedback, most commonly in teacher education and least in natural sciences. The conclusion of this brief review of findings in the Quality Reform evaluation combined with our portfolio survey is that the writing and feedback culture in Norwegian higher education has changed quite dramatically in just a few years. There were, however, unintended consequences that became visible through the interview studies.
What factors improve academic writing? 37 Unintended consequences The main problem for students as well as teachers was that more writing assignments and more feedback meant more work, especially because it very often came in addition to and not as a replacement for other activities in the courses. When the writing assignments were graded, students acted strategically and prioritized writing instead of going to lectures or seminars or attending groups, with negative consequences for the learning environment. On the other hand, if the writing was just a requirement to take the final exam, which was the case in many courses, students reported doing “just enough.” Students also complained that a too-high workload of writing resulted in mediocre quality because they lacked time to revise before the next assignment. Teachers complained that students read less than before, and concentrated their reading around the assignments. The unintended negative consequences were used by critics of the strong element of writing. Particular attention was given to the “time dilemma” for the teachers who now spent more time giving feedback to students’ texts. Some departments went for teaching assistants, but in many cases the professors who taught the courses also handled the paper load. The greatest threat to the new writing, feedback, and portfolio culture is still the conflict experienced by the teachers between increased expectations of research and international publishing on the one hand and the extra time spent on writing intensive courses. The introduction of an incentive-based funding system, where some of the money to each department is dependent on the amount of publishing points, clearly aggravated this conflict. This is just one example of how macro-level university policies impact on teaching and learning. It is not likely that the changes involving a substantial increase in student writing will be reversed, but every department now has to be very aware of the time constraints on their faculty as they try to balance the demands of teaching and research. This is, of course, particularly the case in the universities, while in many of the university colleges teaching is still seen as most important. A brief account of what happened in two departments will bring us closer to the realities of the decisions that were possible to take at the meso level as regards students’ writing. Two case studies of university disciplines where writing has high priority I have chosen two very traditional university disciplines, history and law. Neither of them is typical of what happened after the Quality Reform; instead they could be said to represent “best practice” examples for those who think writing is indispensable for student learning in higher education.
38 O. Dysthe The history department at the University of Bergen This department is an example of how the Quality Reform legitimized changes in writing and assessment practices that had already begun and also supported further development, in this case of electronic writing portfolios. History is a traditional academic discipline and not intuitively the place to look for new pedagogy, but this department has a tradition of great concern for the teaching and learning of students. It was the first department to introduce ICT (Information and Communication Technology) as a pedagogical tool on a large scale and an in-house electronic Learning Management System, called Kark, was developed from 1992 to 1995. Assessment systems are notoriously resistant to change (Gipps, 1994; Shepard, 2001) but already in 1999 portfolios had been introduced in this department, administered through Kark. The purpose was to support more and better student writing. The students had to share their drafts as well as their comments with fellow students and post them in Kark, in order to get teacher feedback. Because writing is regarded as an indispensable tool for learning history, Kark was designed to fit the needs of teachers giving feedback to student papers as well as facilitating students’ access to each other’s drafts. Kark had two main functions, an administrative and organizing function that made it possible to handle the great number of texts in one course, and it facilitated peer and teacher feedback on student essays and to make the comments accessible for all students in the course. When portfolio assessment was announced as an alternative assessment form in the Quality Reform, several of the freshman and sophomore history courses were assessed through portfolios, consisting of a selection of three out of five essays, peer feedback texts, and entries in asynchronous discussions. Highly qualified teacher assistants who had their Master’s degree in history were hired to give feedback to students.6 The number of texts in the portfolio was, however, radically reduced because modularization (courses of 10–15 ECTS) meant that students had less time to write, as the writing-intensive course competed with a course assessed through a traditional end-of-semester exam. History kept the system of full online transparency regarding student drafts and teacher and peer comments. Internal course evaluation has shown that a high percentage of students say that they learn almost as much from reading fellow students’ texts as from teacher comments, a finding that corresponds to studies in other countries. The weak point in the system is the lack of time to write multiple drafts and thus hone the argumentation which is one of the stated purposes of writing in this history department. At the outset the writing- intensive, technology-supported courses were advocated by a handful of enthusiasts, but the Quality Reform legitimized this approach and now it has full support: “Learning history without extensive writing is impos sible” (Oldervoll, 2003).
What factors improve academic writing? 39 The faculty of law This case illustrates a total change of system from a lecture-based and individual-study model to an integrated model inspired by problem-based learning where writing is central. Before the Quality Reform, organized student activities were restricted to lectures and seminars, individual study was the norm, writing was voluntary and assessment was postponed until the end of the third year where a major sit-down exam had a very high failure rate. Student dissatisfaction and bad results were strong incentives for radical changes. In the new model, writing and discussion of writing assignments became compulsory and totally integrated into the teaching and learning environment. The study activities of law students in the undergraduate courses are structured in weekly cycles of lectures, individual writing of drafts, group discussions of drafts published in the electronic Learning Management System (LMS) as well as feedback from peers and teaching assistants (Vines & Dysthe, forthcoming). Groups of ten students are led by a paid teacher assistant (TA) who is an advanced student (year 4–5). When the writing assignments are posted in the LMS, students prepare individually for group meetings where the assignment is discussed and possible outlines are negotiated. Each group is divided into three “commentator teams” to ensure student feedback on all assignments. Because the teams alternate, a student gets feedback from different peers. The mandatory process is strictly regulated. The group assignments are case- based, i.e., authentic legal problems constructed by experienced law teachers. Student papers are not graded. The exam consists of a take-home group exam which is a prerequisite for the individual, traditional sit-down exam, where students write essays similar to the kind they have been practicing (Vines & Dysthe, forthcoming). Weekly compulsory writing assignments, compulsory group attendance, feedback system, and frequent assessment have paid off in terms of a dramatic reduction of students who fail. In this respect there is no doubt that the strict writing and feedback regime has been effective. The fact that comments to student texts in one group amounted to approximately 15,000 words testifies to its importance.7 Student evaluations ascribe the good exam results to the structured writing and feedback processes, and also that expectations and criteria are no longer a guessing game for the students (Sadler, 1998). There are, however, some unintended effects, although different from the ones mentioned earlier as common complaints after the Quality Reform. In law, loss of motivation because of the repetitive nature of the highly structured teaching–learning sequences and too many writing assignments (one per week) have caused some student dissatisfaction. Overall student evaluations show that students appreciate the opportunity to train in both written communication and the specific legal language and genres.
40 O. Dysthe
Discussion In light of international research on educational reforms it may seem surprising that the Quality Reform has led to such substantial changes in writing and feedback practices. Larry Cuban (1990, 1993, 2002), who has studied 100 years’ worth of reforms in American educational institutions, claims that top-down reforms do not result in substantial changes in what happens in classrooms. Teachers hold on to their former practices regardless of political reform expectations. There are, however, exceptions from this: reforms may contribute to changes if they build on teachers’ expertise and take account of the specific character of this type of work: Intended curricular policy changes do influence classroom practices when policy makers develop coherent and integrated strategies that build upon teacher expertise, acknowledge the realities of the school as a workplace, and accept the wisdom of those teacher adaptations that improve the intended policy. (Cuban, 2002, p. 1) In Norwegian higher education I claim that the substantial and rapid changes in writing and feedback practices did not happen because the higher education sector is unusually loyal to top-down expectations, although there is an element of this. The main reason appears to be that the reform documents met perceived needs in the institutions and that they were in accordance with teacher expertise. Another important reason was students’ positive attitude toward more writing and feedback. In other words, the soil was ready. In the reform documents an overarching goal was that “students were to succeed,” with the double meaning: (1) learn better, (2) finish in a shorter time. Many teachers agreed that structuring students’ work through regular writing and providing constructive feedback were two important factors to achieve this. This explains why there was little resistance to this part of the reform, even though there was not much enthusiasm for a centrally initiated process of change. Other factors have also contributed to changes in writing practices. Cuban (2002) emphasised that a top-down reform is only effective if “the soil is ready” for it. Considerable development work in the area of academic writing had also taken place, and a number of books and articles on process-oriented academic writing were published in Norway (i.e., Dysthe, Hertzberg, & Hoel, 2000; Hoel, 1990, 2000; Lie, 1995). The international focus on academic competences had also just begun to filter through to the average higher education teacher. The evaluation survey and interviews revealed that there were large variations among institutions and also among departments in the same institution. This shows that even though the macro level has great influence, the meso level in a knowledge organization can ignore, minimize, or implement and develop the ideas and policies from the top. My two case studies
What factors improve academic writing? 41 illustrate that where the reform led to deep pedagogical changes, the foundations for this had in effect been laid before the reform. In the Law Faculty this meant that weaknesses had been identified and alternative solutions discussed. In the History Department the writing, assessment, and feedback model had already been tried out, and the LMS had been introduced. Thus a certain readiness for change facilitated consensus among the faculty, or at least it neutralized opposition. There is no doubt that the reform served to legitimize the pedagogical leaders who fronted the changes in writing, feedback, and assessment. But the situation in my two case studies was rather rare. A more common pattern was that the faculty and departments were confronted with the reform demands and went straight into the implementation phase, with very little discussion. Time limits were short and students had to be taught while all the courses were revised. A pertinent example is the implementation of disciplinary writing portfolios. Often the knowledge of what a portfolio was or how it could be used was thin, and the result was that in some disciplines portfolio just meant that traditional, graded writing assignments were collected in a folder and there were no innovations whatsoever, except the name. Regular writing is a necessary but not sufficient condition for improvement in writing, and there seem to be a growing awareness in all European countries that explicit teaching of writing in higher education may be necessary. The European Qualification Framework and the growing interest in generic academic competences have resulted in a new focus on writing as a tool for learning and on how to improve students’ academic writing. The European tradition is writing in the disciplines. The strategy therefore has to focus on training some teachers in the various disciplines to teach writing. A newly funded project at the University of Bergen is called “Writing development and feedback from bachelor to PhD” where a writing consultant will work with the Biology Department and the Humanities Faculty to develop a realistic and systematic plan for students’ academic writing from their first year to the doctorate. This may strengthen awareness and competence at the meso level, but the biggest hindrance may be that teachers feel that they have to prioritize their own research instead of students’ needs. Paradoxically this is a result of conflicting macro-level policies, where the Quality Reform pushed quality in teaching and now quality in research is highlighted.
Notes 1 In 1999 European education ministers met in Bologna to discuss a common European education policy for the future. This meeting gave the name to a process which is gradually changing higher education in Europe. 2 The European Credit Transfer System (ECTS) is the EU system for transfer of study credits and grades between countries. The system is meant to supplement, not replace national systems, and plays an important role in creating mobilization between European institutions and creates a European education area: www.europa.eu.int/comm/education/programmes/socrates/ects_en.html.
42 O. Dysthe 3 A problem for Norwegian students seeking further studies at US universities is that the distribution of grades in Norway differs from the United States. In Norway it is supposed to follow the normal distribution scale, which means that a C is an average grade at all levels, while A and B are just for exceptional performance. 4 The eight institutions are: the universites of Oslo, Bergen, and Stavanger, the Norwegian University for Life Sciences and the University Colleges of Sogn and Fjordane (HSF), Bodø (HSB) and Lillehammer, and the Norwegian Business College (BI). 5 The question is ambiguous as to who gives feedback, teacher or peers or both. 6 More about this case can be read in Dysthe and Tolo (2007), where the focus is on the portfolio and writing and in Wake, Dysthe, and Mjelstad (2007), where the focus is on technology. 7 Student essays in the same group counted in total 17,000 words.
References Aamodt, P. O., Hovdhaugen, E., & Opheim, V. (2006). Den nye studiehverdagen. Evaluering av Kvalitetsreformen. Delrapport 6. (The new study situation. Evaluation of the Quality Reform). Cuban, L. (1990). Reforming again, again and again. Educational Researcher, 19(1), 3–13. Cuban, L. (1993). How teachers taught. Constancy and change in American classrooms 1880–1990. New York: Teachers College Press. Cuban, L. (2002). What have researchers and policy makers learned about converting curricular and instructional policies into classroom practices? Unpublished manuscript. Dysthe, O., Engelsen, K. S., & Lima, I. (2007). Variations in portfolio assessment in higher education. Discussion of quality issues based on a Norwegian survey across institutions and disciplines. Assessing Writing, 12(2). Dysthe, O., Hertzberg, F., & Hoel, T. L. (2000). Skrive for å lære (Writing to learn: Writing in higher education). Oslo: Abstrakt Forlag. Dysthe, O., Raaheim, A., & Lima, I. (2006a). Pedagogiske endringer som følge av Kvalitetsreformen (Pedagogical consequences of the Quality Reform). In S. Michelsen & P. Åmodt (Eds.), Kvalitetsreformen møter virkeligheten. Delrapport 1. Oslo: Norwegian Research Council. Dysthe, O., Raaheim, A., Lima, I., & Bygstad, A. (2006b). Undervisnings- og vurderingsformer. Pedagogiske konsekvenser av Kvalitetsreformen. Evaluering av Kvalitetsreformen. Delrapport 7 (Teaching and assessment. Pedagogical consequences of the Quality Reform. Report no 7). Oslo: NIFU STEP—Norsk institutt for studier av forskning og utdanning/Senter for innovasjonsforskning/ Universitetet i Bergen. Dysthe, O., & Tolo, A. (2007). Digital portfolios and feedback practices in a traditional university course. In M. Kaankanranta & P. Linnakylä (Eds.), Perspectives on ePortfolios. Jyväskylä, Fi: Agora Center, University of Jyväskylä Press. Gipps, G. (1994). Beyond testing. Towards a theory of educational assessment. London: Falmer Press. Hoel, T. L. (1990). Skrivepedagogikk på norsk: prosessorientert skriving i teori og praksis (Writing pedagogy in Norway: Process writing in theory and practice). Oslo: Landslaget for norskundervisning (LNU)/Cappelen. Hoel, T. L. (2000). Skrive og samtale: responsgrupper som læringsfellesskap
What factors improve academic writing? 43 (Writing and talking: Response groups as learning communities). Oslo: Gyldendal akademisk. Lie, S. (1995). Fri som foten: om å skrive fagtekster (About writing of disciplinary texts). Oslo: Ad notam Gyldendal. Michelsen, S., & Åmodt, P. (2006). Kvalitetsreformen møter virkeligheten. Delrapport 1 (The Quality Reform meets reality). Oslo: Norsk forskningsråd. Oldervoll, J. (2003). Mappevurdering i eit skrivebasert studium. Erfaringar frå Historie grunnfag (Portfolios in a writing-based history curriculum). In O. Dysthe & K. S. Engelsen (Eds.), Mapper som pedagogisk redskap. Perspektiver og erfaringer (pp. 295–310). Oslo: Abstrakt forlag. Sadler, D. R. (1998). Formative assesment: Revising the territory. Assessment in Education, 5(1), 77–84. Shepard, L. A. (2001). The role of classroom assessment in teaching and learning. In V. Richardson (Ed.), Handbook of research on teaching, 4th edition (pp. 1066–1101). Washington, DC: American Educational Research Association. Ugreninov, E., & Vaage, O. F. (2006). Studenters levekår 2005 (Students’ life situations). Oslo: Statistisk sentralbyrå. Vines, A., & Dysthe, O. (forthcoming). Productive learning in the study of law: The role of technology in the learning ecology of a law faculty. In L. Dirckinck- Holmfeld, C. Jones, & B. Lindstrøm (Eds.), Analyzing networked learning practices in higher education and continuing professional development. Rotterdam: Sense Publishers. Wake, J. D., Dysthe, O., & Mjelstad, S. (2007). New and changing teacher roles in a digital age. Educational Technology & Society, 10(1).
4 Mapping genre research in Brazil An exploratory study Antonia Dilamar Araújo State University of Ceará, Brazil
Genre analyses of academic or professional writing describe the differences of texts and examine the correlated uses of language. Several studies have also generated theoretical models for analyzing genres that serve as a basis for their description (Bakhtin, 1992/1986; Bhatia, 1993; Bazerman, 1988; Bronckart, 1999; Swales, 1990; Halliday & Hasan, 1989, among others). An increasing interest in the study of genres in Brazil both in Portuguese and foreign languages—mainly English—has contributed to the production of a significant number of dissertations and theses in graduate programs in the field of language studies, and the publication of research papers and books on genre and discourse. In addition, since 2001 a conference entitled SIGET (International Symposium of Studies on Genres) has been held every two years as a forum for researchers and teachers to discuss and publicize their research in the area. However, even with the growing interest in genre studies in Brazil, scholars do not have a precise idea of their extent and nature. This chapter aims to report preliminary outcomes of a research project on genre studies conducted in Brazil from 1980 to 2007. In this exploratory research, I set out (a) the identification of studied genres; (b) the focus of each study; (c) their underlying theoretical framework; and (d) the type of methodology adopted for analysis. For this overview, the information was collected mainly from dissertations and theses written for graduate programs in the fields of linguistics, applied linguistics, and language studies, as well as from scholarly journal articles and books in the area. It is worth mentioning that this is the first attempt to systematize studies related to text and genre in Brazil, and it is hoped the results can illuminate the agenda for future investigations into the topic.
Brief historical contextualization Brazil does not have a long tradition of developing research in writing, and scholars only started to pay attention to this skill in the very recent past, within the last 25 years, when graduate programs in linguistics, applied linguistics, and education spread all over the country with new lines of inquiry implemented by young PhDs who qualified at American and
Mapping genre research in Brazil 45 uropean universities. Until the 1970s, most studies were conducted in colE leges and at universities, and Brazilian scholars teaching Portuguese or foreign languages were mostly focused on analyses and descriptions of grammatical aspects following a structuralist and positivist research paradigm. To understand the scholars’ interest in genres and text analysis, we need to learn about the context of writing-instruction practices in Brazil. Before 1980, teachers in schools did not worry about writing. The teaching at any level (elementary, secondary, and higher education) was concerned with reading and grammatical structures due to the great percentage of illiterate people outside schools. With the spread of graduate programs in linguistics and education that began across the country after 1980, many professors sought qualifications in writing development and instruction at American and European universities. The consequence was a number of scholars with a new view of how language should be studied as well as an increasing number of studies on genres adopting different perspectives on writing at different levels of teaching. Another important fact that contributed to the interest in text studies was the college entrance exam that started including a composition as a compulsory and eliminating stage in which students must demonstrate that they have writing abilities and have mastered academic genres. More recently, Brazilian education laws and statements have oriented educational curricula (Brazil LDB, 1996; Brazil PCNs, 1999)1 to focus on the teaching of reading and writing from a socio-interactionism perspective (Bakhtin, 1992/1986), which has been stimulating teachers to see texts as the product of social relations, as real and concrete units of verbal communication characterized by relatively stable types of oral or written utterances influenced by the purpose, theme, and relations among interactants. However, as scholars have earned degrees under different theoretical perspectives on text and genre analysis, the studies carried out in Brazil have reflected the application of these perspectives.
Epistemological characterization of genre studies The view that text genres are forms of communication associated with people’s sociocultural life and that they are seen as ways of social action, acknowledged and used in any communicative situation, has been cultivated among Brazilian scholars in their inquiries and teaching experiences. For the purpose of this chapter, theories that have strongly influenced Brazilian scholars are grouped into four main perspectives or traditions: social interactionism (Bakhtin, 1992/1986; Bronckart, 1999; Maingueneau, 2002), social rhetoric (Miller, 1984; Swales, 1990; Bazerman, 1988), social semiotics (Halliday & Hasan, 1989; Martin, 1992), and critical discourse analysis (Fairclough, 1992; Kress, 1985; van Dijk, 1990). A summary of these theories on text genre analysis is presented; then a report on the data collection and findings is given.
46 A. D. Araújo One of the most influential theories on text/discursive genres in the world has been Bakhtin’s dialogic view of language. By claiming that language has a functional, dynamic, and interactive nature, Bakhtin sets out that verbal communication is only possible by means of some discursive genre. Although his object of analysis was literary texts, he regards genres as social discursive actions used by people to act on the world and to signify the world, representing it in some way. In his Speech Genres and Other Later Essays, Bakhtin highlights that texts are concrete and empirical realizations of genres produced in the spheres of interaction in a specific social circumstance and at a specific time with their particular ideological function. As unique social events, genres are characterized by constitutive elements such as thematic content, verbal style, and compositional structure. These ideas have influenced Brazilian scholars (Brait, 2000; Rodrigues, 2001) that have developed analytical studies on genre and also applied the Bakhtinian perspective to teaching writing (Rojo, 2001) at elementary, secondary, and tertiary schools. Based on Bakhtinian dialogism and Vygotsky’s activity theory, Bronckart (1999) and Dolz and Schneuwly (1998), who represent European (Geneva) scholars, postulated a trend known as “social discursive interactionism” that accounts for genres as products of social activities of language, and as tools that allow people to realize language actions and participate in different social activities; such actions and activities may contribute to change the genre. These authors propose a nonlinear model for textual analysis that accounts for language mechanisms functioning as the basis of orientation for language action and influencing on the shape and content of texts. Their model consists of examining: (a) the content with which, the place where, and time when the participants engage in interaction; (b) the participants in their physical space; (c) the social place in which the interaction takes place; (d) the participants’ social roles; and (e) the writing effects. Genre studies in Brazil, both in Portuguese and in English, based on these ideas focus on: (a) text genre characteristics to be used as teaching tools (Dionísio, Machado, & Bezerra, 2002; Machado, 1988; Rojo, 2000); (b) learners’ development in different language practices; (c) teachers’ qualifications; (d) the interaction between teacher, tool, and learner in the classroom; and (e) evaluation of instructional materials (Barbosa, 2001; Cristovão, 2002; Machado, 1988, 2000). The second perspective of genre studies—named “socio-rhetoric” and represented by Miller (1984), Bazerman (1988), and Swales (1990)—has also strongly influenced many Brazilian scholars in their investigations into text studies. The notion of genre seen as a typified social action postulated by Miller (1984) is seminal for those interested in studying text in classroom and for whom genres are responses to the needs and expectations of participants of the discursive community. As genres are responses, they emerge from and are integrated in the cultures in which they are inserted, also locating and reformulating such cultures. Miller’s view generated new trends for genre studies. One interesting view—different from conventional
Mapping genre research in Brazil 47 notions of genre as static categories of text—is that cultivated by North American scholars from the New Rhetoric (Bazerman, 1988, 1994, 2005; Russell, 1997), for whom genre is conceived as “operationalized social action” (Russell, 1997, p. 512); in order to comprehend it, one needs to examine the rhetorical situation to look at not only the features of the context but also the motivation and intended purposes of the participants. Aligned with Miller and Vygotsky’s activity theory, Bazerman (1988, 1994, 2005) states that genre sets within genre systems are part of systems of human activities and can be investigated from interactions in sociocultural situations without being restricted to individual forms. Bazerman’s theory of genre systems is productive and applicable to any object of investigation, especially analysis of writing produced in school and other social practices. Advocating a social interactive view of genre, Bazerman (2005) acknowledges that people are engaged in society through genres and that when they are comprehended within systems and situations, readers and writers may best understand the aims, expectations, values, and shape of the texts that circulate within systems of genres; they also know how and when texts may be renewed by exclusion, addition, or reformulation. Influenced by studies on rhetoric, discourse, teaching of language skills, ethnography, Geertz’s anthropological ideas, and Miller’s view of genre, Swales (1990) offers a model for genre analysis in academic contexts. He argues that genre knowledge is an essential tool for those who deal with texts in professional situations. He defines genre as a class of communicative events, whose participants share a communicative purpose by means of language. A communicative event is constituted by the discourse, the participants, and the environment where the discourse circulates and is produced. Swales’ theory of genre analysis brings to discussion important concepts: discursive community, communicative purpose, and prototypicity. As to communicative purpose, it is a less visible feature than form; thus if someone wants to identify it, two procedures can be adopted: (a) textual or linguistic, in which the purpose is examined by analyzing the genre structure, content, and style; and (b) contextual or ethnographic, in which the purpose is revealed in the study of text in its context of use by revising the genre through its values, expectations, repertoire, and features. Swales’ model of analysis called CARS (Create a Research Space)— based on the analysis of introductions to research articles in different fields of knowledge—generated three rhetorical moves and several steps that emerged from the occurrence of information in texts. Such a model was applied to many studies about text genres like Bhatia (1993), Dudley- Evans (1986, 1994), and Wood (1982). In Brazil, the model was adapted to characterize different genres such as abstracts of research articles (Santos, 1995), dissertation abstracts (Biasi-Rodrigues, 1998), book reviews (Araújo, 1996; Bezerra, 2001; Carvalho, 2002; Motta-Roth, 1995), introductions to research articles (Aranha, 1996), newspaper articles (Bonini, 2002), sections of literature review (Hendges, 2001), among others.
48 A. D. Araújo The third perspective of text study is the “social semiotic” approach to language based on Halliday’s (1985) systemic-functional grammar, Halliday and Hasan (1989), and Martin (1992). The concept of genre for this perspective means generic structure potential (known as GSP) that is associated to Halliday’s notions of relevant meanings (semantics), ways of realization of meanings (lexico-grammar), and contexts that activate the meanings (context of situation). This concept is important to help learners to understand different types of texts and relate them to their contexts of situation and culture. The integration of text and context helps to identify compulsory and optional elements of textual structure departing from the analysis of contextual configuration or context of situation based on three variables: field (social practice), tenor (relations between participants), and mode (the role the language plays). The three variables are realized by three language functions: ideational (the content of texts), interpersonal (social interactions people participate in), and textual (the structure and shape of texts). These variables together compose the contextual configuration and allow the reader to make predictions of any kind of text that fits into a given context. Analysis of texts based on a social-semiotic perspective starts with the description of the context of situation, which leads to the unveiling of the meaningful attributes of linguistic events and the obligatory, optional, and recurrent features of texts. In Brazil, the works of Heberle (2000) and Ramos (1997) are illustrations of the application of systemic-functional theory to language analysis. Critical discourse analysis as theory and method of discourse analysis has also influenced genre studies, although Fairclough’s (2002) work does not focus specifically on text genres, but on the role of language in the production, maintenance, and change of social power relations. As one of the representatives of this perspective, he proposes that the study of the function of language in ideological processes is accomplished through three dimensions that establish, from an ethnographic perspective, how power relations work within networks of structures as well as discursive and social practices. Fairclough’s (1992) work was grounded on Halliday’s systemic-functional theory, which considers language as multifunctional, meaning that texts simultaneously represent reality, organize social relations, and establish identities. Fairclough’s contributions lie in the three dimensions of his model consisting of: “textual” (analysis of lexico- grammatical features as well as the organization and articulation of texts), “discursive practice” (analysis of the processes involving production, distribution, and consumption, and textual aspects such as coherence and intertextuality), and “social practice” (analyses of relations of power and ideology). In Brazil, the application of Fairclough’s theory to genre studies can be found in the works of Heberle (1997), Figueiredo (1995), and Magalhães (2001).
Mapping genre research in Brazil 49
Data collection The data for this ongoing study were collected from abstracts of online dissertations and theses completed in 26 linguistics and applied linguistics graduate programs, printed research articles from nine Brazilian scholarly journals, and 19 books on discourse analysis and text genres. The qualitative and quantitative analyses were based on the following aspects: identification of the studied genres, focus of investigation, theoretical perspective of the studies, and type of research method. Even though this research has been under development since 2006, the results represent only a small sample of the analyzed data that will cover a 26-year span: 1980–2007.
Findings Distribution of studies on genres by source By analyzing periodicals, theses, and dissertations, and books on genre and text-analysis studies, we can see the emergence of poststructuralist theories of language that emphasize text instead of isolated sentences that led to a growing number of investigations on text genres, especially since the beginning of the 1990s. Table 4.1 shows the quantitative distribution of published genre studies by source. By looking at Table 4.1, it is worth commenting that the production of researches on genres in graduate programs is significant. From 452 dissertations and theses analyzed in 26 graduate programs at Brazilian universities, the greatest number of studies was conducted at institutions in southern and southwestern Brazil, especially in São Paulo, Rio de Janeiro, Florianópolis, and Belo Horizonte, where there is a consolidated line of inquiry and a group of qualified scholars working in the field of genre studies. Regarding the remaining institutions in the country, the number of studies focused on text or genre is still small due to the recent interest in text studies. It is worth mentioning that we could not find information of dissertations and theses in seven of the researched graduate program home pages. Table 4.1 Distribution of published research article by source Source of publication
Period
No. of studies
Percentage
Periodicals (09)
1977–2003
117
17.0
Books (19)
1999–2007
119
17.2
Theses and dissertations (26 graduate programs)
1973–2007
452
65.6
Total
–
688
99.8
Note Nine Brazilian periodicals, 19 books and 26 graduate programs were analyzed.
50 A. D. Araújo When we examine which genres were already studied in Brazil, the results reveal that there is a great variety that may be grouped into different “discursive domains or spheres” (Bakhtin, 1992/1986), with the preferred ones being—seemingly—the school and the media. Thus, the most preferred and studied school or academic genres are: oral and written narratives, reviews, abstracts, research articles, research reports, interviews, reports, lessons, learning diaries, research projects, dissertations, theses, book introductions, prefaces, autobiographies, course plans, meeting reports, summaries, expository articles for school board, and ceremonial speech. The most preferred genres from the media are: magazine advertisements, newspaper articles, editorials (magazines and newspapers), charges, tourism prospectuses, news, telemarketing, women’s lifestyles magazines, news headlines, newspaper sports articles, cartoons, opinion articles, and magazine or newspaper editor letters. In addition, genres from other discursive domains were also investigated. Genres from daily routine (conversation, job interviews, personal letters, complaining letters, apology letters, business letters, songs, jokes, self-counseling books, recipes, life-story reports); legal genres (administrative acts, international deals, bank agreements, petitions, police news- sections, legal orientations, sentences, court judgments, public audiences, accused reports); religious (television sermons, masses, prayers, litanies, Bible psalms, religious festivals); political genres (debates, candidate speeches, political chronicles, election guides); digital genres (emails, chats, blogs, homepages, online interviews, discussion forums, IMs, electronic journal articles, electronic reviews, electronic personal advertisements); and literary (novels, short stories, fables, poems). Focus of genre investigation The data shows that the focus of investigation for the most frequent genres is still the description of generic structures (71.3 percent), motivated by the need to comprehend discursive practices and social relations associated with the use of different genres (Meurer, 2000, p. 149), as well as by the fact that many genres that circulate and are used in different contexts of human activities were not analyzed and described yet (see Table 4.2). Accordingly, the tendency in examining genres in the fields of linguistics and applied linguistics continues to be the description of genre features, both at the macro and the micro levels, as a way to respond to the scholars’ willingness to know how language uses manifest in different interactive situations among participants of a particular community. But genre studies have also been an object of concern for professors who are interested in applying knowledge stemming from theories of text genres to the classroom in both Portuguese and foreign languages (English and Spanish). Recent studies with a pedagogical focus (19.7 percent) have examined instructional genres—especially those related to the textbook—and the efficacy of genre approaches to language teaching.
Mapping genre research in Brazil 51 Table 4.2 Distribution of researches by focus of genre investigation Focus of investigation
Frequency
Percentage
Theoretical articles
61
8.8
Description of generic characteristics
501
71.3
Pedagogical applications
126
19.7
Total
688
99.8
These outcomes confirm Bhatia’s (2004) three phases in how the analysis of written discourse has been developed in the last few decades. They are: “textualization of lexico-grammar” (focus on surface-level features of language influenced by frameworks in formal linguistics), “organization of discourse” (focus on patterns of organization of larger stretches of discourse), and “contextualization of discourse” (broader look at disciplinary and institutional contexts of genre construction). The first and second phases are represented by 71.3 percent of the studies, and the third phase covers practical applications of genre to language classroom through ethnographic studies represented by 19.7 percent. Theoretical perspectives underlying genre studies Among the several existing theoretical approaches to genre studies, it is worth highlighting that social interactionism (Bakhtin, 1992/1986; Bronckart, 1999; Dolz & Schneuwly, 1998; Maingueneau, 2002) is the most preferred approach to analyze genres. However, this most frequent approach is accompanied by a combination of perspectives, given that a great number of studies describe not only rhetorical and schematic structures, but also lexico-grammatical aspects that characterize genres (see Table 4.3). Common combinations in these studies are: Bakhtin and Swales, or Bakhtin and Vygotsky, Swales and Halliday, and Hasan or Swales and Bazerman, among others. In addition to these perspectives, textual analysis based on social rhetoric (Bazerman, 1988; Miller, 1984; Swales, 1990), critical discourse analysis (Fairclough, 1992; Kress, 1985; van Dijk, 1990) and the systemic-functional approach to language (Halliday, 1985; Halliday & Hasan, 1989) are also very popular. Less frequent but also adopted are authors from French discourse analysis (Ducrot, 1980; Charadeau, 1983; Greimas, 1966; Pêcheux, 1969) and authors from still other perspectives like Labov (1972, oral narratives), Austin (1962), and Searle (1969, speech acts). Out of all studies, 5.8 percent did not indicate the theoretical perspective underlying the investigation in the abstracts analyzed. Methodological approaches to genre studies Although different theoretical-methodological perspectives for genre ana lysis are familiar to Brazilian analysts (sociointeractionism, sociorhetoric,
52 A. D. Araújo Table 4.3 Distribution of research by theoretical perspectives Theoretical perspectives
Frequency
Percentage
Social interactionism
166
24.1
Social rhetoric
113
16.4
47
6.8
Social semiotic perspective Critical discourse analysis
55
7.9
186
27.0
Other perspectives
81
11.7
No indication
40
5.8
688
99.7
Combination of perspectives
Total
Table 4.4 Types of research methods for empirical works Research method
Frequency
Percentage
Descriptive (text analysis)
501
79.9
Ethnography
62
9.8
Action-research
38
6.0
Case study
26
4.1
Total
627
99.8
Note Theoretical articles are not included in the statistics.
systemic-functional grammar, critical discourse analysis, and others), we found that the majority of the analyzed empirical works adopts—as the preferred research method for studying genres—a textual and descriptive analysis of artifacts based on theoretical epistemology with a combination of qualitative and quantitative approaches (see Table 4.4). Table 4.4 shows the methodological orientation for the analysis of genres in the study. The majority of studies (79.9 percent) categorized the adopted research methods as descriptive, that is, they are in fact text analysis or, to be more specific, genre analysis. A body of texts representative of a certain genre is selected and discursive practices or mechanisms of text organization are described. Usually quantitative content analysis serves as ground for interpreting qualitative data. Qualitative studies are starting to be adopted by researchers in Brazil, though with less frequency. An ethnographic approach is adopted in 9.8 percent of the studies to examine discursive processes and practices in professional and academic contexts. Such studies start with the examination of the context, values, intentions, expectations, the material conditions of
Mapping genre research in Brazil 53 the community, repertoire and change of genres, and finish with the characterization of genres that circulate in the focused community (Askhave & Swales, 2001). Data collection includes the observations of contexts through audio or videotape, formal and informal interviews, and relevant documents analysis. Action research in educational settings accounts for 6 percent of the corpus. These studies examine classroom interventions with the aim of increasing their comprehension of the teaching–learning process, finding ways of solving problems, or promoting social changes (Thiollent, 2002; Wallace, 1998). A qualitative case study approach represented by 4.1 percent of studies has the focus on a particular classroom to understand the nature of students’ learning of genres in school settings over time (Dörnyei, 2007). A typical case study research focuses on reports on teachers’ practices and beliefs about genres, observations, interviews, questionnaires, and written texts collected over a period of one year. Methodological chapters of the dissertations and theses sometimes reveal mixed research method designs. This issue will be examined in a future study.
Conclusion This chapter, although limited in quantity of analyzed data, attempted to provide an overview of the beginning of genre studies conducted by Brazilian scholars from 1980 until 2007. The findings show that genre studies are growing all over the country, and the number and variety of genres examined are evidence of teachers’ awareness that the study of text genres fosters understanding of “how lived textuality plays a role in the lived experience of a group” (Devitt, Bawarshi, & Reiff, 2003, p. 542), given that text genres offer insights into the complex interrelationship between language and community. It also seems evident from the mapping that the analytical model for texts continues to be the description and interpretation of discursive practices of genres. This can be explained by the need to understand genres as a site of social and ideological action and to use generic characteristics and rhetorical purpose as tools to help professors teach genres in both native and foreign languages. The reflections about theoretical-methodological constructs on genre indicate that, despite the fact that Brazilian researchers acknowledge different methodological orientations concerning genre analysis, they seem to describe genres in an intuitive way. So the establishment of methodological procedures for genre analysis should be the concern of scholars for the purpose of devising new frameworks for investigation that are coherent with adopted theoretical trends and helping researchers to analyze new genres in an adequate and satisfactory way. Finally, it is worth emphasizing that most studies analyzed are grounded on works by foreign authors, and they do not refer to previous Brazilian studies of the same theme and design. Maybe this is due to a lack of systematization of genre studies
54 A. D. Araújo developed so far; such systematization may guide scholars as for their future studies. This study intends to fill this gap.
Acknowledgments My special thanks to Keith Short, Pedro Henrique Praxedes Filho, and the editors of this volume for their thorough revision and challenging comments on the manuscript.
Note 1 Lei de Diretrizes e Bases da Educação (National Law and Bases for Brazilian Education, 1996) and Parâmetros Curriculares Nacionais (National Curricular Parameters, 1999). Both documents rule and orient elementary and secondary education in Brazil.
References Aranha, S. (1996). Argumentação nas introduções de trabalhos científicos da área de química (Argumentation in research article introductions in the area of chemistry). Unpublished master’s thesis. Pontífica Universidade Católica de São Paulo, São Paulo, Brazil. Araújo, A. D. (1996). Lexical signaling: A study of unspecific nouns in book reviews. Unpublished doctoral dissertation. Universidade Federal de Santa Catarina, Florianópolis, Brazil. Askehave, I., & Swales, J. M. (2001). Genre identification and communicative purpose: A problem and a possible solution. Applied Linguistics Journal, 22(2), 195–212. Austin, J. (1962). How to do things with words. Oxford: Oxford University Press. Bakhtin, M. M. (1992). Estética da criação verbal (Speech genres and other late essays). (M. E. G. Pereira, trans.). São Paulo: Martins Fontes. Barbosa, J. P. (Ed.). (2001). Narrar. São Paulo: FTD. Bazerman, C. (1988). Shaping written knowledge: The genre and activity of the experimental article in science. Madison, WI: University of Wisconsin Press. Bazerman, C. (1994). Systems of genres and the enactment of social intentions. In A. Freedman & P. Medway (Eds.), Genre and the new rhetoric (pp. 79–101). London: Taylor & Francis. Bazerman, C. (2005). Gêneros textuais, tipificação e interação (Trad e organização de Ângela P. Dionísio & Judith C. Hoffnagel). São Paulo: Cortez Editora. Berkerkenkotter, C., & Huckin, T. N. (1995). Genre knowledge in disciplinary communication: Cognition/culture/power. Hillsdale, NJ: Lawrence Erlbaum. Bezerra, Benedito G. (2001). A distribuição das informações em resenhas acadêmicas. Unpublished master’s thesis. Universidade Federal do Ceará, Fortaleza, Brazil. Bhatia, V. K. (1993). Analyzing genre: Language use in professional settings. London: Longman. Bhatia, V. K. (1997). Genre analysis today. Revue Belge de Philogie et d’Histoire, 75, 629–652. Bhatia, V. K. (2004). Worlds of written discourse: A genre-based view. London: Continuum.
Mapping genre research in Brazil 55 Biasi-Rodrigues, B. (1998). Estratégias de condução de informação em resumos de dissertações. Unpublished doctoral dissertation. Universidade Federal de Santa Catarina. Florianópolis, SC, Brazil. Bonini, A. (2002). Gêneros textuais e cognição: Um estudo sobre a organização cognitiva da identidade do texto. Florianópolis: Insular. Brait, B. (2000). PCNs, gêneros e ensino de língua: faces discursivas da textualidade. In R. Rojo (Ed.), A prática da linguagem em sala de aula: Praticando os PCNs (pp. 13–26). São Paulo: EDUC, Campinas: Mercado das Letras. Brazil, Ministério da Educação. (1996). Lei de Diretrizes e Base da Educação Brasileira. Brasília: MEC. Brazil, Ministério da Educação, Secretaria da Educação Média e Tecnológica. (1999). Parâmetros Curriculares Nacionais, Linguagens, códigos e suas tecnologias. Brasília: MEC. Bronckart, J. (1999). Atividades de linguagem, textos e discursos: Por um interacionismo sóciodiscursivo. São Paulo: EDUC. Carvalho, Gisele de. (2002). Resenhas/reviews: Da ação entre amigos ao apontador de defeitos? (Um estudo contrastivo de resenhas acadêmicas escritas em inglês e português). Unpublished doctoral dissertation. Universidade Federal Fluminense, Niterói, RJ, Brazil. Charadeau, P. (1983). Language et Discours. Paris: Hachette. Cristovão, V. L. L. (2002). Gêneros e ensino de leitura em LE: Os modelos didáticos na construção e avaliação de material didático. Unpublished doctoral dissertation. Pontifícia Universidade Católica de São Paulo, São Paulo, Brazil. Devitt, A. J., Bawarshi, A., & Reiff, M. J. (2003). Materiality and genre in the study of discourse communities. College English, 65, 541–558. Dionísio, A. P., Machado, A. R., & Bezerra, M. A. (Eds.). (2002). Gêneros textuais & ensino. Rio de Janeiro: Lucerna. Dolz, J., & Schneuwly, B. (1998). L’oral comme text: Construire um object enseignable. In Pour un enseignement de l’oral. Initiation aux genres formels à lécole (pp. 49–73). Paris: ESFediteur. Dörnyei, Z. (2007). Research methods in applied linguistics. Oxford: Oxford University Press. Ducrot, O. (1980). Les mots du discours. Paris: Minuit. Dudley-Evans, T. (1986). Genre analysis: An investigation of the introduction and discussion sections of MSc dissertations. In M. Coulthard (Ed.), Talking about text (pp. 128–145). Discourse Analysis Monographs, no. 13. Birmingham, UK: English Language Research, University of Birmingham. Dudley-Evans, T. (1994). Genre analysis: An approach to text analysis for ESP. In Malcolm Coulthard (Ed.), Advances in written text analysis (pp. 219–228). London: Routledge. Fairclough, N. (1992). Discourse and social change. Cambridge: Polity Press. Fairclough, N. (2002). Critical discourse analysis as a method in social scientific research. In R. Wodak & M. Meyer (Eds.), Methods of critical discourse analysis: Introducing qualitative methods (pp. 121–138). London: Sage Publications. Figueiredo, D. C. (1995). The use and abuse of your sexual power: Cosmopolitan/ Nova and the creation and maintenance of a conservative view of female sexuality. Unpublished master’s thesis. Universidade Federal de Santa Catarina, Florianópolis, SC, Brazil. Greimas, A. J. (1966). Sémantique structurale. Paris: Larousse.
56 A. D. Araújo Halliday, M. (1985). An introduction to functional grammar. London: Edward Arnold. Halliday, M., & Hasan, R. (1989). Language, context and text: Aspects of language in a social semiotic perspective. Oxford: Oxford University Press. Heberle, V. (1997). An investigation of textual and contextual parameters in editorials in women’s magazines. Unpublished doctoral dissertation. Universidade Federal de Santa Catarina, Florianópolis, SC, Brazil. Heberle, V. (2000). Análise crítica do discurso e estudos de gênero: subsídios para a leitura e interpretação de textos. In M. Fortkamp & L. Tomitch (Eds.), Aspectos de lingüística aplicada (pp. 289–316). Florianópolis: Insular. Hendges, G. R. (2001). Novos contextos, novos gêneros: A revisão da seção de literatura em artigos acadêmicos eletrônicos. Unpublished master’s thesis. Universidade Federal de Santa Maria, Santa Maria, Rio Grande do Sul, Brazil. Kress, G. (1985). Linguistic processes in sociocultural practice. Geelong, Victoria: Deakin University Press. Labov, W. (1972). Language in the inner city: Studies in the Black English vernacular. Philadelphia, PA: University of Pennsylvania Press. Machado, A. R. (1988). O diário de leituras: A introdução de um novo instrumento na escola. São Paulo: Martins Fontes. Machado, A. R. (2000). Uma experiência de assessorial docente e de elaboração de material didático para o ensino de produção de textos na universidade. D.E.L. T.A., 16(1), 1–25. Magalhães, C. (Ed.). (2001). Reflexões sobre análise crítica do discurso. Belo Horizonte: FALE: Universidade Federal de Minas Gerais, Belo Horizonte, Brazil. Maingueneau, D. (2002). Análise de textos da comunicação (Trans. C. P. de Souzae-Silva & D. Rocha). São Paulo: Cortez. Martin, J. R. (1992). English text: System and structure. John Benjamins: Amsterdam. Miller, C. R. (1984). Genre as social action. Quarterly Journal of Speech, 70, 151–167. Motta-Roth, Désirée. (1995). Rhetorical features and disciplinary cultures: A genre-based study of academic book reviews in linguistics, chemistry and economics. Unpublished doctoral dissertation. Universidade Federal de Santa Catarina, Florianópolis, Brazil. Pêcheux, M. (1969). Analyze automatique du discours. Paris: Dunod. Ramos, R. C. G. (1997). Projeção de imagens através de escolhas lingüísticas: Um estudo no contexto empresarial. Unpublished doctoral dissertation. Pontifícia Universidade Católica de São Paulo, São Paulo, Brazil. Rodrigues, R. H. (2001). A constituição e o funcionamento do gênero jornalístico artigo: Cronotropo e dialogismo. Unpublished doctoral dissertation. Pontifícia Universidade Católica de São Paulo, São Paulo, Brazil. Rojo, R. H. R. (Ed.). (2000). A prática de linguagem em sala de aula: Praticando os PCNs. São Paulo: EDUC; Campinas: Mercado de Letras. Rojo, R. H. R. (2001). A teoria dos gêneros em Bakhtin: Construindo uma perspectiva enunciativa para o ensino de compreensão e produção de textos na escola. In B. Brait (Ed.), Estudos enunciativos no Brasil: História e perspectivas (pp. 163–185). Campinas: Pontes. Russell, David R. (1997). Rethinking genre in school and society: An activity theory analysis. Written Communication, 14(4), 504–554. Santos, M. B. dos. (1995). Academic abstracts: a genre analysis. Unpublished master’s thesis. Universidade Federal de Santa Catarina, Florianópolis, Brazil.
Mapping genre research in Brazil 57 Searle, J. (1969). Speech Acts. Cambridge, England: Cambridge University Press. Swales, J. M. (1990). Genre analysis: English in academic and research settings. Cambridge: Cambridge University Press. Thiollent, M. (2002). Metodologia da pesquisa-ação. São Paulo: Cortez Editora. Van Dijk, T. A. (1990). La noticia como discurso: Compreensión, estructura y producción de la información. Barcelona: Paidós. Wallace, M. J. (1998). Action research for language teachers. Cambridge: Cambridge University Press. Wood, A. S. (1982). An examination of the rhetorical structures of authentic chemistry texts. Applied Linguistics, 3, 121–43.
5 The teaching and learning of writing in Portugal The case of a research group Luísa Álvares Pereira, Conceição Aleixo, Inês Cardoso, and Luciana Graça Didactics and Learning Technology Department, University of Aveiro, Portugal
Introduction In Portugal, political attention has focused on the poor levels of literacy and knowledge of the Portuguese language of both students and the general population, as revealed by national exams and international studies (M. E., 2004; PISA, M. E., & GAVE, 2001); to solve the problem, politicians, researchers, trainees, and teachers are focusing on the work that is done at schools. Within this national context we have conducted research on the learning of writing and the underlying teaching practices, taking into account two main areas: •
•
writing as an activity, moving from a focus on the product-text to a focus on the process of writing and the intervening factors, defined and systematized in the socio-discursive interactionism framework (Bronckart, 2006; Pereira & Graça, 2007; Vygotsky, 1987/1934); the analysis of writing as a subjective and sociocultural practice, either in school or in personal, professional, institutional, or social contexts (Barré-de Miniac, 2000; Chapman, 2006; Charlot, 1997; Lahire, 1993; Penloup, 1999).
This research has focused on the three cycles of Basic Education in Portugal (covering nine school years): primary school (ages six to ten); the second cycle (ages ten to 12); and the third cycle (ages 12 to 15). Throughout these cycles, current teaching of writing is “simplistic and reductive” (Pereira, 2000). Additionally, there is a clear distance between classroom practices, research principles, and ministerial directives. Given these circumstances, it is our aim here, first, to analyze policy directives, as well as the research on writing across these cycles. Second, we present data resulting from the research on classroom practices, which we have carried out, in an attempt to articulate research and teaching methodologies, and to gain a clearer understanding of the factors that contribute to the improvement of teaching strategies and learning. In order to accomplish our research objectives, we worked within classroom contexts and
Teaching and learning writing in Portugal 59 implemented varied pedagogical devices: didactic sequence (DS), writing notebooks (WN), and writing workshops (WW).
The “new” pedagogy of writing in Portuguese official documents: priorities In the past two decades, recognition of writing as an object of research has been accompanied by a renewed understanding of writing’s essential role both in school success and in individuals’ lives. This new perspective resulted in changes to the Portuguese Language programs within Basic Compulsory Education, in 1991. By 2001, these changes were corroborated and made explicit in the Essential Competences for Basic Education, which trace the profile of what students should know (how to do) by the end of their compulsory cycle of education (M. E., 2001a). Indeed, these official instructions confer a definite status to writing by considering it a specific teaching object that requires systematic work. The focus is now on processes of textual production, namely the production of authentic texts for authentic receivers. One of the outcomes of this education shift has been extending writing assignments beyond literary texts, and encouraging the production of different texts. Writing workshop practices have also become common both at the level of basic schooling and high school. Official recognition of the value of writing as a teachable competence does not necessarily translate into effective teaching practices. The recent introduction of national exams (at the end of the First Cycle of Basic Education (1st CBE) in 2000; at the end of the 2nd CBE, in 2001; and at the end of the 3rd CBE, in 2002) that include writing assessments highlights this fact by unveiling students’ lacks in regards to the production of specific genres. The published reports of these assessments (M. E., 2004, 2001b, 2000; M. E. & GAVE, 2007) provided a description of the results of the students’ performance, and made some recommendations with regard to writing: diversifying texts according to purpose; monitoring of the writing process by teachers and among students (peer-to-peer learning), and assessing not only written products but also the writing process. It is within this context that the Didactics of Writing emerged as an innovative framework for understanding the conceptions and praxis of teaching and learning, and for providing grounded orientations for good teaching and learning practices.
Theoretical perspectives and methodologies: didactic research/intervention devices The Didactics of Writing approach to the teaching of writing is informed by research from the perspective of a variety of disciplines, especially cognitive psychology and textual-, socio- and psycholinguistics, and is concerned therefore not only with writing processes and written products, but also with the sociocultural environment and its impact on student engagement and
60 L. Á. Pereira et al. erformance. More specifically, the development of pragmatic and textual p linguistic studies and the focus on the dimensions of language that go beyond the sentence have increased the attention on the function of texts and their discursive properties, which have made undeniable contributions in our understanding of writing across content areas and genres. Furthermore, knowledge generated by cognitive psychology regarding the different processes of “expert” and “novice” writers has been transferred to the field of Writing. This knowledge has helped to develop and test teaching models that take into account the different stages of the writing process, the development of the subject’s metalinguistic consciousness of language (Barbeiro, 1999), as well as the identification of working methodologies that facilitate textual production (Carvalho, 1998). Proposals from textual linguistics (Adam, 1999) have drawn greater attention to genres, here understood as organizing macro-categories, as tools of social action, which organize an incredible array of human activities (Bronckart, 1996). Through contact with texts that circulate in their own world, children are able to intuitively construct knowledge regarding text genres (Chapman, 2006; Pereira, 2008). However, in order to develop a greater command of textual production, there is a need for a systematic learning of a wide range of textual genres. Additionally, the development of a textual repertoire includes increased processes of self-regulation and the reduction of task complexity, since the previous knowledge of a specific schematic structure decreases the cognitive effort of writing. Besides analyzing students’ textual production from the point of view of linguistics and psychology, research on writing in Portugal has also valued the conceptual framework and didactic choices made by teachers (Pereira, 2000). Through interviews carried out with 18 teachers from the 3rd CBE and from high schools, together with questionnaires handed out to 127 teachers within the same teaching contexts, Pereira (2000) was able to identify the existence of three types of teachers: (1) teachers who privilege writing as a archetype of written verbal production and who rarely write at a personal level; (2) teachers who value creative text production, and who reveal a tendency to write at a personal level, namely texts of a more poetic or intimate nature; and (3) teachers who associate writing with reading so as to promote students’ knowledge of different genres and who are used to writing texts for academic purposes. From the analysis of teachers’ discourses concerning their practices in the teaching of writing, we were able to conclude that the representations of (teaching) writing evinced by each of these profiles underlie the didactic options made in classroom. Therefore, didactics should continue to focus on teachers’ conceptions at this level, given that they determine choices of action. Without this previous knowledge and reflection upon such representations, didactics will be unable to intervene in didactic practices. As well as having identified the conceptual framework and options behind teachers’ pedagogical approaches to writing, through discourse analysis, we analyzed effective teaching practices with the aim of identify-
Teaching and learning writing in Portugal 61 ing teachers’ typical professional habits and analyzing the teaching devices they use and the way in which teaching objects are (re)configured, in the classroom, and transformed into objects which are in fact teachable (Schneuwly & Thévenaz-Christen, 2006). Given that the teaching activity does not, after all, always coincide with the teaching materials’ aims, more research is being conducted on the relationship between teacher beliefs and practices and student learning outcomes (Graça & Pereira, 2006, 2005; Pereira & Graça, 2005). In addition, more recent studies have revealed the importance of taking into account individual students’ relationship with writing (Barré-de Miniac, 2000; Charlot, 1997; Lahire, 1993) and the way they act in different writing situations. This perspective considers: (1) the role of the subject and the meaning found in writing tasks as essential to the development of writing competence, hence sustaining the notion that writing at school can be a factor of democratization or of exclusion, depending on the ability to establish a bridge between the language acquired in each subject’s social environment—or mother tongue—and the language used at school (Cardoso & Pereira, 2007a, 2007b; Pereira & Cardoso, in press); (2) the analysis of the writing produced by students outside the school context as well as on the type of relationship established with school (con)texts (Cardoso & Pereira, in press, 2007c, 2005; Penloup, 1999). All these research areas focus on the Portuguese Language “classroom,” with the aim of understanding how students develop academic literacy, as well as analyzing teachers’ contributions to that development. It seems essential to study the type of activities that intervene in changing writing habits. As various empirical observations have demonstrated, academic practices of teaching writing are not attaining the desired learning outcomes for the development of this macro-competence (Pereira, 2008). This is illustrated by the expectation that students are able to write a certain type of text exclusively because they have been exposed to and/or read examples of that genre, believing that the reading–writing transposition is immediate. These concerns have motivated the development of didactic materials for classroom intervention, framed by the essential research questions of our work: What are the effects of different didactic devices, which include rewriting projects with the mediation of another person (teacher, colleagues) and socialization, in the sense that texts have a social role ascribed by the dynamics of their circulation inside and outside school contexts and among individuals, in the evolution of students’ writing practices? 1 How do didactic devices implemented in teacher education/training enhance teaching performance and influence the object taught? 2 In what way can the relationship with writing influence students’ school writing trajectory? 3 How can the knowledge acquired on the Subject Writer contribute toward a reconceptualization of more inclusive and resourceful didactic practices?
62 L. Á. Pereira et al. The above questions provide a consistent framework for a variety of studies. For this chapter, we have selected three examples which include the implementation of a didactic sequence, the use of writing notebooks, and the organization of a workshop on/about writing.
The case of a research group: researching in the classroom International, national, and classroom-based assessments have frequently led to negative perceptions regarding the quality of students’ writing. These perceptions contribute toward our research interest in understanding in what ways and to what degree writing “quality” can be improved through different teaching practices, as well as the nature of the changes that take place at the level of teachers’ practices in loco by the use of different teaching devices, in particular WW, DS, and WN. Didactic sequence One of these devices is the use of a DS, where the main goal is to lead students to discover their own writing processes when writing particular genres, which is illustrated in Figure 5.1 (Schneuwly, Dolz, Rodrigues Rojo, & Cordeiro, 2004): The DS is central to our concern because it focuses on the writing process, i.e., it implies and foresees stages of planning, writing, reviewing, and rewriting that impact students’ writing and teachers’ work (discourses and practices). The use of the DS led to considerable improvements in the students’ texts, for example, to a higher level of development in the organization of ideas, a significant increase in the use of specific terms related to the text worked on, better organization, a greater diversity of transitional elements between paragraphs, as well as a tendency to produce longer texts (Pereira & Graça, 2005; Pereira, 2007). In the case of opinion texts, for example, our research focused specifically on the impact of the DS on two main dimensions: the teacher’s choices of action when teaching the opinion text and the differences in the way this genre is presented in the classroom. So, in order to examine changes in both of these areas (teacher’s choices of action and genre features), a set of several sixth-grade classes were taught by six teachers, who were filmed, in two stages. In the first stage, each teacher taught the
Context of communication
Initial production
Textual genre
Figure 5.1 Didactic sequence.
Module 1
Module 2
Module n
Final production
Teaching and learning writing in Portugal 63 opinion text according to their lesson plan. In the second stage, they taught the same content using the didactic tool provided by us—the DS. Interviews were carried out before and after each of these stages; the latter included observation, by the teacher, of some extracts of the videos (see Table 5.1). The analysis resulting from the observation of these two stages provided interesting findings related to the usefulness of the DS to learning and to the learning of writing in particular (Cordeiro, Azevedo, & Mattos, 2004; Graça & Pereira, in press, 2008a, 2008b, 2005). Current data analysis clearly shows different logics both at the level of conceptualizing the didactic object and of transmitting it in a didactic manner. For the purpose of better illustrating our findings, we can highlight that the introduction of the DS resulted in the study of certain characteristics of opinion texts— textual deconstruction in order to understand the structure of the text, arguments, etc.—unlike what had happened in the first stage, in which the emphasis was placed on the motivation for writing and focused on the content of the text to be produced. Writing notebooks The second pedagogical intervention involves the use of WN which approach and perceive writing as an act of personal free will. In this case, it is relevant to note that the official program for the 1st CBE (M. E., 2001a, 1991) establishes that each student should have “a notebook in which they may experience writing as they know it, about what they want, when they want.” This guideline allowed us to expand a previous study (Aleixo, 2005) based on interest around meanings and practices that are perceptible in the use of individual WN by students and teachers. Our research was designed in this case to analyze the use of WN from multiple perspectives: (1) the textual diversity observed in the individual WN, in order to identify the types of texts produced and the emerging genres; (2) the variety of contexts of production generated by the students, Table 5.1 Description of empirical research First stage
Interview with each teacher
Audiovisual recording of a “common” teaching-learning sequence
Interview with each teacher
Audiovisual recording of a teaching-learning sequence using a DS
Interview with each teacher (+ excerpts of the films)
Delivery of a DS to each teacher Second stage
Interview with each teacher
64 L. Á. Pereira et al. so as to register personal writing initiatives; (3) the characterization of the students’ writing options, in order to understand the development of their relationship with writing; (4) the appropriateness of the teachers’ methodological options, in order to consider differences among pedagogical devices in relation to the heterogeneity of interests and competences demonstrated by the students. In the three classes that participated in this study, the use of individual WN included individual and pair writing. During the two school years in which the research was conducted, the use of the WN throughout the classes shared many common features (Figure 5.2), despite the differences identified concerning time, organization, and monitoring. Our analysis focused on four areas: (1) the use of the WN (through direct observation and registers made available); (2) the dynamics created in each class (through observation/selection of episodes, teachers, and students’ reports); (3) students’ relationship with and conceptions of writing (through questionnaires and informal conversations); and (4) teachers’ opinions about their own didactic options (through semi-structured interviews). Furthermore, the time dedicated to classroom observations allowed us to identify similarities and differences in the pedagogical choices made by each teacher in the use of the WN. In Class A, the WN were always present as part of students’ daily school material and were used every day because each student was asked to read and react to at least one text. After every student had read, the class commented on the texts they had heard all at once, i.e., some of the texts or some of the students may receive a global comment. In Class B, the WN were part of each student’s daily activities. The teacher repeatedly suggested that students use the notebooks to write during independent study hours and at the end of each school day; additionally, half-an-hour a day was reserved for five students to read their writing from their notebooks aloud to the rest of the class. The class then orally responded to each student’s writing, text by text, i.e., each student received responses to each one of his/her texts. In Class C, the WN were kept in the classroom and students used them during independent study hours. In this class, half-an-
Individual production
Presentation to the class Revision (possibility)
Individual Presentation (...) production to the class Revision (possibility)
Figure 5.2 Stages of the use of the writing notebooks.
Teaching and learning writing in Portugal 65 hour was reserved for the oral sharing of student writing. At the end of each presentation, the class and the teacher commented on the work presented author by author, i.e, each student received a global comment about all the texts s/he presented. Our current analysis (Aleixo & Pereira, 2007; Aleixo, 2005) suggests that WN are associated with a significant production of texts, both in terms of the quantity of writing that students generated, as well as the range of genres that students have appropriated into their textual repertoire. Moreover, the effectiveness of the WN seems to have been affected by the teachers’ pedagogical and didactic choices. For example, concerning the quantity of texts produced, we confirmed that the three options for commenting on the texts—all at once, text by text, and author by author—led to an increase in the number of texts produced by the students. However, the differences regarding the diversity of texts indicate an advantage in commenting on each text, per se, as we observed in Class B. Workshops on/about writing Our third intervention focuses on the important relationship between the student writer and their conceptions of writing and the writing process. The designing of effective writing pedagogy among heterogeneous and diverse school populations requires knowledge of individual students’ writing practices (Cardoso & Pereira, 2007a). For this reason, our research attends to individual students’ responses to the following questions: What do they think about the writing they practice at school? What questions do they have about writing? What difficulties do they encounter when they write? Do they feel motivated to write when they are not required to do so? Is school alone motivating their writing or do they write for extracurricular purposes? In order to answer these questions, we planned and carried out a teaching experiment entitled “Workshop on/about Writing,” that was framed by our conceptual model, in which we schematized the dynamics of the classroom, the sustaining arguments,1 and the teaching objectives, and strategies used in the WW (Figure 5.3). This WW functions as a tool that embraces research concerns while simultaneously testing pedagogical and didactic motivations, focusing on three major objectives which integrate and mutually complement each other, by going from research to practice, with the intent that the latter will allow us to: • • • •
gain deeper knowledge of the relationship students have with writing; observe the dynamics of the classroom environments and their impact on students’ relationship with writing; investigate the impact of extracurricular writing on students’ acquisition of academic literacy; improve instructional practices.
66 L. Á. Pereira et al.
Reflecting about writing Writing
Speaking Writing Representations Personal uses Socialization of the students’ texts
Me Academic writing Non-academic writing Relationship
The others social uses Taking action through writing
Instruments
Getting to know young people’s writing
Getting to know young people through writing
• • • •
Questionnaire “Your writing” Testimonies on writing Functional genres—authentic texts Languages and discourses of the researcher and other, more or less, experienced writers • Subjects’ and researcher’s diaries • Questionnaire assessing the project
Figure 5.3 Conceptual model of the “Workshop on/about Writing.”
The organization and implementation of the WW aimed at answering the questions as presented in Table 5.2. This WW included individual and group activities including a variety of writing activities and debates. Our research essentially emphasized obstacle-representations, which we identified through individual interviews with students who participated in the activities. These interviews took place after school during the months that the WW were in effect. The information collated from the WW and from the interviews—which is being processed with the assistance of the QSR N-Vivo 7 software (Cardoso & Pereira, 2007b)—shows positive changes in students’ relationship with writing. In particular, we observed that a critical attitude might be contributing to the reconstruction of a relationship with writing, as the activities carried out led students to reflect upon the process and functions of writing. Students’ discussion of these activities allowed us to conclude that there was a deeper understanding of writing processes and of what
Teaching and learning writing in Portugal 67 Table 5.2 Organization and operationalization of the “workshop on/about writing” Title Promoter Schedule
Periodicity
Public Period
Me, Writing and Others The researcher 15 hours, during Portuguese Language classes, at school—or on Wednesdays from 10:15 to 11:45 a.m., or on Fridays from 8:30 to 10:00 a.m. (90-minute sessions)—total number of sessions: ten According to the teacher’s availability and to the school calendar: sessions organized with two day intervals during four weeks maximum 16 students, with ages between 11 and 15, from a seventh-grade form, from a predominantly urban area of the district of Aveiro1 February to June 2006
Note 1 According to the designation which resulted from a collaborative study carried out between the National Statistical Insitute and the Directorate-General for Land-use Planning and Urban Development.
they are able to do personally, professionally, and socially with writing, outgrowing the school perception of writing.
Final considerations As noted earlier, this last decade in Portugal has seen an increased emphasis on teaching, learning, and researching writing. However, this “writing era” coexists in an apparently paradoxical manner with a “writing crisis” among both students and teachers. Indeed, our current educational and research contexts have pointed out the need to emphasize problems posed uniquely by the teaching of different kinds of texts and the importance of understanding individual students’ relationships with writing as a global subject. Our research suggests that these pedagogical approaches—(1) the learning of texts as a process— didactic sequence; (2) the production of texts by students through personal motivation—writing notebooks; (3) the understanding of the activity/ student/subject’s motivation underlying text production—Workshop on/ about Writing) contribute toward the improvement of school programs which in turn can help each student use writing as a citizen with full rights. Nevertheless, in our opinion, there is still the need to establish a “web of interactions” which reinforces the desirable link between research didactics and professional didactics, with the intent of creating conditions for innovation based on the transforming potentials of a concrete line of action—reasons which have led us to present, although in a summarized manner, a research perspective directed toward the study of specific cases, from which conclusions that are consistent with the education praxis might be (re)drawn. Gradually, another line of intervention and research may emerge as a result of the communicative circuits established through the so-called “web of interactions” between research and practice. If the results obtained come
68 L. Á. Pereira et al. to be adequately reformulated as courses of action, guiding an application of more systematic and general approaches to the teaching and learning of writing, we may then consider their impact on the professional development of teachers and on those who are enrolled in teacher-training programs. More than assessing the effects of the pedagogical approaches in isolated cases, it is our intention to accompany and analyze the integration of these teaching methods in different professional development models, in order to contribute to urgently needed research-based approaches to the design and selection of school activities that promote the development of writing in students at all educational levels.
Note 1 Students’ relationship with school and extracurricular writing, students’ conceptions of social uses of writing, potentialities and limitations of writing workshops to reconstruct the relationship with writing, namely by broadening the perception of its social, academic, and personal functions.
Bibliography Adam, J.-M. (1999). Linguistique textuelle. Des genres de discours aux textes. Paris: Nathan. Aleixo, C. A. (2005). A vez e a voz da escrita. Lisbon: Ministério da Educação, Direcção-Geral de Inovação e de Desenvolvimento Curricular. Aleixo, C., & Pereira, L. A. (2007). Da intencionalidade na escrita à emergência de gêneros discursivos. In A. Bonini, D. Figueiredo, & F. Rauen (Eds.), SIGET: 4th Internacional Symposium on Genre Studies (pp. 2094–2105). Florianópolis, Brazil: UNISUL, Universidade do Sul de Santa Catarina. Barbeiro, L. (1999). Os alunos e a expressão escrita. Consciência metalinguística e expressão escrita. Lisbon: Fundação Calouste Gulbenkian. Serviço de Educação. Barré-de Miniac, C. (2000). Le rapport à l’écriture: Aspects théoriques et didactiques. Villeneuve d’Ascq (Nord): Presses universitaires du Septentrion. Bronckart, J.-P. (1996). Activité langagière, textes et discours pour un interactionisme socio-discursif. Paris: Delachaux et Niestlé. Bronckart, J.-P. (2006). Actividade de linguagem, discurso e desenvolvimento humano. Campinas, SP: Mercado das Letras. Cardoso, I., & Pereira, L. A. (2005). Inovar no ensino da escrita. A construção de uma relação positiva com a escrita. Palavras, 28, 63–79. Cardoso, I., & Pereira, L. A. (2007a). O Aluno—Sujeito, a escola e a escrita. O Eu—Jovem e o Outro—Aluno. In R. Bizarro (Ed.), Eu e o Outro: Estudos Multidisciplinares sobre Identidade(s), Diversidade(s) e Práticas Interculturais (pp. 174–187). Porto: Areal Editores. Cardoso, I., & Pereira, L. A. (2007b). Moi, l’écriture et les autres: Apports d’un dispositif didactique conçu par et pour l’approfondissement du rapport du Sujet à l’écriture [Electronic Version]. Colloque AIRDF (Association internationale pour la recherche en didactique du français): Didactique du français: le socioculturel en question, Lille, França. Retrieved July 30, 2008, from http://evenements. univ-lille3.fr/recherche/airdf-2007/PDF/Cardoso%20A8.pdf.
Teaching and learning writing in Portugal 69 Cardoso, I., & Pereira, L. A. (2007c). A relação dos alunos com a escrita (extra-) escolar: Uma experiência de investigação. Aprender. Comunicação e Educação. Educar para os Media, 31, 94–109. Cardoso, I., & Pereira, L. A. (in press). “Vou começar a resolver tudo à escrita”: Trabalhar, na escola, a relação com a escrita. In 2º Congresso sobre Aquisição da Linguagem. Guarda: Escola Superior de Educação da Guarda. Carvalho, J. A. B. S. D. (1998). A adequação do texto escrito à situação de comunicação. Da análise da questão ao desenvolvimento de capacidades em contexto escolar. Doctoral thesis. Universidade do Minho, Braga. Chapman, M. (2006). Research in writing, preschool through elementary, 1983– 2003 [Electronic Version]. L1-Educational Studies in Language and Literature, 6(2), 7–27. Retrieved July 31, 2006, from http://l1.publication-archive.com/start. Charlot, B. (1997). Du rapport au savoir. Eléments pour une théorie. Paris: Anthropos. Cordeiro, G. S., Azevedo, I. C., & Mattos, V. L. (2004). Trabalhando com sequências didácticas: Uma proposta de ensino e de análise de narrativas de aventuras de viagens. Calidoscópio, 2(1), 29–42. Graça, L., & Pereira, L. A. (2005). Algumas determinantes de sucesso no ensino da língua. Palavras, 28, 49–61. Graça, L., & Pereira, L. A. (2006). Écrire ensemble pour mieux réécrire. Caracteres, 24. Graça, L., & Pereira, L. A. (2008a). Como ensinar os alunos a escrever? Do que o professor de português já faz ao que pode vir a fazer? In I. Cardoso, E. Martins, & Z. Paiva (Eds.), Actas do Colóquio “Da Investigação à Prática: Interacções e Debates.” Aveiro: Departamento de Didáctica e Tecnologia Educativa, Universidade de Aveiro. Graça, L., & Pereira, L. A. (2008b). As práticas de ensino da escrita e os materiais didácticos: que (inter)relação?, que futuro? Paper presented at the I Encontro Internacional de Ensino do Português, June 30–July 1, 2008, Escola Superior de Educação de Coimbra. Graça, L., & Pereira, L. A. (in press). O trabalho do professor com o ensino da escrita do texto de opinião. In 2º Congresso sobre Aquisição da Linguagem. Guarda: Escola Superior de Educação da Guarda. Lahire, B. (1993). Culture écrite et inégalités scolaires. Sociologie de “l’échec scolaire” à l’école primaire. Lyon: Presses Universitaires de Lyon. M. E. (1991). Programa Língua Portuguesa. Plano de organização do Ensino- Aprendizagem. Ensino Básico 1º ciclo (vol. 2). Lisbon: DGEBS. M. E. (2000). Provas de aferição do Ensino Básico. 4º Ano. 2000. Língua Portuguesa. Matemática. Relatório Nacional. Lisbon: Departamento de Educação Básica. M. E. (2001a). Currículo Nacional do Ensino Básico: Competências Essenciais. Lisbon: Departamento de Educação Básica. M. E. (2001b). Provas de aferição do Ensino Básico. 4º e 6º Anos. 2001. Língua Portuguesa. Matemática. Relatório Nacional. Lisbon: Departamento de Educação Básica. M. E. (2004). Provas de aferição do Ensino Básico. 4º, 6º e 9º Anos. Análise Comparativa (2001–2003). Língua Portuguesa. Matemática. Relatório Nacional. Lisbon: Direcção-Geral de Inovação e de Desenvolvimento Curricular. M. E., & GAVE. (2007). Relatório nacional sobre os resultados das Provas de Aferição. Lisbon: Direcção-Geral de Inovação e de Desenvolvimento Curricular.
70 L. Á. Pereira et al. Penloup, M.-C. (1999). L’écriture extrascolaire des collégiens: Des constats aux perspectives didactiques. Paris: ESF. Pereira, L. A. (2000). Escrever em português. Didácticas e práticas. Porto: Edições Asa. Pereira, L. A. (2008). Escrever com as crianças. Como fazer bons leitores e escritores. Para crianças dos 0 aos 12 anos. Porto: Porto Editora. Pereira, L. A., & Cardoso, I. (in press). Jovens, alunos e relação com a escrita. In M. Matos (Ed.), Jovens e alunos do Ensino Secundário. Porto: Afrontamento. Pereira, L. A., & Graça, L. (2005). Interacção e mediação didáctica de géneros discursivos no ensino da escrita. In Actas do III Colóquio Nacional de Filosofia da Linguagem e Interacção. Porto Alegre: Universidade do Vale do Rio dos Sinos. Pereira, L. A., & Graça, L. (2007). Da conceptualização do contexto de produção e da sua produtividade na Didáctica da Escrita. In A. M. Guimarães, M. A. Coutinho, & A. R. Machado (Eds.), O interacionismo sociodiscursivo: Questões epistemológicas e metodológicas. Campinas: Mercado de Letras. Pereira, S. C. F. (2007). Efeitos da sequência didáctica em textos escritos no 1º Ciclo do Ensino Básico: O Relato. Dissertação Mestrado em Línguas no 1º Ciclo do Ensino Básico. Universidade de Aveiro, Aveiro. PISA, M. E., & GAVE. (2001). Resultados do Estudo Internacional PISA 2000. Primeiro Relatório Nacional. Lisbon: Ministério da Educação, Gabinete de Avaliação Educacional. Retrieved DATE, from www.gave.min-edu.pt/np3conten t/?newsId=33&fileName=primeiro_relatorio_nacional.pdf. Schneuwly, B., Dolz, J., Rodrigues Rojo, R. H., & Cordeiro, G. (2004). Gêneros orais e escritos na escola (R. Rojo & G. Cordeiro, Trans.). Campinas: Editora Mercado de Letras. Schneuwly, B., & Thévenaz-Christen, T. (Eds.). (2006). Analyses des objets enseignés. Le cas du français. Brussels: De Boeck & Larcier. Vygotsky, L. S. (1987/1934). Pensamento e Linguagem. São Paulo: Livraria Martins Fontes Editora.
6 Spanish research on writing instruction for students with and without learning disabilities Jesús-Nicasio García*, Ana-María de Caso-Fuertes*, Raquel Fidalgo-Redondo*, Olga Arias-Gundín, and Mark Torrance† *Department of Psychology, Sociology and Philosophy, University of León, Spain † Division of Psychology, Nottingham Trent University, UK
Introduction Traditionally, across a broad range of national and educational contexts, writing instruction has focused on features of the completed text: spelling, word choice, syntax, and genre. Some students manage to develop good writing skills through exposure to this kind of teaching. However, many students, and certainly those with more general literacy or learning difficulties, need broader and more sophisticated training. There is evidence that instructional programs that (1) focus not just on written products but also on the processes by which these are produced and (2) aim to develop students’ motivation and self-efficacy alongside their writing skills, tend to result in substantially greater gains in the quality than interventions that are wholly product-focused (Graham, 2006; Graham & Perin, 2007). The success or otherwise of instructional programs depends in part on the educational context in which the instruction takes place. This provides the main motivation for this chapter. The bulk of the studies cited in, for example, Graham’s (2007) meta-analysis of writing instruction evaluations were conducted in North American schools. Research has, of course, been conducted elsewhere but, partly because of publication in non-English journals, receives less attention. In this chapter we present a very brief overview of a number of studies conducted over the last eight years by García and co-workers from the University of Leon in northern Spain. Our work, all conducted with native Spanish school-age students, comprises both intervention evaluations and more general exploration of writing performance within the Spanish education system. Aside from the Spanish context, three broad themes mark out our research. First, and in common with other researchers, we aim to understand what helps students develop into independent competent writers rather than how to achieve short-term goals on specific tasks. Central to this is the concept of self-regulation: alongside understanding of written products and writing processes students need metaknowledge, self-awareness, and
72 J.-N. García et al. motivation to apply this knowledge flexibly across a broad range of writing contexts. Gains in self-regulation require specific teaching methods and are both harder to achieve and harder to assess than short-term gains in text quality. Second, we explore writing in both typically developing and struggling students (low-achieving students or students with learning disabilities, which we will refer to collectively as LD students). For practical reasons a focus on LD students is important. In primary1 teaching it is very easy for low-performing students to slip increasingly far behind in their writing development, particularly in contexts where instruction is product- centered. The fewer words a student gets down on the page the less there is to comment on, and so the less they are able to benefit from teacher feedback. As we discuss below, one of the ways round this is to provide feedback directly on process. Finally, and consistent with recognition of the importance of educational context, we aim to explore performance and evaluate interventions within the context of normal literacy classes and delivered by students’ normal classroom teacher. This introduces a certain amount of additional complexity in ensuring evaluations are properly controlled. However, we believe that this is outweighed by gains in generalizability. Also, and importantly, writing instructors, as end users of our research, are more likely to adopt novel interventions if we can demonstrate that they have already been implemented effectively by teachers like themselves in an everyday school context. In the remainder of this chapter we will first briefly describe the Spanish educational context in which we conduct our research. We will then summarize findings from research aimed at establishing the particular strengths and weaknesses of writers who are learning within this system. Finally, we will provide an overview of evaluations of interventions that we have developed to meet these writers’ needs.
The Spanish context Curricula in Spanish primary and secondary schools are dictated to some extent by central government, most recently by a wide-ranging Organic Law of Education, passed in 2006, which in turn reflects frameworks developed by the European Union. National law is interpreted at regional level—specifically the Castilla y León for most of the schools participating in our research. Literacy development is seen as a central focus of primary education and as less central in secondary schools, and reading is given greater emphasis than writing: of the 17 general aims for primary education promoted by the Castilla y León, only two relate to written production and these only tangentially. There is a required daily time for reading, which aims at developing good reading habits in students, but none for writing. Writing is discussed in more detail in specific notes on teaching Castillian language and literature. Here there is an emphasis on writing correctly at letter (handwriting), word (spelling), and text (order and
Spanish research on writing instruction 73 coherence) levels. Interestingly, guidelines also mention the need for some basic metalanguage to describe and conceptualize text and syntactic structure. The research described in this chapter dates back over the last eight or so years. During this period there has been little general change in approaches to writing instruction. In most schools writing has been a specific focus in the curriculum, but emphasis has tended to be on writing texts in specified genres with teachers tending to provide feedback exclusively on the appropriateness of text structure and correctness of syntax and spelling. Teachers typically do not make reference to or evaluate the strategies that students adopt in producing this text. The 2006 education law also introduced, for the first time in Spanish legislation, specific recognition of the need for provision for students with learning disabilities (LD), with diagnostic assessments in second years of both primary and secondary education. Prior to this there had been little or no specific focus on the needs of LD students. García et al. (2006) asked 131 Spanish kindergarten and primary school teachers about how they taught writing and found that nearly all used similar methods regardless of student ability.
Writing in the normal curriculum The research described in this section aimed to explore relationships among process, motivational, and text variables in students, both typically developing and learning delayed, within the normal curriculum. García and Fidalgo (2004a, 2004b) tested 1,688 typically developing students ranging in age from eight to 16 years. Students produced descriptive and narrative texts which were evaluated holistically, by counting particular coherence markers, and in terms of productivity (total amount of text produced). Students also completed a measure of writing metaknowledge comprising a number of open-ended questions about students’ writing experiences and self-perception of their writing ability (based on a questionnaire developed by Graham, Schwartz, & MacArthur, 1993). They found, as might be predicted, that as students get older, mechanical and other low-level aspects of the writing process feature less in how they talk about writing (and, by inference, in their metacognitive representation of writing processes). Older students were, therefore, more likely to talk about substantive features of their writing processes, including text structure and strategic activities like outlining and revision. Focus on mechanical and low-level aspects of text production was negatively associated both with students’ productivity and with the coherence of their text. Focus on substantive aspects of the writing process was, by contrast, a positive predictor of these variables. Older students were also more likely to express negative motivation in relation to their writing (e.g., “I find it difficult to work hard at my writing”), were more likely to engage in self-evaluation and, specifically, showed lower writing self-efficacy (e.g., “I do not think I can write very well”). As might be expected, motivation and self-efficacy both predicted writing performance.
74 J.-N. García et al. School-based studies have tended not to examine students’ writing processes, and methods that explore process (think-aloud, keystroke logging, guided retrospection) are not easily implemented within a classroom setting. We have been using a concurrent self-report method. Students are given a blank “writing log” comprising lists of, typically, seven different writing activities, represented by icons. While completing a writing task the class hears a short electronic tone played at random intervals of between 60 and 120 seconds. At the sound of the tone each student ticks the icon that best represents their current activity. This allows for an estimate not just of the total amount of time spent in each activity but also how different activities (and particularly, planning, drafting, and revising) are distributed through the course of completing the task. Using this method García and Fidalgo (2008) compared writing processes in 81 sixth-grade LD students with those of 80 typically achieving peers. They also explored students’ metaknowledge, writing motivation, and self-efficacy, and the quality and structure of their completed texts. As might be expected, LD students wrote texts that were both less well- organized and less coherent. They also displayed less well-developed writing metaknowledge, showing similar patterns to the younger writers described above. Less predictably, but consistent with findings from research conducted with students in the United States (e.g., Graham et al., 1993), LD students tended to have higher writing self-efficacy. Process measures indicated that total time-on-task tended to be greater for LD students, but this was due to more non-writing activity (e.g., staring out of the window). LD students spent less time drafting their text, and less time reading and changing what they had written. This is analogous to a developmental trend in the extent to which students revise: García and Arias- Gundín (submitted) assessed revision ability in the last two years of primary and first two years of secondary education (N = 958). As might be expected, older students tended to both make more revisions and generate a more coherent final product. As we discuss below, we have found some evidence that as students get older the extent to which they revise becomes an increasingly important predictor of text quality (Fidalgo, Torrance, & García, in press). There was, however, no difference between LD and non-LD groups in time spent formally planning their texts. Formal planning tends to be absent in all Spanish sixth-grade writers unless they have received specific, strategy-focused instruction (Torrance, Fidalgo, & García, 2007). Findings from research reported in this section suggest, therefore, two principles that can be carried forward into consideration of how best to teach writing. First, metacognitive, motivational, process, and final product variables are strongly interrelated. Second, the kinds of writing processes that are typically associated with mature, self-regulated writing tend not to be well-developed in either LD or non-LD students in the final years of primary school.
Spanish research on writing instruction 75
Interventions In this section we describe studies evaluating interventions targeted at developing students’ writing competence. With one or two exceptions our research has tended to focus on writing in the fifth and sixth grades—ages ten to 13 in the Spanish system. Sixth grade is the final year of primary education in Spanish schools. In evaluating interventions we make extensive use of a set of instruments that we call the Evaluación de los Procesos de Planificación y otros Factores Psicológicos de la Escritura (Assessment of the Planning Processes and other Writing-related Psychological Factors—EPPyFPE; García & de Caso, 2006b; García, Marbán, & de Caso, 2001).This comprises four sets of measures: a writing attitudes scale, a writing self-efficacy scale, a series of open-ended questions designed to assess students’ metaknowledge, and three writing tasks—a narrative, an expository essay, and a description. Interventions for LD students Our work with this group of students is based on two beliefs. First, LD students can, in most cases, learn strategies for producing high-quality texts. Second, development of these strategies requires interventions that motivate students to an extent that goes beyond that typically achieved within the normal curriculum. The studies we describe explore the validity of these convictions. Broadly, we have evaluated interventions which combined one or both of (1) teaching of specific cognitive strategies for regulating writing processes—specifically formal methods for planning and/or revising—and (2) specific strategies for developing students’ writing-related motivation. Interventions typically ran for around 25 sessions, and involved teaching with small groups of students during school hours but taken out of normal lessons. Sample sizes varied but were never fewer than 28 in the intervention condition and normally substantially greater. Our initial studies explored just the effects of training in planning strategies. García and Marbán (2003) found that teaching the kinds of planning strategy that would normally only be introduced much later in students’ education (e.g., Sorenson, 1997) resulted in significant gains in productivity and, to a lesser extent, in the coherence of students’ texts. It is possible that the relatively slight effects on text quality were due to a lack of reflexivity in students’ approach to their writing. In a second study (García & de Caso-Fuertes, 2002), we combined the same planning instruction with a method that required students to periodically stop (we used a colored sign) and apply a pre-learned schema to reflect on the current state of their text. Students again produced better text but on a transfer task we found no evidence of increased reflexivity. García and de Caso (2004) repeated the same intervention with a larger sample of students and added in specific tactics for developing student motivation. These were based in (1) rooting tasks in a practical, “real world” context,
76 J.-N. García et al. (2) maximizing the extent to which students interacted during sessions, and (3) both informal and formal methods for rewarding successful writing. This combined planning-and-motivation intervention resulted in improvements to the coherence of students’ texts, though not this time in their productivity. After training, students in the normal-curriculum control expressed a less positive attitude toward writing than they had at baseline, but attitude improved slightly within the intervention group. We did not, however, find effects on motivation or self-efficacy. One possible reason for this is simply that 25 sessions is inadequate to rewind a long history of failure relative to more able peers. The lack of self-efficacy improvement may also, or alternatively, be due to expectations keeping pace with ability: as students become more able writers they also become more aware of how much better their writing could become. García and de Caso (2006b) brought together all three interventions discussed so far—planning only, planning plus reflexivity training, and planning plus motivational training—in a single evaluation. In addition to teaching planning strategies, all of the interventions also included some instruction in text structuring (e.g., use of paragraphing) and in strategies for revising completed drafts. All three interventions showed benefits for text coherence and structure and for productivity compared to normal- curriculum controls. Productivity increased most in the planning-only condition, and least in the planning-plus-motivational training condition. There were no differential effects on either coherence or structure. Effects on motivation and attitude were more mixed. However, there was again no evidence of an effect for motivational training. Taken together, these studies suggest strong evidence for the value for LD students of training in cognitive self-regulatory strategies. Failure to affect students’ motivation, and particularly their belief in their own ability, was, however, disappointing. We therefore developed a program, described in García and de Caso (2006a), that specifically targeted the four elements that Bandura (1997) argues are necessary for the development of self-efficacy. Instruction comprised ten sessions, delivered this time by the researchers, in which students were provided with a pleasant working environment (to induce a positive affective state), were taught to verbally self-reward for personal effort, were given manageable tasks and were reinforced for good performance, and had their attention drawn to successful peers. Alongside these tactics for developing students’ self-efficacy, they were taught the same planning strategies as in previous studies (but not the “stop and think” reflexivity training). Students in the intervention condition showed very substantial gains in the coherence and structure of their texts, with substantially greater effect sizes than those found in previous studies. Process information from writing logs indicated that intervention students spent proportionally less time reading supporting materials and more time writing text than students studying under the normal curriculum. These effects remained two months after the end of the intervention. Again, however, self-efficacy seemed resistant to intervention.
Spanish research on writing instruction 77 Finally in this section, García and Fidalgo (García & Fidalgo, in press, 2006) report a comparison between the Self Regulated Strategy Development (SRSD) model that has been demonstrated as effective in teaching LD students in a North American educational context (e.g., Graham, Harris, & Macarthur, 1993) and an intervention based on Schunk and Zimmerman’s social-cognitive model of sequential skills acquisition (SCM; Schunk & Zimmerman, 1997). Both approaches to instruction share a focus on teacher modeling, social feedback, and scaffolding, and both have as a central theme the need for students to develop explicit strategies to support the writing process. They differ in that the SCM-based program involves a greater focus on modeling with extensive use of think-aloud by the teacher, by students working in pairs, and by the student working alone. Comparison of the two approaches was motivated by the fact that, unlike SRSD, the SCM program had not been previously evaluated with an LD population. We found that both interventions resulted in marked improvements in the structure, coherence, and overall quality of students’ texts. Process measures indicated that both interventions resulted in increases in time spent writing and revising, but only SRSD resulted in increases in time spent planning. Both groups showed slight gains in self-efficacy, but these were only statistically significant in the SCM condition. Findings from our research with students with LD and low-achieving students therefore provide clear support for our first conviction: LD students can show marked improvements in the quality of the texts they produce if they learn to strategically regulate their writing processes. We have not found such strong support for our second conviction, intuitive though it seems, that writing-skill development in LD students is contingent on developing their motivation. Interventions appear to have been successful despite not directly targeting motivation, and also show gains in text quality independently of students increasing in self-efficacy. Interventions for typically developing students While research with students with learning difficulties has been the major focus of our work over the last few years, we have also conducted several studies exploring the effects of specific kinds of intervention with typically developing students (Fidalgo & García, in press; Fidalgo et al., in press; García & Arias-Gundin, 2004; García & Arias-Gundín, submitted; García & Rodriguez, 2007; Torrance et al., 2007). We will focus on two of these. As in our research with LD students, interventions were delivered by students’ normal literacy teachers and comparison was made with normal- curriculum controls. Fidalgo and Torrance (Fidalgo et al., in press; Torrance et al., 2007) evaluated a strategy-focused, ten-week writing instruction program for sixth-grade students, which they called Cognitive Self Regulation Instruction (CSRI). This involved a combination of product-focused teaching and teaching of metacognitive strategies for managing production, focusing
78 J.-N. García et al. principally on planning and revising. Like the SCM program, modeling was a central component in the way instruction was delivered with extensive use of think-aloud. We found large positive effects of the intervention on holistic measures of text quality, structure, and coherence, and in students’ use of several coherence markers typically associated with mature, reader-focused text. These effects were sustained in a follow-up test at six weeks and decreased, but were still large, two years after students had completed the intervention and returned to instruction within the normal curriculum. Process measures taken from writing logs indicated that students also substantially increased their tendency to plan their text. One obvious conclusion from these findings would be that during the program students learned to plan, and that as a result of planning the quality of their text improved. This was not, however, borne out by analyses exploring the relationship between product and process: we found only a weak relationship between the amount of time students spent planning and the quality of their text, both immediately after the intervention and at six weeks. Two years after the intervention the factor best predicting text quality was the extent to which students spent time reading through the text that they had written, a factor that was not affected by the intervention, and the relationship between planning and text quality was not statistically significant. Therefore, although strategy-focused teaching appears effective in developing typically developing Spanish students’ writing ability, the mechanisms by which this is achieved remain unclear. Following Fidalgo and Torrance’s failure to see an effect of the CSRI intervention on students’ tendency to revise their texts, García, Arias- Gundín, and Torrance (submitted) developed and evaluated instruction focused specifically on revision skills. This involved giving students existing texts to revise and providing—and then gradually removing—prompts for specific kinds of revision activity. We evaluated three different versions of the intervention focusing on surface-level revision, deep (message) level revision, and a combination of both. The program was implemented with students in their second year of secondary education (aged 13 to 15 years) and was evaluated in terms of students’ performance on a task that involved rewriting an existing text. In all three intervention conditions rewritten texts were both longer and substantially more coherent relative to those produced by students in a normal-curriculum control group performing the same task. This then suggests that, in older Spanish students at least, revision is teachable and can give substantial gains in text quality.
Conclusions Specific findings aside, this brief overview of our research points to two more general conclusions. The first concerns methods. The final goal of writing instruction is developing in students the ability to produce good- quality text and to do so independently of external prompts. A range of instructional approaches appears to help students to do this, some of
Spanish research on writing instruction 79 which are reviewed in this chapter, and many more in Graham and Perin’s recent meta-analysis (2007). We know much less about the mechanisms by which these interventions succeed. An understanding of these mechanisms is in itself a worthwhile aim. A more thorough understanding of why interventions work is also necessary if teachers working with diverse students and in diverse educational contexts are to make informed choices about instructional tools. One of the strengths of our research, we believe, is that we have tended to evaluate the effects of interventions across several different dimensions. Our studies have assessed not only writing quality but also process (using the writing-log method), motivation and self-efficacy, and writing-related metaknowledge. Although for reasons of brevity we have tended to omit some of the detail of our findings, it is clear that our research suggests a rather complex pattern of effects of different interventions across these dimensions. We believe that all four of product, process, metaknowledge, and motivation need to be considered if we are to move beyond “what works” to an understanding of “why.” Our second conclusion concerns findings. As has been found in a number of studies conducted elsewhere (Graham, 2006) and as we have now demonstrated in the Spanish context, strategy-focused writing instruction is effective in developing students’ writing ability. This is true for typically developing students, for students who have, for whatever reason, failed to develop their writing skills under the normal product-focused curriculum, and for students with wider learning difficulties. Our knowledge may at present be lacking as regards the mechanisms by which they work. However, findings reported in this chapter demonstrate that the success of strategy-focused programs extend to at least one other educational context. Our research underlines existing findings: successful writing instruction requires a focus not only on written products but also on the processes by which these are produced.
Acknowledgment The research reported in this chapter was supported by funds from Dirección General de Investigación del Ministerio de Educación y Ciencia de España General Research Board of the Spanish Ministry of Education and Science) competitive project MEC SEJ 2007-66989 (2007–2010) and with FEDER funds from the European Union both awarded to the principal researcher (J.-N. García). Correspondence should be directed to JesúsNicasio García, Department of Psychology, Sociology and Philosophy. Developmental and Educational Psychology Area. Email: jn.garcia@ unileon.es; Campus de Vegazana s/n, 24071, León, Spain.
Note 1 Here and throughout we use “primary” to refer to teaching of children between, in the Spanish context, ages six and 12.
80 J.-N. García et al.
References Bandura, A. (1997). Self-efficacy: The exercise of control. New York: Freeman. Fidalgo, R., & García, J. N. (in press). Metacognitive intervention in writing composition. Revista de Psicologia General y Aplicada. Fidalgo, R., Torrance, M., & García, J. N. (in press). The long-term effects of strategy-focussed writing instruction for grade six students. Contemporary Educational Psychology. García, J. N., & Arias-Gundín, O. (2004). Intervention in writing composition strategies. Psicothema, 16(2), 194–202. García, J. N., & Arias-Gundín, O. (submitted). How do students from the 5th grade of Spanish primary education through to the 2nd grade of Spanish secondary obligatory education revise? García, J. N., Arias-Gundín, O., & Torrance, M. (submitted). Efficacy of an instructional program in the textual revision process by means of dynamic assessment. García, J. N., & de Caso, A. M. (2002). Is it possible to improve writing composition in learning disabilities (LD) and/or low achievement (LA) students without changes in reflexivity toward writing? Psicothema, 14(2), 456–462. García, J. N., & de Caso, A. M. (2004). Effects of a motivational intervention for improving the writing of children with learning disabilities. Learning Disability Quarterly, 27(3), 141–159. García, J. N., & de Caso, A. (2006a). Changes in writing self-efficacy and writing products and processes through specific training in the self-efficacy beliefs of students with learning disabilities. Learning Disabilities: A Contemporary Journal, 4(2), 1–27. García, J. N., & de Caso, A. (2006b). Comparison of the effects on writing attitudes and writing self-efficacy of three different training programs in students with learning disabilities. International Journal of Educational Research, 43(4– 5), 272–289. García, J. N., & Fidalgo, R. (2004a). Diferencias en la conciencia de los procesos psicológicos de la escritura: mecánicos frente a sustantivos y otros. (Differences in awareness of writing cognitive processes: substantive vs. mechanical and other, in 8- to 16-year-old students.) Psicothema, 15(1), 41–48. García, J. N., & Fidalgo, R. (2004b). El papel del autoconocimiento de los procesos psicológicos de la escritura en la calidad de las composiciones escritas. (The role of self-knowledge of writing cognitive processes in the quality of written products.) Revista de Psicología General y Aplicada, 57(3), 281–298. García, J. N., & Fidalgo, R. (2006). Effects of two types of self-regulatory instruction programs on students with learning disabilities in writing products, pro cesses, and self-efficacy. Learning Disability Quarterly, 29(3), 181–211. García, J. N., & Fidalgo, R. (2008). The orchestration of writing processes and writing products: A comparison of learning disabled and non-disabled 6th grade students. Learning Disabilities: A Contemporary Journal, 6(2), 77–98. García, J. N., & Fidalgo, R. (in press). Writing self-efficacy changes after a cognitive strategy intervention in students with learning disabilities: The mediational role of gender in calibration. Spanish Journal of Psychology. García, J. N., & Marbán, J. M. (2003). El Proceso de composición escrita en alumnos con DA y/o BR: Estudio instruccional con énfasis en la planificación. Infancia y Aprendizaje, 26(1), 97–113.
Spanish research on writing instruction 81 García, J. N., Marban, J. M., & de Caso, A. M. (2001). Evaluacion colectiva de los procesos de planificacion y factores psicologicos en la escritura (Collective assessment of the planning processes and other psychological factors in writing). In J. N. García (Ed.), Dificultades de aprendizaje e intervencion psicopedagogica (Learning disabilities and psychopedagogical interventions). Barcelona: Ariel. García, J. N., Pachecho, D. I., Díez, C., Robledo, P., Martínez-Cocó, B., Rodríguez, C., et al. (2006, 26–28 October). The role of the teacher’s practice in teaching writing in relation to writing products and processes in students with and without LD. Paper presented at the 15th Congress on Learning Disabilities Worldwide, Boston. García, J. N., & Rodriguez, C. (2007). Influence of the recording interval and a graphic organizer on the writing process/product and on other psychological variables. Psicothema, 19(2), 198–205. Graham, S. (2006). Strategy instruction and teaching of writing: A meta-analysis. In C. MacArthur, S. Graham, & J. Fitzgerlad (Eds.), Handbook of writing research (pp. 187–207). New York: Guilford Publications. Graham, S., Harris, K. R., & Macarthur, C. A. (1993). Improving the writing of students with learning-problems: Self-regulated strategy-development. School Psychology Review, 22(4), 656–670. Graham, S., & Perin, D. (2007). A meta-analysis of writing instruction for adolescent students. Journal of Educational Psychology, 99(3), 445–476. Graham, S., Schwartz, S. S., & MacArthur, C. A. (1993). Knowledge of writing and the composing process, attitude toward writing, and self-efficacy for students with and without learning disabilities. Journal of Learning Disabilities, 26(4), 237–249. Schunk, D., & Zimmerman, B. (1997). Social origins of self-regulatory experience. Educational Psychologist, 32, 195–208. Sorenson, S. (1997). Student writing handbook: The only complete guide to writing across the curriculum (3rd ed.). New York: Macmillan. Torrance, M., Fidalgo, R., & García, J. N. (2007). The teachability and effectiveness of cognitive self-regulation in sixth grade writers. Learning and Instruction, 17(3), 265–285.
Part II
Writing education in political and historical contexts
7 Writing, from Stalinism to democracy Literacy education and politics in Poland, 1945–1999 Cezar M. Ornatowski San Diego State University
Introduction The purpose of this chapter is to examine the relationships between literacy education and its historical and sociopolitical context in Poland from the onset of Stalinism to the transformational education reforms of 1999. As I will suggest in this chapter, inquiry into literacy education in totalitarian and transformational contexts such as Poland may not only be useful to educators consulting or teaching in such contexts or to teachers in the United States and elsewhere whose students come from a wide variety of educational backgrounds; it also has the potential to extend the current paradigms for cross-boundary research in writing studies and complicate some assumptions that have underpinned such research as well as discussions of democratic literacy education in the United States. The idea that rhetoric and rhetorical education have a reciprocal relationship to their political context has been central to rhetorical studies. In the Politics, Aristotle argues that what “most contributes to the permanence of constitutions is the adaptation of education to the form of government”; hence, the young should be “trained by habit and education in the spirit of the constitution, if the laws are democratical, democratically, of oligarchically, if the laws are oligarchical” (V, 9, 1310a 10–20, p. 237). Modern rhetorical theory recognizes that political regimes imply also rhetorical regimens, insofar as they imply specific ways of speaking, arguing, and thinking that foreclose, limit, or proscribe other ways of speaking, arguing, or thinking (Farrell, 1993). As Thomas Farrell (1993) has argued, “no culture or public life project can survive for long without some form of rhetorical practice, some coherent, symbolic manner of securing collaborative public action” (p. 9). In turn, as James Berlin has argued, “no classroom pedagogy can long survive without in some way responding to its historical conditions” (1990, p. 185). In American writing studies, research into the relationship of literacy education (including the teaching of writing) to its sociopolitical contexts has moved, in the last decade, from a primarily Anglo-American orientation (Berlin, 1990; Murphy, 1990; Shannon, 2001) toward a more internationalist one, following Mark Schaub’s call for research into “local models
86 C. M. Ornatowski [of writing instruction] in non-US contexts” (2003, p. 91). Not all of this research has focused on the relationship between writing pedagogy and politics per se; it was largely the influence of postcolonial studies that has led rhetoric and composition scholars to consider specifically the “connections between the project of education and the structures of power, between formal schooling and the cultivation of compliant subjectivity” (Bahri, 2004, p. 69). Under this influence, scholars have sought “to understand the ways in which students (and student writings) are variously constructed, subjugated, and turned into obedient subjects, both within and outside the academy, and to find ways of enabling resistance to such forces” (Lunsford & Ouzgane, 2004, p. 1). The “postcolonial,” however, has generally been identified (perhaps because of the location of the leading scholars) with US and Western (British and French) domination and with the “colonizing” effects of “English” (both as dominant language and academic endeavor) on student readers and writers, especially in terms of issues of race, ethnicity, gender, class, language, or sexual orientation. Few voices have come from territories dominated by other sorts of “colonial” centers and other sorts of “colonization,” although bell hooks and Deepika Bahri have notably called for the inclusion of other areas and kinds of colonization (for instance, the Islamic colonization of North Africa). I suggest that czarist Russia and the former Soviet Union, with their extensive empire of absorbed, dominated, or “satellitized” peoples, represent one such center. In fact, some Polish scholars have recently begun to refer to Poland as “postcolonial” (Janion, 2007) in a dual sense: as having been dominated for long stretches of its history by a foreign political “center” (from 1772 to 1918 by Russia, Germany, and the Austro-Hungarian Empire; later by the Soviet Union) and as located on, and shaped by, the historically fluid boundary between, successively, Rome and Byzantium, Christianity and Orthodoxy (and to an extent also Islam), “East” and “West,” and representing, in a historical as well as cultural and political sense, a “younger” (and implicitly—in terms of self-perception and global image—“secondary” and “backward”) Europe (Kloczowski, 1998). Poland thus represents a site that extends studies of writing instruction beyond the borders of the traditionally conceived “West” and into other histories and sorts of “colonization,” as well as other histories of resistance, including resistance against ideological colonization that, from 1949 until 1989, “constructed, subjugated, and turned” entire peoples into “obedient subjects.” The issues such studies raise are different than those raised under the banner of “transnationality,” defined recently by Wendy Hesford and Eileen Schell as referring to “the movements of people, goods, and ideas across national borders and . . . used to highlight forms of cultural hybridity and intertextuality” (2008, p. 463). In places such as Poland, such concerns are still secondary to issues of national and personal identity, and to what Patrick Shannon has referred to as the “dialectic of state and freedom”—issues that lay also at the foundation of the American
Writing, from Stalinism to democracy 87 Republic and that had shaped early American literacy education (Shannon, 2001).
Literacy education in Poland: historical context In Poland, literacy education has, for the last 200 years, been intimately connected to “freedom” and nation-building. Prior to 1918, when the Polish state was brought back into political existence by the Treaty of Versailles after a 124-year absence from the map of Europe, Poland did not have a national education system. Throughout Poland’s turbulent history, literacy education has served as a vehicle for a variety of historical and ideological projects: denationalization under foreign occupation; preservation of collective memory and struggle for political freedom and national liberation (in the Polish context, the two were usually conflated); political and cultural reconstruction (most recently in 1919, 1945, and after 1989); and ideological transformation (after 1949, with the advent of Stalinism, and after 1989, following the democratic transition) (Axer, 2007). During periods of foreign occupation, school literacy education typically served the interests of foreign empires, while during the real-socialist period it was the vehicle of ideological indoctrination; in both cases, school literacy contrasted with the national tradition preserved, also largely through literacy, in the home (as well as in samizdat publications and clandestine teaching) (Axer, 2007). In this sense also, Poland’s situation, at least as far as literacy education is concerned, may be regarded as “post-colonial.” For historical reasons, the subject “Polish” has, over the last century, been central to the elementary and secondary education curriculum (unlike in the United States, there has never been systematic instruction in reading and writing in higher education) in terms of the total number of instructional hours, the volume of applicable official directives and guidelines, and the number of textbook editions. This curriculum has traditionally consisted of three major components: reading and literature, the study of language, and speech and writing. Pedagogically, the system was characterized by integration, gradation, and continuity, with the overarching aim to develop facility, fluency, appropriateness, and correctness in the native language (Kijas, 1968). While the general framework of this curriculum changed little between the 1930s and the 1990s, after 1949 all of its components received a strong ideological inflection as literacy education became the primary vehicle for the transformation of consciousness in the service of the Marxist–Leninist state.
Literacy education under real socialism The consolidation of Stalinism in 1948 marked the beginning of a major “ideological-educational offensive” aimed at forging a new “socialist consciousness” that consisted of specific beliefs, values, and attitudes toward reality (Radziwill, 1981). Literacy education was considered especially
88 C. M. Ornatowski serviceable in this historic project. Education authorities quoted Joseph Stalin’s dictum that “language as a tool of communication is at the same time a tool of struggle and social development” (Stalin, quoted in Kulpa, 1955, p. 5). Although Stalinism as such ended in 1956, official decrees in the early 1960s explicitly confirmed the role of “Polish” as providing the “ideological foundation” for “life in socialist Poland,” while the sciences and other “hard” subjects constituted “technical” preparation for productive participation in the economy (Ministerstwo, 1963). For the duration of the real- socialist period, this “ideological foundation” consisted of two major components: a “scientific” (that is, dialectical and historical materialist) viewpoint and “socialist morality,” which included “socialist internationalism” (primarily allegiance toward the Soviet Union and other countries of the Soviet bloc, as well as “solidarity” with “anti-imperialist” and “liberation” movements around the world); “class consciousness”; a “socialist attitude” toward property, work, and people; collectivist orientation in thought and behavior; concern with “social justice”; a conviction of the superiority of the socialist system over capitalism; and active participation in “socialist construction” (Wojdon, 2001). The manner in which literacy education was mobilized in service of these precepts, and then transformed in the course of political transition, provides interesting insights into the relationship between literacy education and its political context. Most elementary textbooks for the subject “Polish” over the period of real socialism fell into the category of what in the United States are referred to as “rhetorics”: a combination of readings, questions for analysis, suggestions for activities and/or writing, and information and exercises on various aspects of language (spelling, grammar, style, and so on, depending on grade level). The textbooks were typically organized around a set of relatively standard themes, which, not surprisingly, were strongly ideologically slanted. The following sections, for instance, comprised a 1970 sixth-grade “Polish language” textbook: struggle for “social liberation,” World War II, Polish–Soviet friendship, work, vacations, personal character, patriotism, technology, the arts (press, music, radio, TV, books), travel and exploration (discoveries, mostly of Soviet explorers and “pioneers”), space (Soviet triumphs in space, with no mention of the United States), and sports (the triumphs of athletes from Soviet-bloc countries) (Dembowska, Saloni, & Wierzbicki, 1970). The theme of “travel” focused on Africa, colonialism, racial discrimination (the only text about the United States was about the terror of the Ku Klux Klan), and Soviet “conquest” of nature. No Western author was represented; in another 1970 textbook, this one for seventh grade, the only selection by a Western author was a fragment about Lenin from John Reed’s Ten Days That Shook the World (Wieczorkiewicz, Jaworski, & Rurawski, 1970). The activities that accompanied the readings placed little emphasis on the self or on individual opinion, except as a function of an ideologized, public, and correct “attitude,” which was to be discerned in the texts read or assumed and expressed in relation to events or phenomena, including
Writing, from Stalinism to democracy 89 one’s personal life.1 Consider the following “questions for analysis” from a 1970 seventh-grade textbook: “What do you learn from [the reading] about the life of female workers in the second half of the 19th century?”; “Based on the stories ‘Poverty’ and ‘Johnny the Musician’ talk about the fate of poor children in Poland in the second half of the 19th century”; “Tell about how we celebrate the May Day holiday today” (the last following a reading describing the bloody May Day clashes of demonstrators and police under the czarist regime) (Wieczorkiewicz et al., 1970). Even seemingly autobiographical tasks, such as recalling an “important moment” in one’s life, de facto called for an articulation of the desired “attitude” (for instance, the moment one realized the superiority of collective over individual effort). The collectivist orientation was reflected in the titles of elementary school textbooks: With Collective Energy (1949), In Our Collective (1966), In Our Fatherland (1971), Together, Young Friends (1975), We Grow Up Together (1979), Let’s Grow Up Together (1982). By contrast, compare post-1989 textbook titles: Landscape with a Smile (1993); Tomorrow I’ll Go into the World (1998); I Understand the World (2001); I Like That (2001). Writing assignments and activities also involved strongly ideologized topics and called for articulation of desired attitudes. A 1969 advice book on teaching writing (one of very few texts explicitly devoted to writing as such) quoted Soviet educator Vladimir Nikolski to the effect that “[a] topic presents the task of knowing [an aspect] of reality” (Pojawska, 1969, p. 101). Writing topics were seen in terms of a dual thrust: epistemic and ideological. A good topic had to be univocal, with its epistemic and ideological aspects pointing in the same direction. For instance, a topic such as “Two Worlds in Janko Muzykant” (“Johnny the Musician,” a nineteenth- century novella about a poor village boy whose musical talent was wasted due to conditions of serf life) was not acceptable because it left the student a choice of direction, but “Tell about the life and revolutionary deeds of Karol Swierczewski” (a communist war hero) was good because it pointed in a definite direction and its epistemic and ideological aspects coincided (Pojawska, 1969). Every composition, regardless of the topic or genre, was expected to have a clear “leading line.” This “line” began with the title and proceeded through to the conclusion, which was expected to connect the “line” to the writer’s own “attitude.” The line gave coherence to the composition as well as subordinated all information and detail to a central “idea” that ran through the text and constituted its, ultimately political, “message.” As one expert explained in the mid-1950s: Grasping the directional line while analyzing the topic “spring in the fields of the Collective Farm in Lubartow” involves noting the representation of nature: warmth, birds, greenery; human powers: collective, conscious effort—and the power of the machinery, it is thus at the
90 C. M. Ornatowski same time a conscious aiming to show the beauty of nature’s blossoming in relationship to the usefulness of human and machine labor, which are directed by the will of the collective. (Kulpa, 1955, p. 26) The example shows a specific ideological appropriation of both the principle of textual “coherence” and of the “rhetorical mode” of description. Compare the above conception of description with, for instance, George Hillocks’ strategy of teaching description by giving students seashells in a basket, having students pick one seashell out, and then having them describe their specific seashell in such a way that other children would be able to pick it out of a basket of other shells (Teaching). Hillocks offers this activity as an example of a pedagogy he calls “inquiry,” which, in his meta-analytic study of writing pedagogies, he found to be most “effective” at increasing the quality of student writing (Research; “synthesis”). Hillocks’ pedagogy, with its focus on individuality, specificity, and unique detail would most likely be considered “ineffective” in a Stalinist school. The example reveals both the ideological character of “quality” in writing and the existence of implicit ideological dimensions of texts and pedagogies. Another common assignment, the “life story” (Pol. zyciorys) provides an example of a political appropriation of genre. A “life story” was a personal narrative, but one oriented toward ideological and political (self-) revelation. Bernardo Bertolucci’s 1987 film The Last Emperor shows Emperor Puyi arrested by the Chinese communists and put in a re- education camp, where he is immediately given pen and paper and asked to write his “life story,” the successive drafts of which are torn up as inadequate until he comes up with a version more to the authorities’ liking and one that amounts to an ideological “confession”—a visible sign of compliance. Denizens of real-socialist Poland had to submit their “life stories” to the appropriate authorities on various occasions (such as applying for permission to travel abroad) as a way of both demonstrating the right “attitude” and submitting themselves to official scrutiny. The genre of the life story was practiced in every school grade, beginning in the early grades as a simple narrative of dates and events and becoming more “ideologized,” and politically controlling, in higher grades. Life stories always had to be signed. Many literacy activities were directed at fostering the involvements and behaviors that the authorities wanted to elicit and promote: participation in state ceremonies (such as the May Day parade), commemoration of official occasions (by, for instance, writing and publicly delivering an appropriate speech), or reading or watching official pronouncements in the media (for instance, Stalinist-era “grammatical” exercises asked students to read a party leader’s speech in a newspaper and underline or copy specific grammatical or stylistic features) (Klemensiewicz & Zlabowa, 1949). “Participation” was fostered, elicited, and certified through such activities
Writing, from Stalinism to democracy 91 and genres as reports, protocols, descriptions, speeches, articles for class wall bulletins, performances, or summaries of press and media coverage. In this way, literacy education, as official directives had it, “connected” school learning with the “life of the nation” and made education “practical” (as opposed to “bourgeois” education, which was supposed to be abstract, solipsistic, and divorced from the “life of the people”). While such activities may appear similar to current US trends (such as community service learning or environmental education) the political/ regime context made their meaning different. Polish students and teachers acted primarily in response to exigencies created by the authorities. Unlike American students and teachers, they could not initiate actions on their own (for instance, undertake to measure pollution in a local river or clean up a city park—such activities had “political” implications), nor could they respond in unexpected ways or refuse to respond altogether. “Participation” was mandatory and limited to official organizations, institutions, and organized channels, and took only officially sanctioned forms. In these ways, literacy activities and pedagogies received their character and specific meanings through their location in and functional connection with a specific ideological and political/regime context. Literacy activities promulgated and elicited subscription to the dominant ideology and utilized and reinforced the formal organization of society (as opposed to, for instance, non-formalized and spontaneous structures of “civil society”). In fact, one of the major elements of the “socialist value system” whose development was the avowed goal of education was the subordination of the individual to society, which de facto meant to the state (Wojdon, 2001).
Toward the political transition The economic and political collapse of the late 1970s and early 1980s demonstrated, among other things, the failure of the education project described above. In spite of official calls for the “activization” of students, “socialist” pedagogy ended up promoting passivity, conformism, disengagement, cynicism, and knee-jerk responses to predictable stimuli. In the face of the mounting economic problems, the authorities attempted to introduce changes to stimulate individual initiative, genuine engagement, and creative thinking. By the mid-1980s, ministerial directives, while still emphasizing the task of “shaping socially valuable attitudes,” also stressed the need to train students in “argumentation” and “defending their own point of view,” and left teachers significant leeway in the classroom (Instytut, 1985). A particularly dramatic sign of change was a 1985 eighth-grade textbook devoted specifically to speech and writing. The book focused on the development of “individual style” in speaking and writing, including avoidance of “sloganeering,” “ready-made language,” “empty generalizations,” and “meaningless abstractions” (Nagajowa). It is difficult not to see irony in such exercises as “show that the following slogans contain
92 C. M. Ornatowski falsehoods. Convert them into statements that reflect actual reality: ‘All students in our class try hard to get good grades.’ ‘We will earnestly attempt to improve the learning outcomes in our class’ ” (Nagajowa, p. 8). The statements the students are asked to critique here resemble official rhetoric of the day. Even more provocative were such examples of “slogans” as “Everything for workers” or “Out-of-control youths destroy collective property” (the political context for this was an intensive official media campaign against the “solidarity” movement, which accused workers of, among other things, “hooliganism” and destruction of property). The book characterizes these slogans as, respectively, exaggeration and an “outright lie.” Even though the book never mentions “rhetoric” as such, the examples attempt to promote critical rhetorical analysis under the guise of “personal style.”
Literacy education after 1989 The political transition of 1989 shifted the focus of literacy education, on the one hand, toward the individual, personal, and unique, and, on the other hand, toward sociality conceived of in terms of civic relationships between autonomous moral agents, not as a function of the experience of “class” or a unified “nation” or “people.” New guidelines for elementary education, in the section entitled “Creating a System of Values,” emphasized the need for exploration of a wide variety of “religious and philosophical attitudes,” of “the relationship of a human person to the world and to one’s own existence,” and of “the problems of moral choices and the responsibility of a person to other people and to society” (Ministerstwo, 1990). The textbook title that perhaps most emphatically captured the spirit of transition was Imagination Liberated: The Reality of the Person and the Unreality of Doctrine, a 1992 high school textbook produced by a publisher with an equally symbolic name: The New Era (Chodzko, 1992). More comprehensive changes came with the broad educational reforms of 1997–2001 (especially the reform of 1999), aimed at bringing Poland into “Europe” and into the twenty-first century (Ksiazek, 2001). The goals of literacy education were defined as, on the one hand, helping students find their own answers to who they are and how to live, to give them “roots and wings” (Ksiazek, 2001, p. 7), and, on the other hand, developing “facility in speaking, listening, reading, and writing in diverse communicative situations, private and public, especially ones important for life in a democratic and civic society” (Podstawa Programowa Ksztalcenia Ogolnego dla Szescioletniej szkoly Podstawowej i Gimnazjum, 1999, p. 1; quoted in Tabisz, 2006, p. 10). The reform assigned a prominent place to “rhetoric,” without, however, defining the term. It also decentralized the approval and publication of textbooks, which resulted in a rapid commercialization of the textbook market and a flood of new products of varying quality. In spite of these changes, however, many elements of the literacy curriculum remain the same, although they have lost their “meanings.” Text-
Writing, from Stalinism to democracy 93 books continue to insist on a clear “line” through the text and an explicit final “judgment” or statement of “personal attitude,” except that now there is not one “correct” attitude; thus, what used to be a key element of ideological indoctrination is simply a formal requirement. “Characterization” remains a popular school genre, although it has lost the ideological character it had under real-socialism. Even the “life story” is still taught, although it has lost its ideological utility (ironically, in the competitive context of the new capitalism, the resumé is not usually taught). The topics for writing have changed, however; today, they tend to be open (not presupposing one “correct” answer) and involve choice as well as argument and reasoning (consider these topics from a state high school graduation exam: “European Union: Opportunity or Threat to Poland,” “Are People Today Tolerant?”) (Poznanski, 2003).
Conclusion In their introduction to an important collection of essays on writing from the perspective of “activity theory,” Charles Bazerman and David Russell suggest that it is the context of human activities that gives meaning to texts and discourse and argue that “writing research needs to move beyond texts as ends in themselves” to consider how writing is “embedded in . . . uses and interpretations of texts and the creation of meaning and consequence in carrying out the work of the world” (2003, p. 4). In this chapter, I extended Bazerman and Russell’s suggestion to examine the embedding of literacy education (including texts and writing) in the specificity of the ideological and political/regime context of real-socialist Poland and the Polish transition. In this context, the meanings of texts and writing (as well as of reading and writing activities) were connected to their “uses and interpretations” within the historical and ideological project of “building socialism” and, after 1989, building liberal democracy. The Polish case shows that the meanings of educational practices derive from their “uses and interpretations” within the ideological and political/ regime context. In a different context, practices may assume different meanings or become merely formal requirements. In the early and mid- 1990s, literacy reformers and textbooks in Poland borrowed wholesale from Western models.2 A decade later, it is becoming clear that the task of educational reform is even more difficult: to work out a model of democratic literacy education in realistic response to local histories, concerns, and meanings. Postcolonial and “transnational” scholarship often sees national orientation as oppressive and limiting. However, it is important to remember that it is in the name of “national liberation,” including national sovereignty and national identity, that many anti-colonial and anti-totalitarian struggles have been conducted. While some liberal pedagogies in the United States oppose “Western individualism,” one must remember that most major contemporary totalitarian projects (Fascism, Soviet-style
94 C. M. Ornatowski “real-socialism,” religious fundamentalism) have marched under the banner of such grand abstractions as “the people,” the “community,” or “social justice,” which have no less potential than colonialism to turn people, and students, into “obedient subjects.” The political, economic, and ultimately human failures of such projects, and of their educational foundations, should serve as a warning against totalizing idealisms. Neglecting “national,” local, and “comparative” perspectives in research ignores or distorts the actual situations of many places worldwide, as does sole focus on, or selective application of, “postcolonial.” After I presented a version of this chapter at an international conference on writing, colleagues from the Middle East and Africa came up to say that what I said about literacy education in Poland resonated with their own situations. The tasks and shape of democratic literacy education are not necessarily self-evident; they remain to be worked out in each particular case in view of the local experiences, challenges, and histories (including ideological and political/regime histories). Comparative, boundary-crossing research into literacy education in a variety of historical, ideological, and political/ regime contexts may reveal more about the potential meanings, uses, and consequences of educational practices and values, and provide “distant mirrors” for continuing reflection on the nature of democratic literacy education in a globalizing world.
Notes 1 One must bear in mind that actual classroom practices often varied from official precepts, although the relative range of teacherly freedom was limited. During my own education in Poland in the 1960s and early 1970s, I had many outstanding teachers who taught me in spite of, and often against, the confines of textbooks and official programs. 2 The present author was one of the many Western “experts” who conducted workshops organized by the Polish Ministry of Education’s Center for Teacher Training as part of the 1999 reform.
References Aristotle (1943). Politics. Trans. Benjamin Jowett. New York: The Modern Library. Axer, Jerzy, with the assistance of Katarzyna Tomaszuk (2007). Central-Eastern Europe. In Craig W. Kallendorf (Ed.), A companion to the classical tradition (pp. 132–155). London: Blackwell. Bahri, Deepika (2004). Terms of engagement: Postcolonialism, transnationalism, and composition studies. In Andrea A. Lunsford & Lahoucine Ouzgane (Eds.), Composition and postcolonial studies (pp. 67–83). Pittsburgh, PA: University of Pittsburgh Press. Bazerman, Charles, & Russell, David (Eds.). (2003). Writing selves/writing societies: Research from activity perspectives. Fort Collins, CO: WAC Clearinghouse, http://wac.colostate.edu/books/sselves_societies.
Writing, from Stalinism to democracy 95 Berlin, James A. (1990). Writing instruction in school and college English, 1890– 1985. In James J. Murphy (Ed.), A short history of writing instruction: From Ancient Greece to twentieth-century America (pp. 193–222). Davis, CA: Hermagoras Press. Chodzko, Ryszard (1992). Wyobraznia Wyzwolona: Rzeczywistosc Osoby i Nierzeczywistosc Doktryny. Bialystok: Nowa Era. Dembowska, Janina, Saloni, Zygmunt, & Wierzbicki, Piotr. (1970). Swiat i My. Podrecznik do Nauki Jezyka Polskiego dla Klasy VI. Warsaw: Panstwowe Zaklady Wydawnictw Szkolnych. Farrell, Thomas B. (1993). Norms of rhetorical culture. New Haven, CT: Yale University Press. Hesford, Wendy S., & Schell, Eileen E. (2008). Configurations of transnationality: Locating feminist rhetorics. College English, 70, 461–470. Hillocks, George, Jr. (1986). Research on written composition: New directions for teaching. New York: National Conference on Research in English; Urbana, IL: ERIC Clearinghouse on Reading and Communication Skills, National Institute of Education. Hillocks, George, Jr. (1987). Synthesis of research on teaching writing. Educational Leadership, 44, 71–82. Hillocks, George, Jr. (1995). Teaching writing as reflective practice. New York: Teachers College Press. hooks, bell (1995). Third World diva girls: Politics of feminist solidarity. In Lorraine Edwards & Midge Stocker (Eds.), The woman-centered economy: Ideals, reality, and the space in between (pp. 265–280). Chicago, IL: Third Side. Instytut Programow Szkolnych Ministerstwa Oswiaty i Wychowania. (1985). Programy Szkoly Podstawowej, Czesc I. Warsaw: Wydawnictwa Szkolne i Pedagogiczne. Janion, Maria (2007). Niesamowita Slowianszczyzna: Fantazmaty Literatury. Krakow: Wydawnictwo Literackie. Kijas, Juliusz (1968). Cwiczenia w Mowieniu i Pisaniu. In Wladyslaw Szyszkowski & Zdzislaw Libera (Eds.), Metodyka Nauczania Jezyka Polskiego w Szkole Sredniej (pp. 183–203). Warszawa: Panstwowe Zaklady Wydawnictw Szkolnych. Klemensiewicz, Z., & Zlabowa, J. (1949). Nasz Jezyk: Podrecznik Gramatyki Jezyka Polskiego z Cwiczeniami dla Klas VI I VII Szkoly Podstawowej. Warsaw: Panstwowe Zaklady Wydawnictw Szkolnych. Kloczowski, Jerzy (1998). Mlodsza Europa. Warsaw: Panstwowy Instytut Wydawniczy. Ksiazek, Wojciech (2001). Rzecz o Reformie Edukacji, 1997–2001. Warsaw: Adam Marszalek. Kulpa, Jan (1954). Jak Znajomosc Socjalistycznej Teorii Pedagogicznej Wplynela na Organizowane Przeze Mnie Powtarzanie Materialu Jezyka Polskiego. Nowa Szkola, 3. Kulpa, Jan (1955). Z Doswiadczen nad Organizacja Cwiczen Stylistycznych (Metoda Pisania Wypracowan). Centralny Osrodek Doskonalenia Kadr Oswiatowych, Biblioteka Metodyczna Nr. 16. Warsaw: Panstwowe Zaklady Wydawnictw Szkolnych. Lunsford, Andrea A., & Ouzgane, Lahoucine (2004). Composition and postcolonial studies: An introduction. In Andrea A. Lunsford & Lahoucine Ouzgane (Eds.), Composition and postcolonial studies (pp. 1–8). Pittsburgh, PA: University of Pittsburgh Press.
96 C. M. Ornatowski Ministerstwo Edukacji Narodowej (1990). Program Szkoly Podstawowej: Jezyk Polski, Klasy IV–VIII. Warsaw: Wydawnictwa Szkolne i Pedagogiczne. Ministerstwo Oswiaty (1963). Program Nauczania Osmioklasowej Szkoly Podstawowej (Tymczasowy). Warsaw: Panstwowe Zaklady Wydawnictw Szkolnych. Murphy, James, (Ed.) (1990). A short history of writing instruction: From Ancient Greece to twentieth-century America. Davis, CA: Hermagoras Press. Phelps, Louise Whetherbee (2008). Interpreting Transformational teaching practices in Armenian classes: Methodological considerations in a cross-cultural observation. Paper presented at the Writing Research across Borders conference, University of California, Santa Barbara, February 22, 2008. Pojawska, Karolina (1969). Cwiczenia w Mowieniu i Pisaniu w Klasach V–VIII. Instytut Pedagogiki, Biblioteka Metodyczna Nauczyciela Jezyka Polskiego. Warsaw: Panstwowe Zaklady Wydawnictw Szkolnych. Poznanski, Jacek (2003). Wzory Wypracowan z Jezyka Polskigo dla Wszystkich Klas Gimnazjum. Warsaw: Skrypt. Radziwill, Anna (1981). Ideologia Wychowawcza w Polsce w Latach 1948–1956 [Proba Modelu]. Warsaw: Zeszyty Towarzystwa Kursow Naukowych. Rytlowa, J., Piorunowa, A., & Lewandowski, Z. (1949). Wspolnymi Silami. Warsaw: Panstwowe Zaklady Wydawnictw Szkolnych. Shannon, Patrick (2001). Turn, turn, turn: Language education, politics, and freedom at the turn of three centuries. In Patrick Shannon (Ed.), Becoming political, too: New readings and writing on the politics of literacy education (pp. 10–30). Portsmouth, NH: Heinemann. Schaub, Mark (2003). Beyond these shores: An argument for internationalizing composition. Pedagogy, 3, 85–98. Stalin, Joseph. (1950). W Sprawie Marksizmu w Jezykoznawstwie. Warsaw: Ksiazka i Wiedza. Tabisz, Anna (2006). Kompetencja Tekstotworcza Uczniow na Przykladzie Rozprawki. Opole: Uniwersytet Opolski. Wawrzykowska-Wierciochowa, D. (1954). Materialy do Metodyki Nauczania Jezyka Polskiego w kl. V–VIII: Zajecia Pozalekcyjne Zwiazane z Nauczaniem Jezyka Polskiego. Warsaw: Panstwowe Zaklady Wydawnictw Szkolnych. Wieczorkiewicz, Bronislaw, Jaworski, Michal, & Rurawski, Jozef (1970). Wczoraj i Dzis: Podrecznik do Nauki Jezyka Polskiego dla Klasy VII Szkol dla Pracujacych. Warsaw: Panstwowe Zaklady Wydawnictw Szkolnych. Wojdon, Joanna (2001). Propaganda Polityczna w Podrecznikach dla Szkol Podstawowych Polski Ludowej (1944–1989). Warsaw: Adam Marszalek.
8 A pilot investigation A longitudinal study of student writing in a post-totalitarian state Gil Harootunian McDaniel College, Westminster, MD
Overview A three-year educational exchange project between Syracuse University, New York (SU) and Yerevan State Linguistic University (YSLU), Armenia, funded by the US Bureau of Educational and Cultural Affairs (ECA), afforded a rare long-term, on-site opportunity to study the reception of Western-style methods into a post-totalitarian culture. To my knowledge, I administered the first longitudinal study of student writing in a post-Sovietnation university. A total of 194 students in this project chose to sign Letters of Informed Consent, generating 454 samples during the two-year assessment.1 The two main findings are (1) that it takes three semesters of sequenced and sustained instruction for students to understand and adapt the Western-style methods to their own situation, and (2) a startlingly consistent trajectory, with clearly identifiable stages of mirroring, testing, and mediated mastery, became evident over the course of those three semesters. Mirroring is used to describe the student’s entrance into the classroom community, denoting both the social and physical aspects of learning: the imitative learning in zones of proximal development (Vygotsky, 1978) and the “mirror system” that fires up in the human brain during social stimulation (Knoblich, 2006; Jaffe, 2007, p. 22). Testing denotes that stage when students begin to try out our methods and techniques (Phelps, 1988, pp. 218–241). Mediated mastery denotes that the Armenian students now understand and can modify the methods to serve their own ends (Vygotsky, 1978; Rogoff, 1995; Wertsch, 1991). I would clarify that Armenia is best termed a post-totalitarian state (not a postcolonial one). YSLU was founded during the Soviet era (in 1935) as a two-year teacher training institute whose main mission was to instruct Armenian teachers to teach in Russian throughout Armenia. During Armenia’s first decade of independence (1991–2001), English and other Western languages were a source of opposition to the Soviet Union and its imperative Russian. By the time I arrived as a Fulbright Scholar (2001), English had become one of ten languages being taught at Yerevan State Linguistic University.2 Therefore, after I returned from the Fulbright experience, I drafted a project not for the teaching of English at YSLU but for helping to
98 G. Harootunian reform the totalitarian pedagogy lingering there. The ECA awarded $275,000 for a three-year project (fall 2003–spring 2006) to pilot in select YSLU classrooms Western-style democratic and writing-intensive methods, for example, written course guides and student evaluations and videotaped teaching observations for the purposes of constructive critique. A handy lens is needed to give international educators context for this study conducted in a former eastern republic of the Soviet Union. In The Rhetoric of Reason (1996), James Crosswhite describes the transformative potential of Western-style writing, contending that argument essentially means a “giving up of the ideas that one has a metaphysically fixed identity, or that one’s ideas do, in favor of a recognition of the self- transformation and learning that occur in reasoning” (p. 24). Crosswhite considers argument to be a process of inquiry that can “cultivate an intelligent citizenry” and enable “productive, nonviolent” resolution of social conflicts (pp. 17, 256, 294). The progress of Armenian students toward Crosswhitean “self-transformation” will be seen as explicit because they were starting from a radical point: the epideictic rhetoric of the Soviet Union. Ong described the ways in which the totalitarian rhetoric of the Soviet Union was magical: if one proclaims that the people love Stalin, they will. It was crammed with epithetic formulas: “the Glorious Revolution of October 26” (1982, pp. 38–39). Crosswhite explains that the purpose of such epideictic rhetoric is “keeping one’s intent hidden from one’s enemy” so those in power can strengthen preferred values without making explicit claims about those values so the audience could then engage in open argument about the values (pp. 108–109). The Armenian students were being asked no less than to move from the epideictic rhetoric of the Soviet Union to the productive, open conflict of Crosswhitean argument (p. 281). The material conditions under which this project was conducted will help to clarify its administration and the results. Research in the post- Soviet bloc, especially former republics vs. former satellite countries, remains introductory for many reasons, including poverty, corruption, and ethnic conflicts. These conditions have virtually pre-empted longitudinal studies by US scholars. Harrington (2005) used technology to begin the long-term distance-learning “Global Classroom Project.” The small but growing body of short-term on-site research includes Rodman (1996), Stevens (2000), and Hagen (1998), who visited the post-Soviet republics on Fulbright programs. They detailed the ways in which the Soviet oral culture facilitated political, economic, and educational corruption. In the former satellite countries, either located on the EU boundary or now members of the European Union, increased scholarship is becoming available to English-speaking educators. Romania’s Minister of National Education, Andrei Marga, described ongoing comprehensive efforts to overhaul Romanian higher education to eliminate its “communist heritage,” from centralization to endemic corruption (2002, p. 123). The goal for Romania was similar to that of Armenia: “to stimulate the development of a civic culture, a culture of opinions, the need for which has been
A longitudinal study of student writing 99 intensely felt” (p. 125). Other recent studies confront and analyze the lingering oral (vs. writing) culture in higher education in Eastern Europe: Gilder (1995), Verdery and Kligman (1992), and Marin (2005) in Romania; McKinley (1995) in post-communist Hungary; Ornatowski (1995, 2008) in Poland. Clearly the former Soviet bloc offers an educational zone different from that of the West. Given the tough conditions in Armenia, this study was a truly pilot investigation conducted to help fill the gaps in knowledge about rhetoric and rhetorical education in post- totalitarian states.
The pilot investigation Of the students who participated in the project long term, 58 percent showed a consistent trajectory leading to mastery of Western-style, democratic methods: the students progressed from mirroring Western-style pedagogy in the first semester to testing the methods in the second and finally to mediated mastery in the third semester. This trajectory was clearly marked by the Armenian students actively mediating the Western-style writing methods:3 they were filtering the instruction to acquire those concepts and tools that enabled them to further understanding of their new nation Armenia. Finally, these 58 percent began as the weakest or generally undistinguished writers; they had either the least history (or most resistance) to the Soviet-style instruction.4 The student outcomes assessment was conducted during the last three semesters of the project (Spring 2005, Fall 2005, Spring 2006), after the arduous work to set up and stabilize the pilot course offerings. Students were asked to participate in the longitudinal study by signing a Letter of Informed Consent, then drafting two writing samples per semester, one at the start and one at the end. The Letter of Informed Consent guaranteed students that only the US Project Director, plus the individual teachers in their classrooms, would have access to their writing, and that only pseudonyms would be used in any publication. During 2004–2006, a total of 17 students generated 70 samples over all three semesters.5 The samples were asked for and collected at the beginning and end of each course. Some students who had refused to participate at the beginning of a semester later gave one final sample at the end after they had come to understand and accept the study. The number of samples per student per semester breaks down in the manner shown in Table 8.1. This exchange grant brought all four Armenian teachers in the project’s pilot classrooms to SU for the Spring 2004 semester. They observed classes and talked with colleagues, and we worked to generate course materials appropriate for YSLU classrooms. The course and assignment topics and lessons discussed below are in the main SU approaches and materials adapted for Armenian classrooms. When the teachers began piloting the project’s courses at YSLU, they had SU graduate students to co-teach with them.6 These Armenian teachers and the US co-teachers delivered the
100 G. Harootunian Table 8.1 Distribution of student samples per semester Pseudonym [alphabetical]
Spring 2005 Fall 2005 Spring 2006 Total
Linara Balakian Ara Bedrosian Victoria Davtyan Kristine Galstyan Eva Gasparyan Diana Gatyan Rose Goshgarian Silva Haroyan Lilit Karchyan Alina Kosyan Karina Martirosyan Gregory Mikoyan Marianna Parzyan Anna Sahagian Laura Simonyan Meri Tashjian Anna Vardanyan (enrolled in two courses in fall 2005) Total
2 1 1 2 1 2 1 1 1 1 1 1 1 1 2 1 2
2 1 1 1 2 2 1 2 2 2 2 1 2 1 1 2 4
1 1 1 1 1 2 1 1 1 2 1 1 1 1 1 1 1
5 3 3 4 4 6 3 4 4 5 4 3 4 3 4 4 7
22
29
19
70
Table 8.2 Courses referenced in narrative, with teachers and semester identified Semester
Student (alphabetical)
Teacher (alphabetical)
Writing-intensive course
Spring 2005
Gasparyan, Eva Parzyan, Marianna Mikoyan, Gregory Vardanyan, Anna Bedrosian, Ara
Khoyetsyan, Arthur Sargsyan, Nelly
Argument
Sarsgyan, Nelly
Public Relations
Political Science
Fall 2005
Simonyan, Laura
Kananyan, Sophia International Journalism: Newswriting Mikoyan, Gregory Sargsyan, Nelly International Vardanyan, Anna Relations Gasparyan, Eva Parzyan, Marianna Soghikyan, Intercultural Vardanyan, Anna Christine Communication
Spring 2006
Simonyan, Laura Kananyan, Sophia Newswriting Mikoyan, Gregory Sargsyan, Nelly International Relations II Vardanyan, Anna Soghikyan, Intercultural Christine Communication II
A longitudinal study of student writing 101 courses and collected the writing samples. I observed on-site the implementation of these materials in YSLU classes, and depending on scheduling of SU and YSLU semesters, I would either be present during the collection of samples or would collect the samples directly from the teachers. I observed all pilot classes to meet the students and clarify any questions, in person, about the exchange project.
Stage one: mirroring and the Armenian community (Spring 2005) N.B.: Samples of student work are presented and summarized, then analyzed. Sample student work In her first sample this semester on “Nationalism: How Many Faces Does It Have?” Anna Vardanyan commenced: “In the crossroads of the world there is some great diversities.”7 In her second sample, Anna opened the essay with a quote from a famous Armenian writer: “No one could win the strongest king but the woman and wine.” She used this as a springboard for discussion of women’s rights in Armenia, and at one point anchored her argument in real-life examples (e.g., female deputies in the Ministry of Culture and National Assembly). Gregory Mikoyan opened his essay on “Nationalism: How Many Faces Does It Have?” in a manner that would remind US readers of a worn ethnic joke: “Nationalism. What is it? An American, an indian, a jew, an armenian. . . .” The essay consisted of one large handwritten paragraph, in which he used two examples: the primary one of the World War I extermination of the Armenians and the secondary one of the World War II Fascist Germany (and implied extermination of the Jews). He qualified that nationalism can have a “sustaining importance,” meaning that if it weren’t for the Armenian diaspora, he wouldn’t be writing this essay in an exchange project (a reference to the Armenian-Americans who support these projects). On the topic, “What Are the Top Three Things You Consider You Need To Do in Your Life,” Eva Gasparyan stated that she found it “very difficult to answer directly” this question. When she arrived at the answer, Eva stated that she wanted to have a good profession; to have a stable family (which helps the Armenian nation to be stable); and to live according to the Bible. In his first essay on “What Is the Professionalism of a Writer Measured By,” Ara Bedrosian declared that a public relations professional should “read lots of books.” Throughout, Ara gave no examples and one figure of speech: “the professional should know his audience and his demands like his own five fingers.”
102 G. Harootunian Analysis The students’ writing typifies the mirroring stage: the students first attempt to give back to the teacher the expressed ideas and instructions from the teacher’s class materials and lectures: one can almost hear (as I often did in the classrooms) the YSLU teachers lecturing the students on the “great diversities” in the world or that public relations experts should “read lots of books.” Moreover, caveats like Eva’s that it is “very difficult” to answer the topic were repeated verbatim in the papers of many students in the same class. The writing might seem plainly empty of substance in this mirroring stage, and even seem like parroting, particularly as the students possess limited ESL skills. Yet, mirroring was critical to these students because it provided them with the space and time needed to absorb change and acquire understanding. For example, the subject of Ara’s course (public relations) was non-existent in the Soviet Union and still weakly understood in Armenia. It is understandable that Ara would begin in the main by mirroring his teacher’s declarations about public relations. I use “mirror” to denote kindred social and physical processes. “Mirroring” begins the transformation process because the students’ imitation is an act of imagination that allows them to enter and grow into the intellectual life around them (Vygotsky, 1978). The term simultaneously refers to the mirror system in the human brain that fires up during social action, exposing the empathetic nature of the human brain (Knoblich, 2006; Jaffe, 2007). In this case, the social action occurred in Armenian classrooms: in what ways might the mirroring be distinctly Armenian? The project’s external evaluator, Cezar Ornatowski (San Diego State University), singled out for comment the “vast resources of good will” evident in the classrooms. Armenian students had formed a genuine, positively charged network with each other and the teachers, one that facilitated the learning process despite the harsh conditions. Anna’s first sample might read like a podium speech, and her second like a public oration, but that is because her “we” referred not to an unknown reader but to her surrounding, homogeneous Armenian population (e.g., “Besides, we have very active women working abroad”). Ara’s image of holding up one’s hand also recalls the traditional Armenian oral culture, reinforced by the Soviet epideictic oral culture, more than the new writing culture of the pilot classrooms (especially as I saw such hand and finger gestures accompany daily speech in Armenia). But importantly, this is an initial sign of mediation: gestures (visual signs) forerun writing (textual signs) (Vygotsky, 1978; pp. 107–108), and so Ara’s gesture assigned Armenian meaning to his writing, for his immediate, homogeneous audience. Gregory’s audience is also the Armenian community that would understand his references (to the Armenian diaspora, 1915 genocide, etc.). Eva’s writing showed a rare “I” that is critical of the Armenian “we,” yet her “I” is in opposition to that Armenian “we” and so boxed within that context.
A longitudinal study of student writing 103 This “we” that operates in ancient and homogeneous cultures such as Armenia’s is like a double mirror, an all-encompassing one that facilitated both the strong, mediated network in the Armenian classroom but also ironically the student’s inward perspective. The issue was salient for me as the US Project Director, for this perspective is antipodal to a Western-style writer’s low context, outward perspective (Hall, 1976).
Stage two: testing social and linguistic bridges (Fall 2005) Sample student work Anna Vardanyan wrote on Armenia and international relations: “But still no powerful country can ‘afford’ itself to develop and resolve both inner and external conflict without the ties with outside world.” In an intercultural communication class taken simultaneously, Anna wrote: “I guess everyone has at least once asked the same question: ‘What is the meaning of his being?’ ” She concluded with a comment on nationalism: only Armenians are geniuses, etc. These are the words that I’ve heard hundred of times. And if you dare to mention, “What about Chopin, or Mozart, etc.,” one would have an answer, “O.k. O.k.; this is the only name in the whole nation.” Laura Simonyan, in a news-writing course, wrote: First I want to say what I understand by saying censorship. To my mind, everybody must act with the framework of the human code, ethical code, to be unbiased whether he belongs to the government staff or he is simply a journalist. Laura opened her essay by quoting the First Amendment, then proceeded to illustrate the extent of its realization in both nations, but particularly Armenia, noting that an “expensive dinner at a restaurant, a certain amount of money, a promise to help his child to enter the University is enough to overcome any resistance” to corruption in an Armenian journalist. She then argued a series of remedies that should generate “better results.” Eva Gasparyan in intercultural communication wrote: It is of common knowledge that society is very powerful. I am very anxious about this problem because I see what is going on with my own eyes. . . . I knew a boy who played the violin. Every time he took his violin and went to school, the other boys who were hanging around in the streets, started shouting, whistling, and making fun of him. He couldn’t bear this suffocating situation more than two years. He gave it up and became mentally ill. Now he is in the lunatic asylum.
104 G. Harootunian In her second sample, on “Are Armenians Ethnocentric?,” Eva stated there are “different types” of ethnocentric Armenians, ranging from extreme to moderate, then included herself among the latter. She developed her theme of “within-group” differences in Armenia by discussing how boys are more ethnocentric than girls in Armenia, for the girls change their dress, music, and a lot more: “Most of the girls are more flexible than the boys. Girls undergo different global changes more easily and quickly than the boys.” Marianna Parzyan, on the topic “My Expectations from this Course,” listed basic personal goals, for example, to graduate from YSLU. She then discussed the discouraging challenges she faced when she entered YSLU, before describing her successful adjustment. When Marianna elaborated on the challenges, she presented them in Homer-esque terms: “I thought I was in a tossing sea and that I was going to die there.” In her second sample on “Are Armenians Ethnocentric?” Marianna began with a retrospective analysis: If one wants to answer this question, he must know what is ethnocentrism. If anyone asked me the same question last year, I was not able to give a proper answer to it, as I really had no idea of ethnocentrism. She explains that it took her “much time” to generate an answer, then proceeded to discuss the ethnocentric reactions of her fellow students to the Arab, Indian, and Persian students in Armenian universities, concluding with a discussion of her friendship with an Indian student. Gregory Mikoyan moved into a regional discussion of the issue of nationalism and conflicts in his International Relations course. Gregory first noted the formation of modern nation states, then the international system of political blocs in the second half of the twentieth century. He concluded with an analysis of a specific conflict and its resolution between Armenia and neighboring Georgia in the 1990s. Analysis The student samples showed slight to moderate increases in English fluency, but a more significant journey from mirroring to testing. The students began making the critical shift from interpersonal (social) learning in the classroom to intrapersonal (individual) transformation (Vygotsky, 1978), and this was accompanied by an increasingly outward perspective. Marianna’s writing contained the phrases that first suggested to me (on the linguistic level) that testing operated as a bridging technique. Her “tossing seas” might be Homer-esque clichés in Western writing, but they are also the first “bridges” for the Armenian students influenced by the Soviet Union’s epideictic rhetoric (e.g., “the Glorious Revolution of October 26”), and Armenian culture’s long history of oral drama (opera, poetry, song, theater, toasting). Such phrasing suggested the students’ active medi-
A longitudinal study of student writing 105 ation of Western-style education: they first assimilated with more ease those techniques closest to Armenian styles and techniques. Closer analysis of writing styles and techniques made apparent that testing operated as bridge-building in the social aspect, too. Marianna’s writing simultaneously showed an increasing awareness of international relations as she moved from an old, interpersonal “we” (ethnocentric Armenian peers) to a new, intrapersonal “we” (she, plus others who wish to move beyond the larger peer group’s ethnocentrism). Parallel movement surfaced in the writing of the other students. Anna began to test English punctuation marks in her first sample: she deployed continually quotation marks or underlining to call attention to words or points whose meaning she was exploring, and she tried out conventions like rhetorical questions, quotations, and more advanced English punctuation such as the semi-colon. As she subdued and mastered English punctuation in her second sample, Anna also opened the essay with an announcement of her “research” into the subject to be investigated, then launched into a direct line of argument on the subject. This testing of punctuation and argument suggested that Anna was now trying to author a Western-style written conversation, and that means acquiring an expanded mental model beyond the Armenian one. Likewise, Laura made her first attempt to map her individual understanding for readers, one freed from the homogenized Armenian “we.” Laura was genuinely taking up the prompt, for herself, trying out her own definition of the assigned topic. She launched the essay on “Roles and Responsibilities of the Journalist in Armenia” with a detailed and balanced assessment of the state of journalism in Armenia vs. the United States (“we” vs. “they”). This suggested that Laura used Western-style methods, once grasped, for her own ends: her new and strengthened identity as a writer is used to interrogate the Armenian “we” and make recommendations for change. Laura has begun building bridges for the future. Eva’s writing showed parallel development. In the first essay of the semester on “The Influence of Society on an Individual,” she turned to a theme that will resurface (the fate of individuals who “differ” from the dominant types in her society), offering a more enthusiastically detailed response that shows new strength in her “I” that is oppositional to the dominant Armenian “we.” Finally, Gregory moved from last semester’s “mirror” perspective on international relations (e.g., the Nazi Germany extermination of the Jews in World War II mirrored the Ottoman Turks’ extermination of the Armenians in World War I) to this semester’s new perspective of “interdependence,” for example, the interdependence of Georgia on Armenia for energy allowed successful resolution of an incipient conflict in the 1990s. Westernstyle writing education, then, is one prompt to help Gregory build a new bridge: he shifts from an inward-looking Armenian identity to one that interacts with and sees—or tests out—possible inclusion in the global village (Phinney, 2002, pp. 129, 202).
106 G. Harootunian
Stage three: mediated mastery (Spring 2006) Sample student work In Intercultural Communication II, Anna Vardanyan wrote. “There are different factors that have a great role in intercultural communication (culture, language, etc.) and factors that hinder intercultural communication, such as prejudice, ethnocentrism, discrimination, stereotyping, etc.” She explored these factors (e.g., “It is considered that 93% of communication is non-verbal and only 7% is verbal”), then asked readers, “so what should we do? One may think that it would be useful to learn all kinds of gestures and its corresponding meanings.” She concludes, “The only thing one can do is to be ready to learn . . . other cultures, languages, modals of life[,] preserving his/her own.” On “What Is the Meaning of Free Speech and Responsible Journalism in a Democratic Society?” Laure Simonyan began with multiple and integrated textual references and devices, ranging from quoting the First Amendment to posing a rhetorical question on free speech in democracies. She concluded by deploying a US idiom for Armenian society: “Corruption is the axis of all evils.” In International Relations, Gregory Mikoyan began by defining the topic neither through the mirror of “we” the Armenians, nor through regional interdependence, but through a defining international moment: 9/11. “I think that leader or representatives of religious world should be involved in the antiterrorism campaign and the attention must be shifted from military sphere to ideological one.” He clarified, “Most often common people support terrorists who are convinced they are doing it for the sake of their nation.” Analysis The students had achieved mediated mastery, meaning that they had experienced a culture-bound, clear secondary socialization into the Western-style methods and approaches. Anna had internalized the learning to the extent that she now had the ability to negotiate with Western audiences, theses, and forms, for her own purposes. She opened her sample by setting common ground for inquiry not with her homogeneous Armenian peers but with a community of unseen readers. She no longer presented the topic as a word, around which she placed over-emphatic quotation marks, but instead defined that concept and its factors for readers: As she proceeded, Anna no longer announced that she had done research, but presented it. In short, Anna contextualized her discussion for Western-style readers with definitions of concepts, increased use of subtle and varied punctuation marks, and so on. Critically, however, she cycled back to Armenia: one preserves one’s own culture. Or, as Gregory shifted from inward to outward looking, he actively signaled that he was interrogating concepts by placing quotation marks
A longitudinal study of student writing 107 around its words (albeit awkwardly), then referenced the attempts of other nations to deal with terrorism, before offering his own recommendation. His final words cycle back to Armenia as he returns to the concept of “the nation.” In other ways is mediated mastery marked as a culture-bound secondary socialization. The writing conventions that surface and resurface in Anna’s essays—quotes, proverbs, rhetorical questions, a lively and engaging “written” conversation—are all markers of the Armenian oral culture (and its rich tradition of poetry, speech, toasting, hymns, storytelling, and so on) (Hagen, 2005; Harootunian, 2005; Rodman, 1996; Stevens, 2001). Coupled with the absence of other writing conventions (e.g., technical analogies that are not markers of the Armenian oral culture), this reinforces the suggestion that Armenian students first mediated Western-style materials and methods through the “bridge” techniques and styles. Consider that Laura wrote that governments that operate by means of terror should be called “terror-ian” (the Armenian linguistic equivalent of “terror-ist”). Her linguistic play—changing English words to Armenian “forms” or making the challenge of post-Soviet corruption the axis in an Armenian classroom (Harootunian, 2007; MacWilliams, 2001; Ornatowski, 1995; Rodman, 1996)—continued to indicate that students are mediating Western methods to give themselves a wider window of understanding on Armenia: they adapted methods to modify and develop their identity as Armenian writers.
Conclusion Given the enormity of the task for the Armenian students’ grasping a new cultural framework, such “bridging” elements would be critical to assimilating Western-style methods. Once the first mediations have occurred, students like Anna immediately began to apply the new elements to explore with increased depth Armenian questions. The Armenian students made the learning in the project’s classrooms transformative precisely because they created an intellectual loop between the immediate pilot classrooms and the external Armenian culture (Vygotsky, 1978). And so my final recommendation would be that we direct our efforts toward sustained, long-term projects, not sporadic and short-term ones, and the focus of such exchanges should be human resources. How can we create classrooms that are zones of proximal development so that foreign participants can mediate our offerings, in ways targeted to produce the most valued change in their own cultures? That Western-style methods helped Armenian students acquire an expanded mental model they subsequently used to make recommendations for the benefit of Armenian society clarifies what we sometimes take for granted about our methods: they are essentially democratic and therefore laden with counter-cultural power. The power is realized not by our passing these methods and tools onto foreign participants; it is realized when foreign participants take up and master these methods for their own purposes.
108 G. Harootunian
Notes 1 Every one of the 454 samples was read and cataloged. 2 YSLU offers English, Russian, German, Spanish, French, Italian, Greek, Persian, Korean, and Chinese. 3 Research into the neurobiology of cognition could offer further insights into the process of mediated mastery, particularly how culture and learning sculpt and resculpt brain structure over time (Park, 2007; Weinberger, 2004; Schlaug, 2003; Sloboda, 2003; Shellenberg, 2003; Gaser & Schlaug, 2003; Pascual-Leone, 2003; Juslin and Sloboda; 2001; Tramo, 2001; Jourdain, 1997; Besson, Faita, & Requin, 1994; Sergent, Zuck, Terriah et al., 1992; Bever & Chiarello, 1974). 4 Most of the other 42 percent had excellent ESL skills and had at some point been educated in literary-style essays, by Armenian teachers who often reminded me with pride that the Armenian translation of Shakespeare was considered the best in the Soviet Union. This literary admiration often trumped a democratic approach to writing instruction when I was at YSLU. Mastery of a literary style and ESL competence was, simply, more valued. For example, one student stated that she already knew how “to write” essays because she had studied British and US literature during her schooling. 5 The main challenge was this limited number of samples due to student resistance. Many students completed the essay for the teachers but declined to sign the Letter of Informed Consent, and the writing competence of such students ranged from poor to excellent, indicating that writing skills per se were not the reason that students declined to sign. Comments made by the students suggested to me that the causes of student resistance to signing were their lack of experience with and belief in democratic authority and their unfamiliarity with advanced writing methods and pedagogy, particularly Western-style. 6 The US graduate students/co-teachers were Maxwell Garcia and Micah Rubin (Journalism) and Susan Runkle (Anthropology) in 2004–2005 and Michael Apicelli and Thomas Bezigian (International Relations/International Law) in 2005–2006. 7 Student writing is reproduced exactly.
References Casimir, Fred L. (Ed.). (1995). Communication in Eastern Europe: The role of history, culture, and media in contemporary conflict. Mawhah, NJ: Lawrence Erlbaum. Crosswhite, James (1996). The rhetoric of reason: Writing and the attractions of argument. Madison, WI: University of Wisconsin Press. Gilder, Eric (1995). Turning personal experience into social reality: Communication as a “third-culture building” tool in the Romanian classroom. In Fred L. Casimir (Ed.), Communication in Eastern Europe: The role of history, culture, and media in contemporary conflict (pp. 197–221). Mawhah, NJ: Lawrence Erlbaum. Hagen, P. (1998). Teaching American business writing in Russia: Cross-cultures/ cross-purposes. Journal of Business and Technical Communication, 12, 109–126. Hall, Edward T. (1976). Beyond culture. New York: Anchor Books. Harootunian, Gil (2005). Re-designing our technical and individual screens: The new “windows” opened by teaching in a former Soviet republic. In C. Lipson & M. Day (Eds.), Technical communication and the world wide web (pp. 193–208). Mawhah, NJ: Lawrence Erlbaum.
A longitudinal study of student writing 109 Harootunian, Gil (2007, January). Dancing the Kochari: Challenging the U.S. perspective on communication in newly-democratic cultures. Journal of Business and Technical Communication, 21, 91–105. Harrington, Ty (2005). Linking Russia and the United States in web forums: The global classroom project. In C. Lipson & M. Day (Eds.), Technical communication and the world wide web (pp. 167–192). Mawhah, NJ: Lawrence Erlbaum. Jaffe, Eric (2007, May). Mirror neurons: How we reflect on behavior. Observer, 20(5), 20–25. Knoblich, G., & Sebanz, N. (2006). The social nature of perception and action. Current Directions in Psychological Science, 15, 99–104. Lipson, Carol, & Day, Michael (Eds.) (2005). Technical communication and the world wide web. Mawhah, NJ: Lawrence Erlbaum. MacWilliams, B. (2001, December 14). Corruption, conflict and budget cuts afflict academe in former Soviet republics. The Chronicle of Higher Education, p. A43+. Marga, Andrei (2002). Reform of education in Romania in the 1990s: A retrospective. Higher Education in Europe, 27(1–2), 123–135. Marin, Noemi (2005). Communication adaptation and competence in global times: Intercultural challenges for Eastern European foreign graduate students in American academia. Journalism & Communicare, 4(13), 52–61. McKinley, Mary M. (1995). Hungarian culture in communication. In Fred L. Casimir (Ed.), Communication in Eastern Europe: The role of history, culture, and media in contemporary conflict (pp. 115–131). Mawhah, NJ: Lawrence Erlbaum. Ong, Walter (1982). Orality and literacy: Technologizing the word. New York: Methuen Press. Ornatowski, Cezar (1995). Democracy and discourse: A rhetorical perspective on political change and the emergence of democracy in Poland. In J. Kutnik & C. Ornatowski (Eds.), Re-visioning democracy, Central Europe, and America: Critical perspectives (pp. 57–65). Lublin, Poland: Maria Curie-Sklodowska University Press. Ornatowski, Cezar. (2008, February). Literacy education and the state: Native language pedagogy and politics in Poland, 1945–1999. Paper presented at Writing Research across Borders, University of California, Santa Barbara. Park, Denise C. (2007, May 25). Eastern brain/Western brain: Neuroimaging cultural differences in cognition. Paper presented at the 19th Annual Convention of the Association for Psychological Science, Washington, DC. Phelps, Louise Wetherbee. (1988). Composition as a human science. Oxford: Oxford University Press. Phinney, J. S., & Devich-Navarro, M. (2002). Variations in ethnic identity. In Jeffrey Jensen Arnett (Ed.), Adolescence and emerging adulthood (pp. 120–131). Upper Saddle River, NJ: Prentice Hall. Rodman, L. (1996). Finding new communication paradigms for a new nation: Latvia. In D. C. Andrews (Ed.), International dimensions of technical communication (pp. 111–121). Arlington, VA: Society for Technical Communication. Rogoff, Barbara, & Chavajay, Pablo (1995, October). What’s become of research on the cultural basis of cognitive development? American Psychologist, 50, 859–877. Rohde, Carl C., & Pellicaan, Carsten R. C. (1995). Advertising and the legitimacy crisis of Eastern Europe. In Fred L. Casimir (Ed.), Communication in Eastern
110 G. Harootunian Europe: The role of history, culture, and media in contemporary conflict (pp. 133–162). Mawhah, NJ: Lawrence Erlbaum. Stevens, B. (2000). Russian teaching contracts: An examination of cultural influence and genre. Journal of Business and Technical Communication, 14, 38–57. Verdery, Katherine, & Kligman, Gail. (1992). Romania after Ceausescu: Post- communist communism? In Ivo Banac (Ed.), Eastern Europe in revolution (pp. 117–147). Ithaca, NY: Cornell University Press. Vygotsky, Lev. (1978). Mind in society: The development of higher psychological processes. Cambridge, MA: Harvard University Press. Wertsch, J. V. (1991). Voices of the mind: A sociocultural approach to mediated action. Cambridge, MA: Harvard University Press.
9 The continuum illiterate–literate and the contrast between different ethnicities Maria Sílvia Cintra1 Universidade Federal de São Carlos, Brazil
Introduction I present the results of a two-year postdoctorate research project developed at the Institute of Studies on Language (IEL/Unicamp, Brazil) and also of action research I have been organizing since 2006 when I started to work as a professor at the Federal University of São Carlos, also in Brazil.2 Relying on recent developments in new literacy studies, I explore the concept of the continuum illiterate–literate and argue that it implies elements of transformation, as well as conservation. I also argue that three intersecting continua must be considered together: the continua oral– written, rural–urban, and restricted–full literacy. Drawing on ethnographic fieldwork in Brazil, I show how elements of restricted literacy are still present on the threshold of the twenty-first century; how they entertain relation with rural to urban migration and with a marked contrast between different ethnicities; and in what sense this fact may be visible in the everyday use of language. My initial project was entitled “From Speech Acts to Literate Practices” and I intended to understand the interface between the written and the oral on the basis of the conceptions developed by Voloshinov (1973) and by Bakhtin (1981, 1986) concerning speech acts and discourse genres. Two years ago, I was not yet aware of the fact that oral and written languages consist of overlapping realities, which are impossible to understand as separate unities. In this sense, I still defended the idea of a continuum of discourse genres which would range from oral to written language or from informal to formal registers in a somewhat separate fashion. That is the way I imagined a transition from speech acts to literate practices when I first started to visit a community on the outskirts of São Carlos, a city of about 220,000 inhabitants in the southeast of Brazil, 95 percent of whom reside in the city and only 5 percent in the rural area. As I began to collect field data, I continued to study, together with the group of students I supervised, Bakhtin and Voloshinov’s theory, and to improve my understanding of the sociohistorical approach from Fairclough’s tridimensional proposal (1992, 1995, 2003). Some important
112 M. S. Cintra insights were given by Certeau (1980), and by Bortoni-Ricardo (1985) and Goffman (1959), as I gradually enlarged my understanding of language in general and started to envisage it more and more as a part of culture and ethics.
Method The methodology used for the action research was adopted since I not only intended to collect data regarding literacy, but I was also interested in education and in transformation (cf. Freire, 1972; Cameron et al., 1992; Thiollent, 1986). In his “Pedagogy of the oppressed,” Freire (1972) proposes a method of research which is intimately related with education, inasmuch as the limit between the research itself and education is very subtle and most often we have an overlap between both processes. In principle, it is argued that the educator should first visit the community where they intend to develop an educational practice in order to raise awareness of the “generating themes” they will work with. However, as the process of research is supposed to be carried out in a dialogical fashion, and as the researcher is expected to be attentive to the community’s problems and tensions, there is not only the movement of grasping the questions present in such a reality, but problem- solving is also very intensively present, regarding the necessity the researcher feels to discuss the questions that emerge in order to find the best solutions together with the subjects of the research. As a result, there is a rich blend of research and education. Cameron et al. (1992), on the other hand, defends empowering research as a form of academic research that endeavors to respect the subjects’ agenda, i.e., the emphasis is not centered on the researcher’s priorities, but rather on the problems and tensions present in a certain community and on the ethical compromise the researcher feels concerning questions of marginality, hegemony, contra-hegemonic forces, and empowerment. Thiollent (1986) defines action research as a kind of social research of empirical basis which is conceived and carried out in close relation with an action and with the solution of a collective problem. In this case, the researchers and the subjects of research are involved in a cooperative and participative fashion. In the case I will discuss, there was a constant movement from theory to practice, which is also a characteristic of action research. As a consequence, the results I will present will reveal this constant relation. I will show theoretical data together with certain vivid examples that illuminate these data and make it possible to think about theory in a more in-depth and detailed fashion. Taking this into account, I strongly believe in the importance of doing ethnographic research as a way of constantly checking the theory we deal with and building new conceptualizations. The initial research project I mentioned was based on the assumption that we still lack better knowledge of the heterogeneous reality of language
Continuum illiterate–literate and ethnicity 113 as it is practiced by individuals who are in the initial phase of written language. I have chosen a group of adults considered illiterate or semi- literate; I expected, however, that the results of my research would contribute to a reflection about literacy in general, and to the understanding of the continuum illiterate–literate in particular. I also believed that some aspects of the use of adult language could bring a better understanding of children’s and teenagers’ use of language and of their difficulties with school literate practices. It is worth mentioning here that according to a recent government survey from the IBGE (Brazilian Institute of Geography and Statistics), 7.4 million young Brazilians aged between 18 and 29 (from a total of 34 million urban youngsters, or 21.7 percent) have had only one to seven years of school study—which means abandoning school before completing the basic level of nine years. Moreover, 813,200—almost one million young people—are considered illiterate. Alagoas, a state in the northeast of Brazil, leads the ranking with 46 percent of youngsters in either case. On the other hand, São Paulo state, which is located in the southeast region and is the richest and most developed Brazilian state, has 15 percent of youngsters who are subject to what we may call “restricted literacy.” I coordinated biweekly meetings at a cooperative of manual workers on the outskirts of São Carlos, an average-sized city in the southeast of Brazil, with the double objective of teaching language and collecting research data. At the same time, I gave a course at the university and some of my students participated in the Saturday meetings with me. There were ten students who alternated in attending meetings with me, so that two or three academic researchers were always present. In the community, there were about three or six people. We attempted to convince more adults to participate in the meetings, but over time we understood that, although limited in number, the group could in a way serve as literacy agents in their everyday contact with the rest of the community. The cooperative consisted of about 260 manual workers, most of whom came from a recent process of urbanization. We used different strategies for collecting data, including semi- structured interviews, filming, tape-recording, and note-taking. Preference was given to semi-structured interviews which were considered more appropriate to action research in the sense that they imply more freedom and flexibility. Every meeting was filmed and the group of researchers alternated in taking notes and engaging in dialogues with the subjects. Another group of researchers transcribed the recordings.
Results The partial results I will present here involve, on one hand, discussion of the theory I explored while developing action research; on the other hand, they involve data collected in fieldwork.
114 M. S. Cintra
From speech acts to literate practices In “The Problem of Speech Genres,” Bakhtin (1986) proposes the difference existing between primary and secondary or complex genres. According to him, secondary genres of discourse appear historically in a circumstance of more complex and relatively more advanced cultural communication. It is worth paying attention to the observation by the Russian philosopher that secondary genres absorb and transform the primary genres previously constituted. Spontaneous verbal communication and informal letters are given as examples of texts pertaining to primary genres, whereas romance, theater, and scientific discourse would represent secondary genres. The former are defined on the basis of their spontaneous constitution and of a direct relation with immediate reality, and it is said that the latter are mainly written and entertain rapport with a world represented by language, rather than with the surrounding world. It is also worth remembering that, as a philosopher, Bakhtin was mostly interested in a phylogenetic approach and not so much in an ontogenetic one. It is certainly possible to think of language as ranging smoothly from informal to formal registers. However, when we assume Bakhtin’s theory as pertaining to a sociohistorical approach (together with Vygotsky’s, 1985) it is necessary to consider that the transition from one to another phase implies rupture as well as conservation. Furthermore, it also implies envisaging language as an inherent part of social relations and culture. Taking this into account, I gradually began to understand that it would be misleading to think of Bakhtin’s discourse genres without taking into consideration, simultaneously, Voloshinov’s theorization on social psychology and on ideology. I also understood that, in a certain sense, primary genres pertain to everyday life ideology, whereas secondary genres are part of the ideology of super-structures, but this assumption may also be misleading when lacking a dialectic view. Inspired by Freire’s previous work with adult learners, we called our biweekly meetings “cultural sessions.” At each session, I took a different text, always pertaining to a secondary genre: a short film, a poem, a map, an interview. Some students suggested that we should take three films together with a short explanation about each one in order to encourage the community participants to choose which they would prefer to see and therefore involve them more. They suggested the films Modern Times, Thank You for Smoking, and a short Brazilian film entitled Island of Flowers (Ilha das Flores) about nature preservation. The students understood that it was important to take films which involved some sort of social criticism; I pointed out that it was not only the content that was at stake, but the structure of each genre, in the sense that the inhabitants of that community were not used to the kind of irony and simulation which certain films explore. Nevertheless, we took the three films and, after a short talk, the participants chose the film Thank You for Smoking. At a certain point of the session, I noticed that the community participants
Continuum illiterate–literate and ethnicity 115 seemed somewhat bored and indifferent, so I decided to start making some intermediary comments in order to provoke their involvement. As a matter of fact, it was necessary to make some metalinguistic interference and some purposeful joining between the oral language of everyday life and the film so that such a text could be fully understood by those adult learners. It was plainly visible that a film, as a text pertaining to secondary genres, involves elements of dialogue typical of primary genres. From an ontogenetic point of view, however, we did not feel that this explanation was reasonable to make us understand why there still remains some difficulty for adults considered semi-literate to read and understand such texts. At first sight, it seemed to us that exactly because of the multimodality present in them, they would be easier to understand for the kind of students we dealt with. Practice, however, denied our assumptions and brought to light the fact that even humor was not as accessible as we imagined beforehand—at least a certain kind of intellectual or ironic humor. In the course of the process involving several “cultural sessions,” I gradually developed the understanding of two fundamental questions. First, the evidence illustrated that primary genres persist in secondary genres only in a simulated form—and not in a more direct or quite accessible fashion. In other words: following the dialogue in films similar to the one we had presented apparently requires a kind of experience different from the most simple one people deal with in everyday, face-to-face communication. A certain kind of dialogue explored in films as well as in advertisements, in cartoons or in comic strips is not as straightforward as it may seem to people who have benefited from the availability and the access to most complex literacy practices. Second, the evidence indicated that the simulation present in language must be accompanied by simulating new roles in social interrelation. If it was true—as I believed—that the continuum illiterate–literate should be seen together with the continuum restricted–full literacy, I was more and more convinced that language certainly has to be seen as part of social interplay, but not in a harmonious way, and rather in the sense of conflicting social struggle for hegemony and for full citizen participation. Taking this into account, restricted literacy as seen inside an ideological model must mean a restricted form of participation and of access to different opportunities in society in general. Language, thus, is not generally accessible unless it accompanies the playing and the simulation implicit in the roles we engage in as we take part in literacy practices inherent to determined social groups. In this sense, it may seem available without being plainly accessible (Kalman, 2004)—and accessibility has to do with real pertaining to social groups and sharing power.
Some considerations on restricted literacy Goody (1968) referred to West Africa and India as cases of restricted literacy. On the other hand, Gumperz (1986) referred to minority groups situated in the urban outskirts or in rural areas in the United States, and
116 M. S. Cintra pointed out the apparent difficulty for learning in these communities. I argue that in any of these cases, it is the same reality that may be envisaged by means of the continuum restricted–full literacy. In Brazil, as I understand it, we have a case of restricted literacy similar, up to a certain degree, to the Indian one. There are characteristics proper to restricted literacy (cf. Goody, 1986), such as: appeal to magic/religious conceptions; resource to formulaic style; tendency to secrecy; persistence of oral modes of instruction; emphasis on rote learning (even at university); oral residues in a literate culture; tendency toward preciosity. At first sight, however, there seems not to be, in this case, a strong association between writing and religion. It is possible, anyway, to recognize the intersection of three continua: the continuum oral–written, the continuum rural–urban and the continuum restricted–full literacy, which we can say intersect. Restricted–full literacy deals with questions related to power, to secrecy, to cipher, and to initiation. Without defending the autonomous consequences of literacy but, rather, understanding literacy from a sociological approach (Gee, 1990; Street, 1993; Barton & Hamilton, 1998), I started to consider some data related to the socioeconomic structure of Brazilian society that might explain why, in a twenty-first-century global society, in an emergent country, there are still traits characteristic of restricted literacy, even when the widespread code is alphabetical. Concerning the appeal to magic-religious conceptions, an incident that called my attention (and similar incidents occur with a certain frequency in adult classes) was when Iraci (a pseudonym) declared “I pray that Jesus may open my head so that I can read.” At first sight, it seemed only an emotional expression, but as I met other people from a rural background who made similar statements, I started to suspect that they represented a broader reality. Many of the women have their first contact with literate events when attending church. They cannot read, but they are literate in the sense that they take part in events centered on literacy (cf. Heath, 1983). As suggested by Kalman (2004), we can say that they have contact with literacy, but not real access to it. Other characteristics of restricted literacy—resource to formulaic style; tendency to secrecy; tendency toward preciosity—appear strictly related to the same question when we observe the behavior of the illiterate, for whom literacy seems to be inaccessible; on the other hand, I was shocked when Jessica (another pseudonym), a postgraduate student who belonged to our research group, said: “Here at the university, most teachers seem to hide their knowledge.” Again, it could be seen just as a common complaint of students, but as I had been aware of these questions related to restricted literacy and secrecy, I began to conceive the idea of a continuum ranging from restricted to full literacy, which is more visible in the case of illiterate people who get in contact with literacy, but which would also exist in other situations and even at a university level. The two other characteristics which Goody pointed out—persistence of oral modes of instruction
Continuum illiterate–literate and ethnicity 117 and emphasis on rote learning—are certainly present at lower levels—primary and high schools—but we can say they still persist at university when we consider that an oral mode of instruction does not only imply the behavior of teachers and students in the classroom, but also the way written culture is still frequently regarded inside an untouchable bell jar, particularly in what concerns western European or American literate culture. As I understand it, this fact is partly a consequence of our belonging to a Third World country whose academic culture is mainly imported from First World nations and often arrives in our country not as a result of autonomous field research, but rather as ideology in the sense constructed by Foucault (2002).
The idea of assimilation: ideology and hegemony When taking into account the concepts of ideology and hegemony (Fairclough, 1992, 1995, 2003), I could envisage discourse genres in a more complex way. I gradually understood that genres assimilate characteristics previously existing (in the sense of the assimilation of primary genres existing in secondary genres) and that they reveal these characteristics because they are crossed by vectors of ideology and hegemony which, in principle, belong to the social dimension. It is interesting to point out that for Voloshinov (1973), ideology and power are constantly crossing social relations, which is different from the orthodox Marxist standpoint concerning infrastructure and superstructure relations. Considering this, if it is true that secondary genres assimilate and absorb all primary genres previously constituted (cf. Bakhtin, 1986), this is only possible because certain elements proper of power and ideology persist in social relations in general. Amanda (a fictitious name) was a very active member of the cooperative of cleaners in Jardim Gonzaga—the neighborhood on the outskirts of São Carlos where I developed my field research. She lived alone with her two children and showed much interest when coming to our meetings. She used to say that she did not like to engage in much conversation with her neighbors, that she had her private life and was not fond of much contact with others. With time, however, we understood that, paradoxically, she was a supportive woman: her eyes lit up whenever she mentioned her plans concerning either her family (brothers she had left far away in the northeast of Brazil and that she wanted to help) or the neighborhood (one of her dreams was to set up a kindergarten for children to stay in while their parents studied). One Saturday afternoon, Amanda stated: “I don’t like to talk, because my speech is full of mistakes. People listen and notice that I say several words wrongly. I am ashamed, so I prefer to be silent.” I listened to her and could hear no mistakes. I commented to her that her speech was absolutely right, but she said she still wanted to learn the “right” way, only then she would be satisfied. What she understood as the “right” way involves the accent proper to southeastern speakers and, sure
118 M. S. Cintra enough, a certain way she recognizes in the speech of Brazilians who entertain a more sophisticated literate condition. Or, in other words, the form of speech proper to citizens who are most involved in complex genres of discourse.
Ethos and ethics Brazil is still considerably shaped by traits belonging to rural communities and culture.3 Since the beginning of the twentieth century, there has been a permanent movement characterized by rural exodus and by the continuous insertion of illiterate people into the literate universe of different cities. This fact has meant complex consequences for communication as different ethnicities are involved implying different ethé which are often ignored when education is centered on certain matters considered of universal reach, and also when developing ethnographic and action research. There are some different traits pertaining to each ethnicity—rural illiterate ethnicity and urban literate ethnicity—which point to a link between ethics, on one hand, and ethnicity, on the other. I think such an approach can contribute to a better understanding of literacy and of the continuum illiterate– literate and that this understanding is fundamental for education as a whole, and not only for beginning literacy instruction. Moreover, it can provide a better understanding of the communication between academic researchers and the subjects of their research, as in the case of an experiment developed by another group of professors and academic researchers of a Brazilian university who wanted to contribute to building cheap and comfortable houses for rural inhabitants, as well as developing popular cooperativism, but have had difficulties implementing their proposal.4 I had been reading about the concepts of ethos and ethics when I started visiting the community under discussion. In principle, I understood ethos as a part of language itself, inasmuch as language and identity are intimately linked. As I observed conversation between a professor and the president of a popular cleaning cooperative, I began to see that ethos is related to questions of ideology and hegemony as well, in the sense that once it represents a habitus, it is concerned with the culture and the ethics present in a certain community. The professor addressed the popular worker in what we can call a professoral ethos: he asked certain questions for which there was an expected answer, as if he were in fact in a classroom environment. In this case, it is interesting to think of what Fairclough (1992) calls hybridism. It is worth mentioning the fact that not only genres undergo hybridization: there are social spheres that intersect. It is as if the professor had forgotten that he was engaged in the genre of an informal conversation (pertaining to primary genres) and started behaving in tune with a classroom ethos, i.e., in accord with secondary genres. As I will discuss below, this fact also has to do with the continuum restricted–full literacy, in the sense that, partially, we can say that the professor assumed a literate ethos whereas the cooperative worker played an oral role; on the
Continuum illiterate–literate and ethnicity 119 other hand, it is possible to see, in this behavior, a difficulty of communication proper to a place where literacy is still restricted to a few. In fact, there seems to be an abyss between our literate ethics and logical thinking and the way some illiterate people think and behave. Furthermore, this transcends logic in the sense Luria (1979) brought to light: it has to do with logic, but it also has to do with ethics. In this case, I concluded that the difference between oral and scriptural ethé might be at stake. Oral ethos implies reference to family and community bonds whereas the acknowledgment of rules present in broader social coexistence is typical of scriptural ethos. In either case, however, we have strict rules which are learnt implicitly or explicitly belonging to particular social groups. It is, therefore, impossible to think about different discourse genres belonging to different social practices and behaviors as smoothly ranging from one to another pattern. The transit from restricted to full literacy implies changing one’s participation in a certain ethic in order to start one’s familiarity with another one. Social roles, however, are not so smoothly or harmoniously interchangeable.
Discussion: when rural illiterate meets literate culture— confrontation and assimilation My previous discussion concerns the fact that it is not just genres that incorporate elements present in pre-existing realities; it is social reality that involves the coexistence of old and new elements, of archaic and modern social relations and ethé. Taking this into account, when illiterate rural adults start to have contact with urban literacy, it is not only the written language they have to master: there are different ethnicities in confrontation or, making use of the Bakhtinian conception, there is an arena of social struggle and conflict. There is a fight for hegemony, which does not only imply a struggle for participation in literate alterity and ethos, but it also implies a confrontation and an attempt toward maintaining old structures with which one is most concerned. Again, we have the intersection of three continua: illiterate– literate, rural–urban and restricted–full literacy, which, in a certain sense, consist of the same continuum, as far as we take into consideration social reality as a whole and all the elements it is formed of. Postmodernity emphasizes diversity and, at this point of my conclusion, it is pertinent to raise the question: in what sense is it possible to coexist with diversity or to provide that different ethnicities coexist when we know that confrontation and a struggle for hegemony is involved? Or in other words: in what sense should different ethnicities coexist when they involve questions related to ideology and to power? Many years ago, when I taught at a secondary school in a very poor neighborhood located on the seashore of a southeastern Brazilian town, I was explaining to my 14-year-old students sociolinguistic concepts related to variation and diversity. I said that there was a verb form in Brazilian
120 M. S. Cintra Portuguese that had more prestige (eu pus) and another that was also correct, but suffered stigmatization (eu ponhei). I explained that their use depended on social context, so that each of them could be adequate for a different situation, but neither could be considered incorrect. In a peculiar form, I was defending postmodern relativism and diversity. I said that “eu ponhei” could be used at home or in the neighborhood, together with friends or with their family; and that “eu pus” should be preferred in more formal contexts.5 A tall Afro-Brazilian student raised his hand and said: “I do not agree with this concept. Before, I used to say ‘Eu ponhei,’ but as soon as I knew this was a wrong verbal form, I did not want to pronounce it any longer.” The prejudice against the popular variety of Brazilian Portuguese is very deep-rooted, characterizing restricted literacy. Ideology and the struggle for hegemony are definitely present in everyday relations and conversation, in the arena Bakhtin refers to, of dialectic and dialogical social confrontation and struggle. In this context, it becomes very difficult to defend the pacific coexistence of diverse registers, because one is not able to convince citizens on the basis of argumentation, when the very struggle for recognition in society depends on the domination of the language of the other and of his proper ethos and alterity. The continuum restricted–full literacy as far as it also involves rural to urban migration points to the fact that history is at stake and that linguistic studies are part of a broader reality that calls for a sociohistorical and anthropological approach. The defense of a pacific coexistence of the diverse may be possible in countries where a certain social equality is present. In the case of Brazilian Third World reality, however, we are still in the middle of a historical transition from rural to urban life which involves the difficult conquest of a place of citizenship and of more thorough social participation. This reality implies, on one hand, the recognition and acknowledgment of diversity and alterity; on the other hand, however, it requires the very struggle for hegemony, which means the struggle to ensure a prominent place reserved for protagonist roles and not just for supporting or subservient ones.
Notes 1 Department of Modern Languages—UFSCAR, São Carlos, State of São Paulo, Brazil, CEP 13.564-060. Email:
[email protected]. 2 I acknowledge Fapesp (04/15539-3) for supporting the initial two years of fieldwork. 3 In the city of São Carlos where my research was developed, and which is part of the richest region of our country, only 5 percent of the population resides in the rural area, contrasting, for instance, with Cabeceiras, a town of about 8,300 inhabitants in the northeast of Brazil, where more than 85 percent of the population still live in a rural area. 4 I take part in an interdisciplinary research team which includes professors, undergraduates, and postgraduate students from different subject areas: engineering, architecture, health, and social sciences. The experience implying the
Continuum illiterate–literate and ethnicity 121 c onstruction of houses was developed in a sector related to mine. It was not part of the literacy project, but my group of students and I were able to follow the development of their action and take notes related to our interests. 5 The contrast between “he isn’t” and “he ain’t” may be given as an equivalent for this variation in Brazilian Portuguese. It should be mentioned that this is not a direct translation of “eu pus”/“eu ponhei.”
References Bakhtin, M. (1981). The dialogical imagination. Austin, TX: University of Texas Press. Bakhtin, M. (1986). The problem of speech genres. In Speech genres and other late essays (pp. 60–102). Austin, TX: University of Texas Press. Barton, D., & Hamilton, M. (1998). Local literacies: Reading and writing in one community. London and New York: Routledge. Bortoni-Ricardo, S. M. (1985). The urbanization of rural dialect speakers: A sociolinguistic study in Brazil. Cambridge: Cambridge University Press. Cameron, D., Frazer, E., Harvey, P., Rampton, M. B. H., & Richardson, K. (1992). Researching language: Issues of power and method. London and New York: Routledge. Certeau, M. (1980). L’invention du quotidien: Arts de faire. Paris: Gallimard. Cook-Gumperz, J. (2006). The social construction of literacy. Cambridge: Cambridge University Press. Fairclough, N. (1992). Discourse and social change. Cambridge: Polity Press. Fairclough, N. (1995). Critical discourse analysis. London: Longman. Fairclough, N. (2003). Analyzing discourse: Text analysis for social research. London: Routledge. Foucault, M. (2002). The archaeology of knowledge. London: Routledge. Freire, P. (1972). Pedagogy of the oppressed. London: Penguin. Gee, J. P. (1990). Social linguistics and literacies: Ideology in discourses. London: Falmer Press. Goffman, E. (1959). The presentation of self in everyday life. New York: Doubleday Anchor. Goody, J. (1968). Literacy in traditional societies. Cambridge: Cambridge University Press. Heath, S. B. (1983). Ways with words: Language, life and work in communities and classrooms. Cambridge: Cambridge University Press. Kalman, J. (2004). Saber lo que es la letra: una experiencia de lectoescritura con mujeres de Mixquic. Mexico: Siglo Veintiuno. Luria, A. R. (1979). The making of mind. Cambridge, MA and London: Harvard University Press. Martins, M. S. Cintra (2007). Ethos, gêneros e questões identitárias. Revista Delta, 23, (1). São Paulo: EDUC. Street, B. (1993). Cross-cultural approaches to literacy. Cambridge: Cambridge University Press. Thiollent, M. (1986). Metodologia da pesquisa-ação. São Paulo: Cortez. Voloshinov, V. I. (1973). Marxism and the philosophy of language. New York: Seminar Press. Vygotsky, L. S. (1985). Thought and language. Cambridge, MA: The MIT Press.
10 Strategies, policies, and research on reading and writing in Colombian universities Blanca Yaneth González Pinzón Universidad Sergio Arboleda, Colombia
The university does not know how to fulfill its own requirements! If it knew how, students would be able to read and write as she demands them to. Gloria Alvarado
How and why did Colombia redefine the teaching and learning processes of reading and writing? In Colombia, a new network has been established to advance reading and writing in higher education. Previously, as in other Central and South American countries, teaching reading and writing has for years been associated with linguistic and philological erudition in the mother tongue. This has left the responsibility to work on these skills to a single subject in the curriculum: Spanish or Castilian, as the subject has been defined since the days of conquest and colonization. In this subject, reading and writing exist alongside phonetics, semantics, morphology, syntax, orthography, etc. Nor has the rise of literary studies alongside language studies helped reading and writing gain in significance in primary or secondary schooling. Later, we shall see how this same way of conceiving reading and writing within the realm of language is brought into the university, largely as a remedial subject. However, reading and writing did not always appear to be matters associated with the process of acquisition and development of the language. In many cases, we find them linked with different daily and cultural activities of the peoples (let us remember the role of scribes and the role of reading and writing for the commercial activities) (Viñao, 2002). Educational practices with origins in medieval Europe and imported into Spanish colonies in the Americas such as cartillas (a type of grammar drill) continue to be fundamental classroom tools. Spelling was also imposed, in search of good diction. Reading aimed to bring excellence in reciting (at the end of the nineteenth century, reciting was still very common in schools). Viñao says that in 1893, there were eight types of textbooks for reading: the spelling books or cartillas, short stories, libros
Reading and writing in Colombian universities 123 de cosas (the “books of things,” which are books with information or lessons on diverse topics), biographies, miscellaneous books, books in verse, manuscripts, and treatises related to one or more school subjects— today’s textbooks. Moralizing and religious topics were interspersed in the lessons that were developed in such reading. Writing, for its part, was intended to perfect handwriting and improve the orthography. Literacy education in general, was subjected to a process of normalization, concentrating on good speech, vocalization, intonation, and orthography. At the beginning of the twentieth century, with increased class sizes (to this day, 50 or 60 children per class is typical), silent reading became preferred to facilitate classroom order and study practices. Tests which valued comprehension and speed took a central role in defining educational goals. Reading practices were influenced by physiological and social hygiene, emphasizing such concerns as posture and good manners in using notebooks. As reading took priority over writing, writing activities were limited. In primary and secondary schools, the most common acts of writing were dictations and compositions, along with taking notes of the lecture and written exams, which replaced oral examination, again due to the increase in class sizes. This inheritance has influenced social attitudes toward literacy instruction, has limited the educational opportunities for literacy instruction to a single subject, and has defined a primarily grammatical approach to writing education. Various linguistic approaches have been used to achieve grammatical goals, including structural grammars, generative- transformational grammars, and functional grammars. In the 1990s, a communicative semantic focus expanded linguistic approaches beyond grammar, and gradually other disciplines such as philosophy and logic were seen as relevant to reading and writing. Psychopedagogy also turned our attention to the practices of teaching and learning reading and writing, the evolutionary processes of the acquisition of the code and the cognitive modifications involved (Ferreiro, 1999). Currently, discourse analysis of several varieties has given us new tools to examine pedagogical and learning processes. While these developments have changed the thinking of many educators, they have also generated some resistance among those who seem reluctant to change. Whatever position teachers may hold, these controversies have brought attention and discussion to reading and writing and their pedagogic implications. Political guidelines and regulations, even while being implemented have also generated energetic inquiry suggesting that they have been adopted without much reflection (Bustamante, 2002), and in some cases are simply translations of foreign laws, such as the LOGSE from Spain, accommodated to the national context. In particular Law 115 (1994) redefined many of the paths through which the country’s education was being directed, including improving reading and writing at each educational level. The law introduced into the professional educational discussion terms such as PEI (Proyecto Educativo Institucional—Institutional Educational Project),
124 B. Y. González Pinzón a cademic calendar, interdisciplinarity, remedial courses, promotion, process, achievement, achievement indicator, performance, skills, guidelines, competency, and standard. Initially the Law generated all sorts of reactions among primary and secondary educators. Those in higher education at first ignored it because they believed that the Law did not invade their professional space. However, symposiums, conferences, discussions, and orientation workshops for teachers gradually permeated higher education, and the consequences became evident of the strong work that was promoted at other levels of education. When our secondary school students entered university, they were faced with a system of evaluation and promotion considerably different from what they had known in high school. In universities, terms like remedial courses and evaluation of achievement were not used. This forced higher education to participate more in the conversation and accommodate so that, 14 years later, the differences are not so drastic. The Colombian Institute for Promotion of Higher Education (Instituto Colombiano para el Fomento de la Educación Superior, ICFES), in charge of designing and administering the state exam for university admission, emphasized reading and writing in the examination and analysis of results in the national exams and also in the international exams such as the TIMS and the PISA. Different researchers, the majority from public universities, contributed to this turnaround with the publication of materials (Jurado & Bustamante, 1995) and the preparation of the educators for the new challenges that faced them in the development of these abilities. The Evaluation of Basic Competences in Language, Mathematics, and Sciences (Evaluación de Competencias Básica en Lenguaje, Matemática y Ciencias) was introduced (from 1988 to 1999), and a document (Bogoya, 2000) that analyzed the published data on the pilot exams was circulated in each Colombian high school in order for each institution to organize their plans of restructuring. Formerly, the Pruebas Saber (Knowledge Exam) was used at different levels, with the official slogan of “Evaluate to Improve.” During these years, the Colombian Network for Transformation of Educators’ Formation in Language (Red Colombiana para la Transformación de la Páctica Docente en Lenguaje) was also created. All of these processes of change, with their strengths and shortcomings, also mobilized reflections in universities and pedagogic institutions where the teachers are trained. These approaches had to be redefined and, although a decade is short for generating definitive changes, a different focus for thinking about these processes does exist in Colombia due to this.
The context of higher education in Colombia Almost five million Colombians are between 18 and 23 years old, which is the age of professional development (10.9 percent of the total population). There are 276 institutions of higher education; the private institutions account for 195 of them, while the public institutions account for only 81.
Reading and writing in Colombian universities 125 There are 51 professional-technical institutions, 60 technological institutions, 91 university institutions, and 74 universities as such. Universities account for the highest number of students and the technical institutions come after them. Between 1996 and 2005, registration in higher education jumped from 15.3 percent to 24.6 percent. Of the 433 accredited programs (the majority of which are engineering, architecture, urbanism, and other related fields), 204 are private universities and 229 are public. Only 11 universities have the accreditation of high quality for the moment. Others are in the process of obtaining it. One out of every two students does not complete their studies, although the annual drop-out level decreased from 16.5 percent in 2003 to 13.4 percent in 2005. In the first semester in 2005, the ten academic undergraduate programs with the highest number of students were: Law (69.87 percent), Business Administration (60.23 percent), Public Accounting (52.30 percent), Industrial Engineering (34.81 percent), System Engineering (32.96 percent), Medicine (28.56 percent), Psychology (26 percent), Electronic Engineering (21.49 percent), Civil Engineering (18.24 percent), and Nursing (16.36).1 In terms of postgraduate education, the most recent data published by the SNIES (Sistema Nacional de Información de la Educación Superior—National System of Information about Higher Education) indicate that there are 3,937 active postgraduate programs in Colombia; 3,412 are diploma programs, 443 are Master’s, and 82 are doctorate programs. In higher education, before the General Law of Education (la Ley General de Educación; Law 115 of 1994), other laws and decrees had already been published, such as Law 30 of 1992 (still in force today), that organized the public service of higher education, establishing the principles and guidelines by which universities would govern themselves. For its part, the National Council of Higher Education (Consejo Nacional de Educación Superior, CESU) collaborated with the State in its function of promoting and orienting scientific and technological development, in agreement with what was established by Law 29 of 1990. A relatively recent decree, Law 2566 of September 2003, established the minimal condition of quality and other requirements for the offering and development of academic programs of higher education. Article 6, in particular, makes a reference to the development of research skills: The institution must explicitly present how to develop the research culture and critical, autonomous thinking, which allow students and professors to develop new knowledge, taking into account the methods of the development of such skills. For this purpose, this program should incorporate measures to develop research and to gain access to the advances of the knowledge. This law is seen as calling for the promotion of scientific thinking. The CRES (Conferencia Regional de Educación Superior—Regional Conference of Higher Education), recently held in Colombia, makes very specific
126 B. Y. González Pinzón demands on Latin American and Caribbean countries with regard to “politics that reinforce the social commitment to Higher Education, its quality and pertinence, and the autonomy of the institutions,” and in the same way, make an urgent and emphatic call for the members of the educational community, particularly to those who are in the position of political and strategic decision making . . . to consider the approaches and the lines of action that have been derived from the sustained debate about the priorities that Higher Education must assume, on the base of clear awareness about the possibilities and contributions that it makes to the development of the region.2 All of this was strongly associated with the demand to watch over the development of science and research. What is curious about all these discussions is that reading and writing is never put in a direct relation to scientific advancement and the production of knowledge. The project described below is an attempt to reunite reading and writing with science.
Reading and writing in higher education: a return journey to the essence of university Reading and writing have always been associated with academia. Nonetheless, despite being so integral to its existence, reading and writing have come to be taken for granted in academia as automatic, silent devices, which do not deserve much attention. What is being said now with a marked emphasis about reading and writing in higher education clearly suggests two ideas: on one hand, the effects of the inattention to reading and writing are now uncontrollable, and on the other hand, it is necessary to recognize the importance of reading and writing in higher education. In Colombia, to address this problem with will and determination we have already taken up a community-building team approach. Almost 40 projects, completed or in progress, are seeking concrete classroom solutions from the classrooms that fit the local context of each university. These projects have been sharing their experiences and have been reporting results as part of national accountability. This initiative has recognized the role of institutions in teaching reading and writing to contribute to an academic culture in the country, and has recognized that each discipline is a discursive space with its own practices of reading and writing, not comparable to the practices that are carried out at other levels of education (Carlino, 2005). While research is new and limited, with much still in progress, a number of studies have been published. Some studies describe the students’ level of performance in comprehending and producing texts (Rincón et al., 2005). Other studies have attempted to diagnose difficulties so as to design correctives (Cisneros, 1995). A third group of studies examines students’ use
Reading and writing in Colombian universities 127 of reading and writing in new technologies and the ways new technologies enhance learning; other studies develop instructional models to advance abilities; and, finally, a few case studies examine reading and writing in some disciplines (Narváez & Cadena, 2008). It is apparent that, unlike in other countries,3 the focal point has not been analyzed profoundly: the problems with the type of frequent practices of reading and writing in university, the presence or absence of such practices in some disciplines, their quality when they are present and what they promote in their dissimilarity, and something very important—the type of readers they create. Many research projects and actions still start from erroneous assumptions such as the belief that there is a deficit in the cognitive processes of the students, when in reality, students are finding it difficult to identify the multiple literacy paths they must take in the university. There is a sort of “schizophrenia” that torments the students, who must engage different practices of reading and writing in each academic space, with no clear sense of what knowledge or practices can travel from one context to another. From these studies, it is easy to identify the importance of the culture of reading and writing that slowly and without warning has been installed in the institutions. From my point of view, two problems emerge strongly in these studies. The first coincides with Foucault’s explanation of a developing discipline’s problematization of the knowledge of prior knowledge in the domain, and of the emergence of its own specialized language of that domain, which can only be accessed with guidance from specialists. There is not yet much awareness about this situation in our institutions. The other is strongly related to what is called comprehension. We all know about the recurring complaint from educators about the problems of comprehension that students bring (“bring” refers to what students learned in the secondary schools) and about the complaints that students cannot write. The concern for comprehension has dominated thinking about reading (Argudín, 2001). “Reading comprehension exercises” and “workshops on reading comprehension” have dominated curriculum. But comprehension is only an aspect of a process, not the ultimate goal. Thus, it is necessary that the universities clarify the objectives of reading. After comprehension, a student must know what to do with the information he now understands. Reading serves many purposes: reading for research, reading for organizing information, reading for resolving a problem, reading for conceptual learning, reading for comparing theories or points of view, reading for refuting, and reading for making a comment. The sole focus on comprehension has led to a university culture where required information dominates student thinking about activities. Seldom do we hear students say “this material, which is written to compare these two approaches, is due on this day,” or “the document that we are going to read to comment on is due on this other day”; most of the time, as Professor Alvarado (2007) says, this topic is reduced to the photocopy. Generally, students associate the reading with the physical material, “the
128 B. Y. González Pinzón photocopies are for the coming week.” Universities must reconsider their situation in relation to the preoccupation with comprehension that secondary schools have shown for a long time. Over time, we will find young individuals arriving at university with good academic levels. Is the university prepared to work with high levels of “comprehension”? If the student already comprehends well, what will these spaces in the university be dedicated to? There is a deeper question that lies below the situation I have described. The development of science and the production of knowledge are the essence of the university. The university should develop the processes of thinking that allow individual students to reason, analyze, and argue, as well as to recognize the ideological forces that are mobilized in different types of discourse. To achieve this, students must engage in concrete practices that question the essence of things and the role of language in formulating knowledge, says Alvarado. However, in universities currently, reading and writing are considered isolated skills, in large part to prepare the students with skills to carry out tasks in the working and professional worlds. Of course, we know that the Latin American context does not offer major alternatives to its inhabitants in this respect, as reflected in the common saying: “work and acquire the essentials to survive, or disappear.” Yet this imperative for our students and society to reach toward the highest level of knowledge and thought has to struggle against two tendencies in the Latin American context. Some who see a great divide between work and academic life pressure the university to provide narrow vocational training. Equally troubling is that despite bright spots of excellence, the university in Latin America does not always provide models of the highest academic discourse that ought to exist in a modern university. Juan Ramón de la Fuente (2008) goes as far as to assert that “Latin America exists in the slum of the Age of Knowledge.”
The network of reading and writing in higher education (la red de lectura y escritura en educación superior, REDLEES): work of the institutional policies4 Between 2004 and 2007, in this country, many spaces for discussion were opened through academic conferences, in order to share concerns about the difficulties presented by the students in the processes of development of reading and writing: the First Conference on Reading and Writing in University Education (Primer encuentro sobre lectura y escritura en la educación universitaria), organized by the Universidad Autónoma de Occidente in the city of Cali (May, 2006); the Regional Conferences on Teaching of Reading and Writing in University (Encuentros regionales de la enseñanza de la lectura y la escritura en la universidad), that was held for three consecutive years in Antioquia, organized by the Universidad de Medellín (2004), EAFIT (2005) and the Universidad Pontificia Bolivariana
Reading and writing in Colombian universities 129 (2006); the work with the UNESCO Lectures that the Escuela de Ciencias del Lenguaje de Univalle put forward and the work of the Network for the Transformation of the Educators’ Practices in Languages. National Conference on Institutional Policies for the Development of Reading and Writing in Higher Education (Encuentro Nacional sobre Políticas Institucionales para el Desarrollo de la Lectura y la Escritura en Educación Superior), held at the Universidad Sergio Arboleda on April 26–27, 2007, under the auspices of the Colombian Association of Universities (Asociación Colombiana de Universidades, ASCUN), gave rise to the structure of REDLEES (Red de Lectura y Escritura en Educación Superior—the Network of Reading and Writing in Higher Education)5 that currently brings together 67 IES (Instituciones de Educación Superior— Institutions of Higher Education) in the country. The April 2007 conference allowed us to make a very close inventory of what has been done in this respect so far. In the event, 190 representatives from 73 universities participated. Among the discussion topics put forward by the conference was that of Institutional Policies and Experiences of Development of Reading and Writing in Higher Education. At that time, proposals on institutional policies were being advanced at eight universities. The Conference produced a manifesto stating that the discussion on this topic should be continued with the universities that participated in the Conference as well as others in the country. Why did REDLEES make the institutional policies its theme or flagship for work? First, it is because we recognize that the enormous and well- intentioned efforts, which are isolated from the educators in the institutions, would not propel a culture of university reading and writing if it did not take into account the intervention and investment on the part of the university directors. And second, it is because there are a considerable number of people who are involved in a particular local situation who deserve a planned intervention on the part of those who influence their destiny in a certain way, as the experts in policies such as Kingdon’s (1995) explain. The creation of policies in each institution to work on reading and writing is urgent because reading and writing are associated with very sensitive aspects of university life such as the dropout rate, low academic performance, research enhancement, production of knowledge, and development of professional competitiveness. Even more, difficulties in reading and writing exclude large numbers of students from entering a public university. Given the importance of these policy issues, we have grappled with whether Colombia is in a position to create institutional policies that will respond to the need. Kingdon suggests that the success of a policy proposal depends first on the political climate, which has a bearing on the receptivity that a proposal can achieve. A single institution looking for solutions cannot create enough of a climate for change. Second, it depends on the interests organized around those policies, such as the unions, groups, associations, and/or academic groups. Third, it depends on the communities of
130 B. Y. González Pinzón experts or specialists (educators, researchers, academics, consultants, and analysts). Fourth, it depends on the continuity of ideas and their renovation. According to Kingdon, it also depends on “softening up” on the part of the directors of the institution, and in our case, of the state and private entities associated with the topic. Similarly, it depends on the capacity to cope with the restrictions and obstacles that are involved in these processes (budget, regulations, procedures, necessities, etc.). And finally, it depends on the ability to view the problem as something structural rather than temporary, which requires policies that are adopted long term. To this I would add that it is necessary to rely on previous attempts that have demonstrated an interest in reflecting upon a problem. We are committed to addressing these challenges. The discussion of policies at the April 2007 Conference raised numerous further questions:6 how should we understand an institutional policy aimed at developing reading and writing in higher education today in our country? Who should participate in its development and execution, and how? What would be the objectives that the said policy would have to aim for? Toward whom will it be oriented and to what necessities will it respond? What components would it have? How should we work on the development of reading and writing in terms of the curriculum? Questions extended from immediate policies to knowledge about underlying needs for reading and writing: what is the place for reading and writing in the professions and disciplines, and how must we arrange the practices for its development? What place does it hold in society? Are we aware of the interdisciplinary treatment required in those processes in which we must create a relation between concepts and practices of a language professor and those of a professional or an expert on the knowledge specific to each specialization? Other questions we are addressing include: what is reading in the university and what are its objectives? How should we follow up with the processes of the development of reading and writing? What do professors and students do with what they read and write? Are reading and writing mere active tools for the evaluation? How do students work on assigned homework on reading and writing? What are the formative interests of the educators? Are educators trained to follow up with the practices of reading and writing in each discipline and evaluate them? Does the specialist clarify for the students how to comprehend, associate, and incorporate the codes of the discourse in their disciplines and professions? That the questions exceed the answers is symptomatic of the recognition of just how much more we must research and how large a field we are entering.
Notes 1 Ministry of National Education. Bulletin no. 6, January–March 2006. 2 Declaration of the Regional Conference on Higher Education in Latin America
Reading and writing in Colombian universities 131 and the Caribbean held in Cartagena de Indias on June 4–6, 2008. Retrieved June 12, 2008, from www.udual.org/Anuncios/DeclaracionCRES2008.pdf. 3 To learn more about the state of research in other countries, see Carlino (2007) ¿Qué nos dicen hoy las investigaciones internacionales sobre la escritura en la universidad? Available at www.ascun.org.co. 4 The definition of institutional politics is as follows: a way to comprehend and conduct the development of the skills in reading and writing that includes two types of approaches as a minimum: (i) explicit approaches on the relations between the university institution and the society, and within them, on the role that reading and writing practices should have in the professional performance and in the graduates’ actions as individuals in our societies, (ii) general approaches on how this development should be advanced in the study plans of the institution (in other words, the appropriate types of pedagogy, the forms of support for the independent research, the profile and the training of the required educators, the recognition and the treatment of the formative differences that exist among students, how a curriculum should be organized to carry out the training on projected reading and writing,—whether or not it deals with a transversal development that goes beyond what is done in the traditional course of reading and writing, etc.). 5 REDLEES is a Colombian network of institutions of higher education that promotes dialogue on reading and writing at university. The network acts as a community of orientation and cooperation in the actions implemented by the institutions and the educators to strengthen these processes of superior development. 6 As a result of the first conference on the topic of Institutional Politics for the Development of Reading and Writing, the first document produced by the REDLEES, elaborated by Alejandro Gordillo from the Universidad Central, was circulated. The document synthesized the principle questions on the topic.
Works cited Alvarado, G. (2007). Los lugares de la Lectura y la Escritura en la Formación Universitaria: Didácticas y Pedagogías de la Enseñanza v.s. Pedagogías de la Acción. Presentation in the National Conference on Institutional Policies for Development of Reading and Writing in Higher Education, Bogotá, Colombia. Available at www.ascun.org.co. Álvarez, M. (1998). La interdisciplinariedad en la enseñanza aprendizaje de las ciencias en el nivel medio Básico. La Habana: Instituto Superior Pedagógico. Argudín, Y. et al. (1995). Aprender a pensar leyendo bien. Habilidades de lectura a nivel superior. México: Plaza y Valdés editores. Argudín, Y. et al. (2001). Desarrollo del pensamiento crítico. Libro del profesor. México: Plaza y Valdés editores. Bode, J. (2001). Helping students to improve their writing skills. Retrieved February 11, 2008, from www.jcu.edu.au/school/sphtm/documents/gctt01/article5.htm. Bogoya, D. (Ed.). (2000). Resultados de la evaluación de competencias básica en lenguaje, matemáticas y ciencias. Bogotá: District Secretary of Education. Bustamante, G. (2002). Estándares curriculares: ¿inofensivos? Revista Educación y Cultura, 63. Fecode, pp. 7–12. Carlino, P. (2005). Escribir, leer y aprender en la universidad. Una introducción a la alfabetización académica. Buenos Aires: Fondo de Cultura Económica.
132 B. Y. González Pinzón Carlino, P. (2005). Representaciones sobre la escritura y formas de enseñarla en universidades de América del Norte. In Revista de educación N. 336. Madrid. Carlino, P. (2004). Enseñas a escribir en la universidad. Cómo lo hacen en Estados Unidos y por qué. In Revista Iberoamericana de Educación. Carlino, P. (2004). Escribir y leer en la Universidad: responsabilidad compartida entre alumnos, docentes e instituciones. In Textos en contexto, 6. Carlino, P. (2004). Escribir a través del currículo. Tres modelos para hacerlo En la Universidad. In lectura y vida. Year 25. Carlino, P. (2002). Leer, escribir y aprender en la Universidad. Cómo lo hacen en Australia y por qué. In Investigaciones en Psicología. Revista del Instituto de Investigaciones de la Facultad de Psicología UBA. Year 7 No. 2. Carlino, P. (2002). Alfabetización académica: Un cambio necesario, algunas alternativas posibles. Third Conference “La universidad como investigación,” Dept. of Psychology, Universidad Nacional de la Plata. Cisneros, M. (2005). Lectura y escritura en la universidad. Una investigación diagnóstica. Pereira: Universidad Tecnológica de Pereira. Ferreiro, E., with José Antonio Castorina, Daniel Goldin & Rosa María Torres. (1999). Cultura escrita y educación: Conversaciones de Emilia Ferreiro. México: Fondo de Cultura Económica. González, B. (2006). Experiencia de alfabetización académica en la Universidad Sergio Arboleda de Colombia. Presentation in the First National Congress “Reading, Writing and Speaking Today,” Tandil, Argentina, September 28–October 1, 2006. González, B. (2008). Dispositivos para pensar una política institucional para el desarrollo de la lectura y la escritura en la Educación superior. Conference held at the Universidad Autónoma Metropolitana de México. México D.F. February 20, 2008. González, B. (2008). La lectura y la escritura como saberes interdisciplinarios. Opening conference held by the Faculty of Education of the Universidad Minuto de Dios, March 3, 2008. Jurado, V. (2005) Los procesos de lectura y escritura. Bogotá: Editorial Magisterio, National Ministry of Education, Bulletin no. 6. January–March, 2006. Kingdon, J. W. (1995). Agendas, alternatives and public policies. New York: HarperCollins. Marruco, M. Enseñar a leer y escribir en el aula universitaria una experiencia en la facultad de psicología de la Universidad de Bueno Aires. Minutes from the National Conference on Institutional Policies for Development of Reading and Writing in Higher Education, Bogotá, Colombia, 2007. Available at www.ascun.org.co. Narváez, E., & Cadena S. (Eds.). (2008). Los desafíos de la lectura y la escritura en la educación superior: Caminos posibles. Cali: Universidad Autónoma de Occidente. Rincón, G. et al. (2005). Enseñar a comprender textos en la Universidad. Cali: Universidad del Valle. Van Dijk, T. (1998). Estructuras y funciones del discurso. Mexico: Siglo XXI editores. Viñao, A. (2002). La enseñanza de la lectura y la escritura: análisis socio-histórico. Presentation at the First National Congress on Written Language, organized by the Center of Professors and Recursos Murcia II, June 3–6, 1999.
Part III
Research on primary and secondary school practice
11 Young children revising their own texts in school settings Mirta Castedo and Emilia Ferreiro Centro de Investigación y de Estudios Avanzados del Instituto Politécnico Nacional
Introduction Revising is a frequent and essential cultural practice. Through revision, it is possible to adjust linguistic register to the communicative situations, change expressions to suit purpose and target readers, solve problems due to transfer from speech to written language, avoid unwanted ambiguities and repetitions, alter the illocutionary force of a statement, and so on. Revision “involves detecting mismatches between intended and instantiated text, deciding what could or should be changed in the text, knowing how to make desired changes, and operating, that is, making the desired changes” (Fitzgerald, 1987, quoted by Allal & Chanquoy, 2004, p. 1). In order to make such modifications, it is necessary to reread, compare the preliminary text with already published texts, consult or ask others to read your writing, and/or turn to support texts (for example, dictionaries or grammar books). Many authors have criticized traditional school practices (Petitjean, 1998; Allal, 2004), but they also claim that when the practices of writing at the school diversify and are carried out with a specific purpose and in real communicative situations, children may be encouraged to go over their own texts and to help their classmates, also revising others’ writings. This practice generates spaces for reflection on the language and is part of the writers’ current responsibilities. Since the pioneering research of Hayes and Flower (1980) and of Bereiter and Scardamaglia (1987), a progressive consensus has been reached on the need to develop teaching sequences where the different moments of the writing process take a specific teaching time and a specific design of situations (Allal & Chanquoy, 2004; Camps, 1989, 1996, 2003; Dolz & Schneuwly, 1998; Garcia-Debanc, 1983; Ribas, 2001; Schneuwly & Bain, 1998; Teberosky, 1992; Tolchinsky, 1993). In children, revision is not spontaneous: “the revision process seems difficult and it is rarely carried out by beginner writers. It implies a high level of developed writing and reading operations” (Negro & Chanquoy, 2005, p. 108). Is revision impossible before children develop such operations? No doubt it is a difficult task, but perhaps we must consider it from a developmental perspective. We will report here that, under certain conditions,
136 M. Castedo and E. Ferreiro young schoolchildren can revise their texts and learn about written language through the revision process. Considering that revision varies according to genre, we chose a specific type of text to be produced, revised, and then edited: captions for an album of family photographs. Captions are short informative texts, related to an image. They are present in different social contexts and can be found not only in textbooks, encyclopedias, magazines, and newspapers, but also in photographs, paintings, and museum exhibitions. In general, these texts need to reach a wide audience. They inform, present, or explain characteristics of the objects or events related to the picture, illustration, or image they are referring to. The caption does not need to inform what the image is showing; it must expand, contextualize, explain, or specify. In turn, the image provides elements that words could hardly express.1 While the photograph represents a piece of reality, the caption limits the possibilities of interpretation. When an event is identified, a date is specified or a proper name is given to an individual or place, some of the numerous interpretations are no longer possible. In this sense, illustrations based on what is written and captions play opposite roles. The illustration sets the meaning of the writing, whereas the caption sets the meaning of the image.
The research2 We designed a sequence of teaching activities to observe the revision process of short texts, produced by seven- and nine-year-old children. The final product of the sequence of activities was the collective production of a family photo album, where each photograph had its corresponding caption. The sequence consisted of 11 situations developed by the teachers during eight weeks, with four groups of students from two different public schools (two groups of second year and two groups of fourth year) in the city of La Plata, Argentina. Each child brought four family photographs for the album: a current picture of the family with the student; a current picture of the family (without the student); an old picture of the family before the student was born; a picture in which the author was a baby. The teaching sequence included the following situations: (a) agreement of the group on the purpose of the task, the audience, and the product to be obtained; (b) analysis and reading of newspaper and magazine captions; (c) individual production of captions for the four photographs (specific for each child); (d) collective revision of some captions produced for the first photograph; (e) revision of those same captions in small groups; (f) similar situations of revision for each of the remaining photographs; (g) finally, the album make up. Such a complex teaching approach was previously studied by one of us with several genres (Castedo, 1989, 1995; Castedo & Bello, 1999; Castedo & Waingort, 2003).
Young children revising their own texts 137 Throughout the collective revisions, emphasis was placed on the following restriction: the album would be passed around the different families and, therefore, the references to the people in the photographs should avoid the grammatical first person (an external reader would not know which was the reference for “my father” or for “I” if all the students used such expressions). Therefore, it is a peculiar situation in which the children must talk about themselves in the third person. Some children reformulated the teacher’s instruction using the following terms: “you have to write about yourself as if you were someone else” (“tenés que escribir de vos como si sos otro”). We chose captions for family pictures, because the caption is a short text that can be revised in depth even by very young children. These captions allow each student to write about events she/he has experienced, people whom she/he is related to, places she/he has been to, and what’s more important, about him/herself. The content to be communicated is familiar and of great proximity to the student. But even so, the selection of the linguistic means to express that content demands a great effort to decentralize, because each child would have to write “for others” about him/herself. That writing was targeted at a reader who was neither involved in the picture nor had a relationship with the photographed people. By sharing the album among the children’s families, the readers would not only be the students who participated in the writing process, but also their relatives who were unaware of the comments and conversations that took place during the process. This situation determined and justified a big part of the decisions taken when writing. We obtained 552 written productions (276 first versions and an identical number of revised versions; 262 produced by the seven-year-olds and 290 by the nine-year-olds). Besides, 40 observations of revision classes were carried out. To facilitate comparisons, all the examples correspond to the first photograph (i.e., a recent photograph of the child with some family members). However, the quantitative results presented in the figures include the data from the four photographs. We are only considering “deferred revision, which takes place once a relatively complete draft . . . has been written” (Allal & Chanquoy, 2004, p. 2).
Results The results will be presented, taking into account four questions: 1 Did the children refer to themselves and to the family members using the third person? (Results show a decisive influence of the revision process.) 2 Can we find certain caption frames according to age and to the photographs? (Results show a typology that was not influenced by the revision process.)
138 M. Castedo and E. Ferreiro 3 What remarks were most frequently made by children to their classmates during the revision process in small groups? 4 What is the impact of revising upon punctuation marks? 1 The use of the first or third person During a face-to-face discursive interaction, the reference to “I” does not bring about problems, because “I” is the one who speaks. In written texts, the signature provides a reference to that “I,” but in the context of these captions the decision to ask the children to write in the third person is justified by the fact that the readers, who cannot identify the individuals in the photographs, have to be taken into account. The enunciator must stand back from the character he is; in a way, he must split into two. For these reasons, teachers first worked with the children to explicitly define the position that the enunciator would take according to the communication circumstance. In spite of this, many captions—especially the first versions of the seven-year-olds—were written using the first person, as shown in Figure 11.1. To categorize the texts according to the grammatical person used, we considered the personal pronouns, possessive adjectives, and verbal inflections as well as the use of the author’s proper name. In Spanish the use of personal pronouns is not obligatory, as the verb includes specific markers for each of the grammatical persons. It is evident that both the students’ age and the practice of text revision are strongly associated with the possibility of solving the problem presented. Figure 11.1 shows that almost all the initial versions of the seven- 100 90 80 70 60 50 40 30 20 10 0
88
85
53
Initial Revised
52 44
24
23 5
First person
Oscillations
7 Third person
7-year-old (N � 262)
6 First person
4
9
Oscillations
Third person
9-year-old (N � 290)
Figure 11.1 Use of the first or third grammatical person in initial and revised versions. Note Values correspond to percentages calculated on the total of initial and revised versions for each age group and considering the four photographs (552 texts).
Young children revising their own texts 139 year-old children are in the first person (88 percent). At age nine, less than half of the children (44 percent) produced versions in the first person at the beginning. Even if the initial percentage at age nine is reduced by half in comparison with the other age group, it is still sufficiently high to indicate that the task is not easy for the nine-year-olds either. The revision process favors the type of reflection necessary to comply with the task requirements. In both age groups, the number of texts in the third person increases as they are revised: at the age of seven, half of the texts (53 percent) are written in the third person; at age nine, this percentage rises considerably (85 percent). In other words, the seven-year-old children manage, through revision, to reach the percentage that the nine- year-olds achieve before the revision. Let us see some examples. In all the cases, we will use Vi and Vr to indicate initial version and revised version respectively:3 Example of a text in the third person: El papá de Matias es el que tiene la remera verde que se llama Oscar, al lado del papá está Matias, la que está al lado de Matias es la mamá, que se llama Marta, están en Brasil. [Matias’ dad is the one wearing a green t-shirt who is called Oscar, next to the dad is Matias, the one who is next to Matias is the mom, who is called Marta, (they) are in Brazil.] (Matías, seven, Vr) Example of a text in first person: Mi familia ellos son mi papá mi mamá y mis dos hermanos [My family they are my dad my mom and my two brothers] (Nicolás, seven, Vi) Some texts present oscillations between the first and third person. The same child, Nicolás, succeeds in transforming his initial version into the third person. However, during the same revision process he decided to add information (name of the place and activities people are performing). The text added to the caption is in the first person plural (same verbal desinence in the three verbs). La familia ellos son el papá la mamá y los dos hermanos de Nico. Estamos en el zoológico de Buenos Aires. Vimos a los lobos marinos, a los patos y jugamos en unos juegos. [The family they are the dad and the mom and the two brothers of Nico. We are at Buenos Aires zoo. We saw the sea dogs, the ducks and we played on the swings.] (Nicolás, seven, Vr)
140 M. Castedo and E. Ferreiro Texts with oscillations grow through revision in seven-year-old children, showing how difficult is for them to keep the task’s requirement.4 2 The captions frames used by children Starting with the photographed scene, all students had to face the problem of solving which elements to choose and how to set them in order of priority. Children made their own selection and arranged the information, creating descriptions of the photographed scene following some basic standards: who are the people in the picture, what type of gathering event is it, where are they and, if possible, some time indication. All these details represent the relevant information. The way of organizing this information in a coherent text constitutes the main problem. The caption frames are listed below (2.1 and 2.2). 2.1 Integrated relevant information These texts inform about the most important features of the photographed scene in a single syntactic structure. In the most elaborate cases, they provide information about what has happened, when, and where. In comparative terms, these are the briefest texts of the corpus. The most elaborate texts compress a lot of information in a few words; the least elaborate provide limited information. Examples of both extremes are the following: Gerardo y su familia están en Argentina festejando el cumple de Gerardo el día 19 de marzo de 1998. [Gerardo and his family are in Argentina celebrating Gerardo’s birthday on March 19, 1998.] (Gerardo, nine, Vi) Estoy en la puerta de mi abuela [I am at my grandmother’s door] (Manuel, seven, Vi) 2.2 Lists of people Basically, these are lists of constructions that name the people in the photograph. They are the most frequent type of list used in our corpus. The lists show a tendency to repetitions of identical or similar constructions that are juxtaposed, linked by conjunctions, or separated by means of punctuation. We distinguish two types of lists of people: simple and complex. In the simple lists, people who were photographed are identified with their name and relationship. An example of minimum list, in the first person:
Young children revising their own texts 141 Estoy con mi hermana y mi papá y mi mamá [I’m with my sister and my dad and my mom] (Mario, seven, Vi) In many cases, these lists have either initial anchorages or closures, typical constructions of the texts that describe the event, place, and time, or they present collective nouns that help to identify the named individuals. When these constructions appear at the beginning of the list, they name the scene, acting as a title does, i.e., guiding the reader’s interpretation. (An example is Nicolás, already cited, whose Vi starts with “My family,” transformed into “The family” in his Vr.) Complex lists of people go beyond the identification of name and/or relationship. The variations attested in our corpus are: biographical lists, lists of location, and combined lists. 2. 2 . 1 B iographical L ists
These are texts that present biographical information of the identified people. Each of the individuals can be associated with his/her name, occupation, place of residence, date or place of birth, or with his/her age. Since they are lists of people, these biographical lists can also present initial anchorages or closures. In some biographical lists, a list of people is displayed, including some special features about them. Then this same list is reopened and some other features are included, until there is no information left. In others, first there is a description of an individual, giving all possible personal information and then another individual is described, as in Lucia’s example: Mi abuela se llama Ernestina no trabaja vive en Luján y nació en La Plata. Yo me llamo Lucía estoy en la Anexa vivo en City Bell y nací en Neuquén. Mi papá se llama Eduardo se está por recibir de profesor de computadora vive en City Bell y nació en La Plata. Mi abuelo se llama Valentín hace tapices vive en Morón y nació. Mi otra abuela se llama Marta es veterinaria vive en Morón y nació. [My grandmother is called Ernestina she doesn’t work she lives in Luján and was born in La Plata. I am called Lucía I’m in Anexa school I live in City Bell and I was born in Neuquén. My dad is called Eduardo he will soon graduate as a computer teacher he lives in City Bell and was born in La Plata. My grandfather is called Valentín he makes tapestry he lives in Morón and was born. My other grandmother is called Marta she is a veterinarian she lives in Morón and was born.] (Lucía, seven, Vi)
142 M. Castedo and E. Ferreiro When Lucía does not know the place of birth of someone in her family, she just writes “was born” with a change of line after that verb, as if Lucía was aware that some information is missing and left the space to complete it at the appropriate time. 2. 2 . 2 L ists of location
These are texts organized in such a way that the reader can identify who is who in the picture; therefore, the text becomes very dependent on the image. They can also present initial anchorages or closures. Ana’s text is an example of this last case, with a closure that identifies the event: “They are at Ana’s baptism.” La que está a la derecha es Beatriz. El que tiene camisa a cuadritos es Franco. La que tiene vestido a cuadritos es Inés. El que está con corbata es el padrino de Ana. El que tiene pantalón celeste. Tania la prima de Ana está al medio de Inés, la de rulitos. La bebé es Ana. El que tiene camisa blanca es el hermano de Ana. La mamá de Ana tiene a upa a Ana. Están en el bautismo de Ana. [The one on the right is Beatriz. The one with a checked shirt is Franco. The one with a checked dress is Inés. The one with tie is Ana’s godfather. The one who is wearing light blue trousers. Tania, Ana’s cousin, is next to Inés, the one with curly hair. The baby is Ana. The one with a white shirt is Ana’s brother. Ana’s mom is holding Ana in her arms. They are at Ana’s baptism.] (Ana, seven, Vr) 2. 2 . 3 Combined lists
These texts constitute combinations of the types of lists described so far. The next example is Micaela’s caption. She opens a biographical list, making reference to the birth and relationship with her brother. Then, she mentions the place where the event was photographed (“they are at their grandmother’s”). She reopens a location list and, finally, she starts again with a biographical list, focused on the age of three of the people. Nicolás nació el 28 de junio Nicolás es el hermano de Micaela están en lo de su abuela Nicolás es el que tiene una remera rayada con rayas azules y amarillas
Young children revising their own texts 143 el papá tiene gorra la mamá es la que tiene a upa a Micaela y ella está a upa de su madre el hermano tiene 8 años la mamá 36 y el papá 36 [Nicolás was born on June 28 Nicolás is Micaela’s brother they are at their grandmother’s Nicolás is the one with a blue and yellow striped t-shirt the dad is wearing a cap the mom is the one holding Micaela in her arms and she is in her mother’s arms the brother is 8 the mom 36 and the dad 36] (Micaela, seven, Vr) 2.3 Distribution of caption frames between Vi and Vr Figure 11.2 shows that the revision process has no effect on the caption frames. Once a frame has been chosen, it remains stable, even if information is added or modified. The analysis of each case shows that only in a few of them does the revision affect the initial frame. Figure 11.2 shows that seven-year-old children’s preferences are distributed among the two main frames: integrated information and lists, while the nine-year-olds show a preference for the first frame. Only 4 percent of the texts produced by the younger children cannot be classified into any of these two main frames. The category “others” was no longer necessary with the older children.
80 70
66
60 50
49
43
47
Initial Revised
70
53
40
34
30
30
20 10 0
4 4 Integrated information
Lists
Others
7-year-old (N � 262)
0 0 Integrated information
Lists
Others
9-year-old (N � 290)
Figure 11.2 Distribution of caption’s frames in initial and revised versions. Note Values correspond to percentages calculated on the total of initial and revised versions for each age group and considering the four photographs (552 texts).
144 M. Castedo and E. Ferreiro 3 Children’s interventions during the revision process in small groups The analysis of children’s interactions in small groups shows that children pay attention to the systematic use of the third person, which was consistent with the teacher’s interventions. They also pay attention to punctuation marks. However, the children also paid attention to other aspects that were not the focus of the teacher’s interventions, avoiding, for example, lexical repetitions. In the following example we will explore these three aspects in detail. Natalia is seven. In her first version, she writes in the first person. During revision she manages to change everything into the third person, without modifying substantially the caption frame. The first version is a list of people with initial anchorage on the event. The revised caption is also a list of people, but with two anchorages, for opening and closure. Aquí estoy en mi cumpleaños con mi mamá Amelia y mi papá Gustavo y mi hermana Graciela (Vi) [Here I am on my birthday with my mom Amelia and my dad Gustavo and my sister Graciela] (Vi) Aquí está Natalia en su cumpleaños de 7 años con la mamá Amelia, papá Gustavo y Graciela La familia de Natalia. (Vr) [Here is Natalia on her seventh birthday with the mom Amelia, dad Gustavo and Graciela Natalia’s family.] (Vr) How did she go from one version to the other? Let’s analyze some snatches of the interactions held in the group where Natalia and her classmates, Ignacio and Julieta, participate. Natalia rereads her text and comments: Natalia: Well,
I think it is wrong to write “my, my” it is wrong “my, my.” Miss, can I write “here”? Teacher: Yes. It is possible, but it depends. We have to check how it sounds. Natalia: I can say . . . “Here on my birthday” Ignacio: No! No, because it is not on “my birthday,” that is not my birthday. Natalia: “Here is Natalia on her/celebrating her birthday”
Young children revising their own texts 145 The first observation of Natalia refers to the repetition of the three possessive adjectives “my.” This observation remains unsolved, because Natalia herself immediately focuses her attention on the first word of the text and turns to the teacher making an explicit question about the deictic “here.” Ignacio strongly objects to the use of the first person possessive adjective, arguing that it refers to the person who is making the statement (“this is not my birthday,” that is to say it is not Ignacio’s birthday). Natalia quickly understands the objection and restates. Natalia starts writing: Teacher: Are
you going to write everything again, Natalia? Aren’t you going to erase and correct it there? Natalia: No, I have not erased it yet. [She refers to her Vi, which she leaves without modifications.] [While Natalia writes, Ignacio and Julieta talk about TV programs.] Teacher: It sounds like a lot of fun, but Natalia is working alone . . . Ignacio: She didn’t ask for help Natalia: You should know I need help Julieta: [Continues singing a TV song] Natalia: OK, I’ll work on my own and that’s it! [She writes and then reads aloud “Aquí está Natalia en su cumpleaños de 7 años”] Julieta: Done, have you finished? Julieta, who is not willing to get involved in the writing of her classmate, considers the task finished. Indeed, the caption could end that way, with a closure that identifies the event. However, Ignacio takes part and, probably based on the previous caption (Vi that remains visible), suggests adding the names of the people in the picture. Ignacio: Your
dad is called Gustavo, and your mom? [Pause] And my sister Graciela, and I Natalia. [Pause] I don’t know what to write to introduce my mom “who is called Amelia,” my sister, “who is called Graciela” [with a special tone stating the written text and emphasizing “who is called”] Ignacio: That is a rhyme . . . what you said. Julieta: It doesn’t matter, don’t put it, some didn’t include it [referring to some of her classmates]. Teacher: Natalia said she didn’t know how to introduce her dad who is called like this, her mom who is called like that, how can she do it? Julieta: Then, she can put “I am with my dad who is called like this, I am there, my mom is there and my sister is there” Natalia: Can I write “Natalia is with her dad, with her mom, with her dad who is called Gustavo, with her mom who is called Amelia, with her sister who is called Graciela”? Ignacio: I see! She is thinking what to write. [Pause] Natalia: Amelia.
146 M. Castedo and E. Ferreiro Once again, Natalia faces the problem of repetitions. She doesn’t want to repeat three times the construction “who is called.” Julieta keeps affirming that the caption is ready and that it is not necessary to add anything else, but the teacher goes back to Natalia’s concern and tries to get the group involved. So Julieta makes the first complete suggestion, but in the first person. Natalia keeps suggesting a text orally, including the names but making the same repetitions she was trying to avoid. She starts writing and completes the caption. Natalia: Done! Teacher: Let’s see.
[She reads aloud] “Aquí está Natalia en su cumpleaños de 7 años con la mamá Amelia, papá Gustavo y Graciela La familia de Natalia” [Pause] Ignacio: “The mom Amelia/the dad Gustavo” and comma, “Gustavo, comma, comma” Natalia: Where? Ignacio: “Gustavo” Natalia: Ugh, I don’t understand anything! Julieta: Have you finished? Ignacio: Did you write “mamá” with accent? “Mamá” goes with an accent.5 Natalia: Oh, you are right! Julieta: I think you have finished. Natalia: I have to write a full stop at the end. When writing the text, Natalia found a solution in order to avoid the repetitions that were so annoying. This solution wasn’t previously verbalized. Ignacio helps with observations regarding punctuation and graphic stresses, while Julieta is focused only on trying to end the revision process as soon as possible. In this brief snatch, we can observe how these three recurrent topics appear in the interaction among a small group: using the third person (initial observation of Ignacio, accepted by Natalia, and ignored by Julieta); avoiding repetitions (initial concern of Natalia, who solves the problem with no help from her classmates); taking care of punctuation (introduced by Ignacio, and accepted by Natalia, who decides to add a final full stop without discussing it). 4 About punctuation marks During the revision process, children pay attention to punctuation marks. Despite the briefness of the text, the comma used in lists is mostly included in the texts based on that frame, and when a lot of information is introduced, the concern about the necessity of indicating limits between such information arises. We know that in narrative texts, punctuation tends to progress from the
Young children revising their own texts 147 external limits of the text inwards (Ferreiro, 1991, 1996; Ferreiro & Pontecorvo, 1999; Vieira-Rocha, 1995). That is the reason why we have classified the texts into the three categories shown in Figure 11.3. Figure 11.3 shows that both age and the practice of revision are strongly related to the use of punctuation marks. The texts with no punctuation marks are more frequent in seven-year-old children (in the Vi and Vr) than in nine-year-olds, while texts with internal punctuation are more frequent in nine-year-olds (both Vi and Vr). Besides, texts with no punctuation decrease between the Vi and the Vr, both at seven and at nine, i.e., revision increases the use of punctuation in both age groups, regardless of the caption’s frame. Nevertheless, it is important to consider that a caption organized as a list usually requires internal punctuation, while a brief caption, organized as integrated relevant information, can be solved without internal punctuation. The comma is the punctuation mark that shows the highest occurrences; based on a frequency order, the full stops come next. This is similar to what has been reported as use of punctuation in narratives (Ferreiro, 1996). However, it is interesting to observe that in captions we can find punctuation marks not very frequently used in a narrative: parenthesis and quotation marks, particularly in the nine-year-olds’ texts. Parentheses were used to specify places of the event, position of a character within the picture, ages, relationships, and times. For example: Acá estoy en Puerto Madero (Buenos Aires) Con mi mamá Susana, mi hermana Noemí y yo. [Here I’m in Puerto Madero (Buenos Aires) With my mom Susana, my sister Noemí and me.] (Marcela, nine, Vi) 80 70
76 67
60
Initial Revised
56
53
50 40
35
30
21
20
12 12
10 0
No punctuation
Only final full stop
23
21 14
10 Internal No punctuation punctuation
7-year-old (N � 262)
Only final full stop
Internal punctuation
9-year-old (N � 290)
Figure 11.3 Use of punctuation marks in the texts. Note Values correspond to percentages calculated on the total of initial and revised versions for each age group and considering the four photographs (552 texts).
148 M. Castedo and E. Ferreiro Acá está la madrina de Micaela que está al lado de la señora (madre de Micaela) . . . [Here is Micaela’s godmother next to the lady (Micaela’s mom) . . .6 (Micaela, nine, Vr) Quotation marks were used with names of places or people, as in these examples: Acá están en el “Restaurante Don Quijote.” Está mi papá, mi mamá y mi abuelo. Están leyendo la carta [Here they are at “Don Quijote Restaurant.” My dad, my mom and my grandfather. They are reading the menú] (Julieta, nine, Vi) Aquí esta la familia de Matías en la comunión de su hija menor: “Jimena.” [Here is Matías’ family in the holy communion of their younger daughter: “Jimena.”] (Matias, nine, Vr)
Concluding remarks The data we analyzed here shows that, under specific teaching conditions, young children are able to adopt a “revising position” toward their own texts as well as their classmates’. We believe that it is important to promote the need to revise every draft produced from the beginning of elementary school. This is not an imposition but the result of two preconditions. First, the strong belief that written language in school settings should be modeled by social practices. When the teacher is the only person in the classroom with the power to revise (and to make normative corrections), children are prevented from becoming autonomous writers. In addition, they are kept away from the metalinguistic activity that consists of “rephrasing” in order to adjust the intended meaning to specific requirements of the communicative situation. Second, to revise a draft is to promote individual responsibility toward a wide audience, as these written productions are usually embodied in long-term projects whose final product usually (not always) aims at people outside the school. The type of text chosen for this research is a peculiar descriptive one. But the main results are not limited by the text’s type: 1 Children as young as seven can be engaged in revision tasks. We must look at revision in developmental terms.
Young children revising their own texts 149 2 The teaching requirements—in this case, writing about themselves in the third person—are kept alive during the revision process in small groups, even if not all the children are able to succeed in the task. 3 Children introduce their own requirements (i.e., to avoid repetitions) that were not the focus of the teacher’s interventions. 4 Punctuation marks are typically introduced during revision, a fact that is along the lines of the history of punctuation in the Western tradition (Parkes, 1992). 5 The initial frame of the text resists children’s revision, at least in cases where teacher’s interventions do not promote a specific model to be followed. 6 The departure point of the two contrasting age groups is clearly different (see Figures 11.1 to 11.3), showing that a developmental factor interacts with the revision process and the teacher’s interventions.
Notes 1 The photographs can be used as “triggers” for various types of texts. For a recent revision, see Van Horn (2008) who, in fact, is more concerned about “creative writing” and does not deal with captions. 2 The data reported belongs to Mirta Castedo’s doctoral thesis, carried out in the DIE-CINVESTAV, Mexico, under the tutoring of Emilia Ferreiro. 3 The names of the family members and places were modified to preserve children’s identity. The children’s texts are presented with conventional spelling to facilitate translation. However, we kept the original punctuation marks and syntactic inflections. We did not add words, even in cases of evident omissions. The original graphical line was preserved only in cases where it seemed to be used as a tool to organize information. We refer to Ferreiro (2008) for a discussion about transcription norms. 4 Béguelin (2000, p. 269) reports similar changes of grammatical person in texts of French teenagers. So, it is an important problem that deserves attention. 5 Mamá is the Spanish word for Mom and is spelled with a graphic accent on the second a. 6 It follows a list of people.
References Allal, L. (2004). Integrated writing instruction and the development of revision skills. In L. Allal, L. Chanquoy, & P. Largy (Eds.), Studies in writing: Vol. 13, Revision. Cognitive and instructional processes (pp. 139–155). Boston, MA and Dordrecht: Kluwer. Allal, L., & Chanquoy, L. (2004). Revision revisited [Introduction]. In L. Allal, L. Chanquoy, & P. Largy (Eds.), Studies in writing: Vol. 13, Revision. Cognitive and instructional processes (pp. 1–7). Boston, MA and Dordrecht: Kluwer. Béguelin, M.-J. (2000). De la phrase aux énoncés: Grammaire scolaire et descriptions linguistiques. Brussels: De Boeck Duculot. Bereiter, C., & Scardamaglia, M. (1987). The psychology of written composition. Hillsdale, NJ: Lawrence Erlbaum. Camps, A. (1989). Modelos del proceso de redacción: Algunas implicancias para la enseñanza. Infancia y Aprendizaje, 49, 3–19.
150 M. Castedo and E. Ferreiro Camps, A. (1996). Proyectos de lengua entre la teoría y la práctica. Cultura y Educación, 2, 43–57. Camps, A. (2003). Miradas diversas a la enseñanza y el aprendizaje de la composición escrita. Revista Latinoamericana de Lectura. Lectura y Vida, 24(4), 14–23. Castedo, M. (1989). Construcción de un texto dramático. Revista Latinoamericana de Lectura. Lectura y Vida, 10(1), 14–23. Castedo, M. (1995).Construcción de lectores y escritores. Revista Latinoamericana de Lectura. Lectura y Vida, 16(3), 5–24. Castedo, M., & Bello, A. (1999). Escribir cosas que corresponden a la verdad o se asemejan a la verdad. Cuadernos de Investigación Educativa, 1(4), 7–31. Castedo, M., & Waingort, C. (2003). Escribir, revisar y reescribir cuentos repetitivos. Revista Latinoamericana de Lectura. Lectura y Vida, 24(1), 31–35. Dolz, J., & Schneuwly, B. (1998). Pour un enseignement de l’oral. Paris: ESF. Ferreiro, E. (1991). L’uso della punteggiatura nella scrittura di storie di bambini di seconda e terza elementare. In M. Orsolini & C. Pontecorvo (Eds.), La costruzione del testo scritto nei bambini (pp. 233–257). Firenze: La Nuova Italia Editrice. Ferreiro, E. (1996). Los límites del discurso: puntuación y organización textual. In E. Ferreiro, C. Pontecorvo, N. Ribeiro-Moreira, & I. García-Hidalgo (Eds.), Caperucita Roja aprende a escribir. Estudios psicolingüísticos comparativos en tres lenguas (pp. 129–161). Barcelona: Gedisa. Ferreiro, E. (2008). Criterios para la transcripción de los textos [Anexo]. In E. Ferreiro & A. Siro (Eds.), Narrar por escrito desde un personaje (pp. 233–241). Buenos Aires: Fondo de Cultura Económica. Ferreiro, E., & Pontecorvo, C. (1999). Managing the written text: The beginning of punctuation in children’s writing. Learning and Instruction, 9, 543–564. Garcia-Debanc, C. (1986). Intérêts des modèles du processus rédactionnel pour une pédagogie de l’écriture. Pratiques, 49, 23–49. Hayes, J., & Flower, L. (1980). Identifying the organization of writing processes. In C. M. Levy & R. Steinberg (Eds.), Cognitive processes in writing: An interdisciplinary approach (pp. 3–30). Hillsdale, NJ: Lawrence Erlbaum. Negro, I., & Chanquoy, L. (2005). The effect of psycholinguistic research on the teaching of writing. Educational Studies in Language and Literature, 5(2), 105–111. Parkes, M. B. (1992). Pause and effect. An introduction to the history of punctuation in the West. Aldershot: Scolar Press. Petitjean, A. (1998). Enseignement/apprentissage de l’ecriture et transposition didactique. Pratiques, 97–98, 105–132. Ribas, T. (2001). La regulación del proceso de composición escrita en grupo: Análisis de la utilización de pautas de revisión. In A. Camps (Ed.), El aula como espacio de investigación y reflexión (pp. 51–68). Barcelona: Grao. Schneuwly, B., & Bain, D. (1998). Mecanismos de regulación de las actividades textuales. Estrategias de intervención en las secuencias didácticas. Textos. Didáctica de la lengua y la literatura, 16, 25–48. Teberosky, A. (1992). Aprendiendo a escribir. Barcelona: ICE-Universidad de Barcelona. Tolchinsky, L. (1993). Aprendizaje del lenguaje escrito. Barcelona: Anthropos. Van Horn, L. (2008) Reading photographs to write with meaning and purpose. Grades 4–12. Newark, DE: International Reading Association. Vieira-Rocha, I. (1995). Adquisición de la puntuación: usos y saberes de los niños en la escritura de narraciones. Revista Latinoamericana de Lectura. Lectura y Vida, 16(4), 41–46.
12 Written representations of nominal morphology by Chinese and Moroccan children learning a Romance language Liliana Tolchinsky and Naymé Salas University of Barcelona, Spain
Second-language learners in a literate community are exposed to the spoken and written varieties of the language(s) of the environment and they must necessarily cope with the two modalities simultaneously. Studies on first language (L1) acquisition and on the acquisition of first-learned writing systems (WS1s) have shown that children are attentive to domain-specific information relevant to language (Karmiloff Smith, 1992), they are sensitive to the formal and functional features of the WS1, and they build different assumptions as to the way in which the WS1 represents the L1 (Ferreiro, 2002; Tolchinsky, 2003). The way in which young second-language (L2) learners make sense/use of the spoken and written modalities of the L2 have been far less explored. Moreover, and but for a few exceptions (e.g., Durgunoglu, Mir, & Ariño-Marti, 2002), research on learning processes in L2/ WS2 has been mostly conducted on adult EFL learners. The present study focuses on young (5–8 years old) L2 learners’ sensitivity to Catalan morphology; that is, to the internal structure of words in Catalan. Specifically, we assess L2 learners’ sensitivity to number inflection—the way in which the alternation singular/plural is expressed in the language they are learning. We were particularly interested in studying this process within constructions—Nominal Groups (NGs)—formed with a noun and an indefinite article. In English, for example, the difference between a car (indefinite, singular NG) and cars (indefinite, plural NG) is expressed by the addition of -s after the noun, and the deletion of the indefinite article a. In Catalan this contrast is expressed by a change both in the noun and in the indefinite article; thus, un cotxe “a car” (indefinite, singular NG) turns into uns cotxes “a-pl cars” (indefinite, plural NG). It is clear, then, that English and Catalan mark number inflection (singular vs. plural NGs) in different ways. Moreover, we also assess L2 learners’ sensitivity to locative-derivatives formation. This is another kind of morphological process, which affects the way words are formed out of other words in a language. Specifically, we were interested in the formation of nouns that refer to the place where something can be bought—locative derivatives—out of nouns that refer to a single object—base nouns. In English, for example, we form the locative-
152 L. Tolchinsky and N. Salas derivative noun shoe store out of the base noun shoe; that is, we add the word store to signal the place where shoes can be bought. In Catalan the same contrast is expressed by adding suffix -eria to the base noun; thus, the base noun sabata “shoe” turns into the locative derivative sabateria “shoe-store.” Catalan and English, therefore, have different ways of forming locative derivatives. Both processes are explored in the spoken and written representations produced by Moroccan and Chinese children who have recently settled in Catalonia. Central to our research interests was the typological distance between the languages involved in the study, as well as the inclusion of different writing systems, since we aimed at exploring possible L1 and WS1 influences in the process of acquiring and becoming literate in Catalan as an L2. The nature of the study requires taking into account both Second Language Acquisition (SLA) theories and children’s developing assumptions on the relationship between writing systems and oral language. Explanations of the role of the L1 in SLA range from those arguing that it has no substantial effects on the initial phases of an L2 learning process (e.g., Perdue, 1993), to those claiming that the mistakes and difficulties found in learners’ interlanguage will differ as a function of their L1 (Liceras, 1997; Tsimpli & Roussou, 1991; Smith & Tsimpli, 1995). Additionally, it has been suggested that experience with a richly inflected language enhances learners’ sensitivity to L2 morphology (Dressler, 2007). On the other hand, children’s early assumptions on the links between spoken and written language as forged in a diglossic context may affect the learning process of a WS2 that represents the L2. In the following we describe some of the features of the languages involved in the study—Chinese, Moroccan Arabic, and Catalan. The descriptions are restricted to those features directly related to the study: number inflection and definiteness, and locative derivation. Number inflection and locative derivation in Chinese, Moroccan Arabic, and Catalan Participants’ L1s strongly differ in their degree of morphological richness— the most important typological property of a language (Dressler, 2004). The level of morphological richness refers to the number of functions that are expressed through morphological means. For example, languages that mark the contrast singular/plural through word-internal elements (affixes) are morphologically richer than those that express such a contrast through other means. The Chinese participants were speakers of Qing Tian Hua which, as any other Chinese language, has a very poor morphology—almost every word consists of a single morpheme. Number distinctions are expressed through different particles—such as cardinals, quantifiers, and a group of classifiers—while definiteness is mainly discursively established.
Written representations of nominal morphology 153 The Moroccan participants were speakers of Darija which, like Catalan, has a much richer morphology. Number changes are expressed solely in the noun, using inflectional affixes that are added to the singular form of the noun. Determiners are not inflected for number, so there is no number agreement in the nominal system. Indefiniteness, on the other hand, is achieved through the use of an indefinite article. Catalan, the target language, is a Romance language that reflects number changes through inflectional morphology. In Catalan, number assumes singular and plural values, and number changes are marked by means of suffix /-s/ and its allophonic variants, affecting both the noun and the determiner; number agreement is thus compulsory. Indefiniteness is achieved through the use of the indefinite determiner un (“a”), which inflects for number and gender. As regards locative-derivatives formation, in Qing Tian Hua they are N+N compounds, whereas in Darija they are currently formed through paraphrases of the base form (e.g., muta’a d’sabato: “place for shoe”). Although there remains some locative Ns inherited from Classical Arabic— which followed typical Semitic processes of derivatives formation, they are no longer productive in Darija. As mentioned earlier, in Catalan locative nouns are obtained by adding suffix /-ə′riə/ to the base form of the noun. Chinese and Moroccan children’s earlier experience with spoken and written language differs to a large extent from the one they undergo learning Catalan. In their home countries, oral language is not written and, conversely, written language is not normally spoken. Becoming literate in China or in Morocco implies, therefore, gaining command of a different language: Mandarin in China; Standard Modern Arabic (SMA) in Morocco. In Catalonia there is not such diglossia: becoming literate means having access to the written modality of the spoken language. We aim at determining the extent to which subjects’ L1 influences their sensitivity to number marking—within indefinite-article NGs—and to locative-derivatives formation in Catalan. Moreover, we intend to establish children’s degree of awareness of the fact that in the target language changes in speech—such as a change from the singular to the plural or a change from a base N to its corresponding locative—should be reflected in writing. Provided that our younger participants—five to six years old—had not had any previous literacy instruction in their home countries, developmental factors are also contemplated. Assuming that L1 characteristics largely shape the process of acquiring an L2, we would expect our Moroccan participants to outperform Chinese children in their accuracy to mark number changes, since Darija and Catalan are both highly inflected languages. Moreover, in spite of the lack of information on the acquisition of L2 derivational morphology, we predict that both language groups will preferably reflect changes from a base N to a locative (and vice versa) through means other than the addition of a suffix. In addition, such means might differ as a function of subjects’ L1. Finally, we anticipate that both Chinese and Moroccan children
154 L. Tolchinsky and N. Salas will frequently fail to show in their written productions changes made to speech utterances.
Method Participants Eighty-four children participate in the study: 28 Chinese children, mostly from Zhejiang—a province south to Shanghai, speakers of Qing Tian Hua; 26 Moroccan children mostly from North Morocco, speakers of Darija— i.e., Moroccan Arabic; and 30 controls—native speakers of Catalan. All three language groups were divided into two age groups: 5–6-year-olds and 7–8-year-olds. The final number of children within each subgroup and mean ages are shown in Table 12.1. Control subjects have been raised in Barcelona or neighboring suburban areas, which are bilingual (Spanish/Catalan). While all controls were born in Catalonia, none of the children in the experimental groups were. Newcomers have been attending Catalan public schools in the above-mentioned areas for 18 months, approximately. Tasks and materials There were four semi-structured tasks: two production tasks that required the child to produce spoken utterances and written representations and two comprehension tasks. The first two tasks relate to inflectional morphology, the two other to derivational morphology. The methodology of each of these tasks is described below with examples and some scoring criteria. Number-inflection task Children were presented with an initial picture with one object (singular condition) or many identical objects (plural condition), with a caption below (e.g.,
“a car”). They were required to say what was in the picture, without any indication on the part of the interviewer. Afterwards, regardless of the kind of description produced by the child, the interviewer said to him/her what was written on the caption (e.g., un cotxe: “a car”). Children were then presented with a target picture with
Table 12.1 Participants’ distribution and mean ages
Chinese Moroccan Catalan
5–6 years old
7–8 years old
Total N
n = 10 (5;6) n = 12 (5;3) n = 13 (5;3)
n = 18 (7;7) n = 14 (7;6) n = 17 (7;4)
28 26 30
Written representations of nominal morphology 155 the same object, but differing in number (plurality of objects, if the initial picture had been a singular presentation, or one object, if the initial picture had been a plural presentation). The target picture had an empty label below the depicted object. Children were then asked to name the target picture and, afterwards, to write a caption for it on the empty label. The initial picture with the written caption remained all the time in front of the child, so that he/she could use it at his/her convenience. The pictures shown to the children were clear photographic images that were chosen after making sure that they were easily identifiable. A total of eight nominal groups—four for each condition—were used: un cotxe (“a car”), una casa (“a house”), un quadre (“a picture”), una poma (“an apple”), un llibre (“a book”), una taula (“a table”), un arbre (“a tree”), una porta (“a door”), and their plural counterparts. They were concrete, countable nouns with a rather simple disyllabic structure—consonant clusters of a maximum of two elements—stressed in the penultimate syllable (the commonest stress pattern for Catalan), and they correspond to levels 1–2 of difficulty in basic Catalan vocabularies (e.g., Rafel i Fontanals, 1998). Derivation task This task had an identical structure to the inflection task: children were presented with an initial picture with an object, such as sabata (“shoe”) (base-noun condition) or the store where the object can be bought, i.e., sabateria (“shoe-locative”; “shoe-store”) (derivative condition) with a written caption below. Again, a total of eight words—four per condition— were used: llibre (“book”), pastís (“cake”), sabata (“shoe”), gelat (“ice cream”), fusta (“wood”), paper (“paper”), barret (“hat”), fruita (“fruit”). The locative derivatives of all these Ns are formed adding suffix /-ә′ria/. Just as with the number-inflection Ns, they all are concrete, disyllabic (only one trisyllabic N), with simple syllabic structure, and correspond to the lowest levels of difficulty in basic Catalan vocabularies (e.g., Rafel i Fontanals, 1998). Comprehension tasks After each one of the production tasks—i.e., inflection and derivation—a comprehension task was conducted. Children were presented with all the pictures used in the corresponding production task—i.e., all singular and plural presentations of the objects, in the inflection task; all base-noun and locative-derivative presentations, in the derivational task. The interviewer asked the child to identify the particular picture that corresponded to his/ her description (e.g., Aquí hi ha moltes fotos. Jo et diré el nom d’una cosa i tu me la dònes. “Here there are many pictures. I’ll say the name of one of them and you must hand it to me”). To prevent order effect, there were two orders of presentation of items within each task. For the number-inflection task, NGs that were initially
156 L. Tolchinsky and N. Salas presented in the singular in order A appeared initially presented in plural form in order B. Similarly, for the derivational task, Ns that were initially presented in a base form in order A appeared initially presented in derivative form in order B. Analysis For each task we first counted the number of target-like responses and, thereafter, we performed a qualitative analysis of deviant responses— except for the comprehension tasks in which only correct identification were considered. Target-like utterances/written productions in the number- inflection task were those that attended to number inflection, indefiniteness, and agreement, maintaining the indefinite article—i.e., un cotxe (“a car”) for a picture displaying a singular object or uns cotxes (“a-pl cars”), for a picture displaying a plurality of objects. For the derivation task only utterances/written productions of the base noun, plus suffix /–ә′ria/ for locative derivation—or without the suffix for the base-form condition—were counted as target-like responses (e.g., llibreria—“book store”—for the picture displaying a book store and, conversely, llibre—“book”—for the picture displaying the object). Finally, in the comprehension tasks, correct identifications of the NG uttered by the interviewer—i.e., handing in the card with a picture that matched both the item in the exact form in which it was uttered—were classified as correct answers.
Results In order to provide a general picture of children’s performance, we compared the mean scores of target-like responses they had obtained in the different tasks, by language group (Figure 12.1). The two tasks that assess children’s comprehension of the verbal expressions used for describing the pictures used for the number-inflection and the derivation tasks were the easiest ones. Both Chinese and Moroccan children were able to identify correctly more than half of the pictures, while Catalan controls were at ceiling. Second in order of difficulty were the two tasks requiring the children to verbally express changes in number inflection and in derivation of locatives. The most difficult tasks were those requiring a written representation. The two morphological domains and the three languages involved in the study showed the same pattern of difficulty. Spoken representation of number inflection Table 12.2 shows the mean score of target-like utterances for each condition of the number-inflection task, according to language and age group. Since each child was presented with four items per condition, the range is 0–4.
Written representations of nominal morphology 157 8 7 6 5 4 3 2 1 0
Chinese
Moroccan Inflection Sp Inflection Wr Inflection Com
Catalan Derivation Sp Derivation Wr Derivation Com
Figure 12.1 Comparison of performance in each task, by language group (range 0–8).
Almost half of the utterances produced by Chinese and Moroccan children reflected the change from plural to singular in the determiner and the noun, respecting number agreement within the NG. This behavior tended to increase with age. Utterances that did not reflect this change were more popular among the Chinese than among the Moroccan children, suggesting that the task itself was more difficult, in general, for the former. A completely different picture is obtained for the production of plural, indefinite-article NGs. Not a single utterance produced by the youngest Chinese children and very few of those produced by the Moroccans reflected the change from singular to plural in both determiner and noun, showing internal NG agreement. Most interestingly, for the Catalan controls also the plural condition was much more difficult than the singular one. Non-parametric tests for related samples revealed that there was a significant condition effect—Wilcoxon Signed-Rank, p < 0.02 for all languages and age groups. However, although controls clearly outperformed Chinese (U-Mann-Whitney, p < 0.05) and Moroccan children (U-MannWhitney, p < 0.001, only for the singular condition), both groups of newcomers did not differ significantly from each other. Qualitative analyses of non-target-like responses served as a means to observe the extent to which children’s resources to complete the number- inflection task differed according to their L1s. Verbatim repetition of the input was the most common response among the youngest Chinese to describe singular objects (48 percent), whereas for plural descriptions they also resorted to modifying the determiner without altering the noun (27.5 percent). Repetition decreases with age (25 percent) but modification of
1.50
0.00
Singular target
Plural target
0.00
1.50 0.33
1.83 0.97
1.20 0.17
1.75
Mean
0.38
1.21
(SD)
Mean
Mean
(SD)
5–6 years n = 12
7–8 years n = 18
5–6 years n = 10
(SD)
Moroccan
Chinese
Table 12.2 Mean score of target-like utterances describing the target picture
0.64
2.36
Mean
1.33
1.21
(SD)
7–8 years n = 14
0.23
3.92
Mean
0.83
0.27
(SD)
1.18
3.88
Mean
1.42
0.33
(SD)
5–6 years n = 13 7–8 years n = 17
Catalan
Written representations of nominal morphology 159 the determiner increases in plural descriptions (31.8 percent). Younger Moroccans also repeat the input utterance but to a lesser extent (25.9 percent) and, unlike the Chinese, who focused on the determiner, they tended to change only the N, with article omission (51.9 percent). These preferences did not change dramatically with age or condition. Moroccan children seem to rely more on N modification for reflecting number changes or, secondarily, on modification of both elements (N and article) for expressing plurality. Controls showed an interesting pattern—in the singular condition they were at ceiling, but for the plural condition more than 50 percent of utterances included a change in the determiner. Spoken representation of locative derivatives If children are already sensitive to the way in which Catalan forms locative nouns from common base nouns, they should add a suffix to the noun to form the locative and, conversely, delete a suffix in order to obtain the base noun. In Table 12.3 the mean score obtained the target-like utterances across words by language and condition. L1 effect—which was very subtle in the degree of success to perform the task—is more noticeable in the deviant strategies children employ. The youngest Chinese tended to repeat what they themselves or the interviewer had said to describe the initial picture when asked to describe the target one. Moroccan children, on the other hand, explored other ways of solving the situation, mainly resorting to paraphrases (e.g., for the target picture “shoe store,” they produced di per comprar sabatas (“say_3rd.sg. for selling shoes”) instead of using the corresponding suffix). Links between spoken modifications and written representations At preschool level, functional literacy experience is rather similar among the three groups but, naturally, when more language-specific reading abilities are required, the gap between Catalan students and newcomers increases (Tolchinsky, Salas, & Perera, in press). However, we expected important differences in the way Chinese and Moroccan children, as compared to Catalan controls, relate to the links between speech and writing. Therefore, to explore the possible influence of children’s WS1s in their understanding of the links between speech and writing in the target language, we assessed the extent to which participants’ modifications to speech utterances correlated with changes in written productions, in both number-inflection and derivational tasks. As Table 12.4 shows, modifications in spoken utterances and modifications in the written captions highly correlate to one another in all languages and conditions, both in the inflectional and in the derivational task. In other words, the more children manage to show changes of number and of derivation in speech, the more they reflect it in writing as well. Nevertheless, qualitative analyses revealed that many Chinese and Moroccan
Note Range: 0–4.
Base target Derivative target
0.80 0.10
1.03 0.31
2.06 1.06
1.69 1.39
1.33 0.50
Mean 1.43 1.24
(SD) 1.64 0.93
Mean 1.33 1.59
(SD)
Mean
Mean
(SD)
5–6 years n = 12 7–8 years n = 14
7–8 years n = 18
5–6 years n = 10
(SD)
Moroccan
Chinese
Table 12.3 Mean of requested utterances for target-pictures naming
2.92 2.46
Mean
1.32 1.50
(SD)
3.47 3.76
Mean
1.06 0.43
(SD)
5–6 years n = 13 7–8 years n = 17
Catalan
Written representations of nominal morphology 161 Table 12.4 Correlation of spoken and written modifications Number-inflection task Chinese Moroccan
Pearson 0.432* 0.522**
Derivational task Chinese Moroccan
0.730** 0.512**
Notes *Level of significance 0.05. **Level of significance 0.01.
children failed to change both spoken and written productions while virtually all of the controls did. We should recall that the initial picture with the caption remained all the time in front of the child, so that it could be used as a guide to complete the task. If the child modified the description of the initial picture to describe the target one and moreover modified the original caption to write the new caption, we regarded this as an indication of the child being aware that changes in oral descriptions should be reflected in writing. All utterances/written representations showing any kind of modification to the input—whether accurate or not—were regarded as instances of awareness of the fact that modifications must be made to reflect changes in either number or derivational status. Table 12.5 shows the percentages for each possible type of relationship between oral and written modifications, according to language and age group in the number- inflection task. These results indicate that the Moroccan children, on the one hand, whenever they failed to recognize that changes should be made to both oral and written productions, mainly modified spoken utterances and produced no changes in their written representations. Chinese children, on the other hand, resorted less to solely changing their spoken utterances and some of them modified the written representation. This strategy was almost nonexistent in the Moroccan group. A clear age effect should be noted: the percentages of modifications affecting the two modalities increases as a function of age in both the Moroccan and the Chinese group. In the control group, responses neglecting a change in either modality were almost nonexistent, as children massively changed both the spoken utterance and its written representation.
Discussion Mastering the language(s) of a literate community implies learning its written and spoken varieties. L2 learners profit from spoken and printed information of different kinds provided by the environment. We assumed that this information is actively assimilated by the learner: it is selected and, probably, transformed, as a function of children’s knowledge and
22.50
15.00
25.00
37.50
Changes in written prod. only
Changes in oral prod. only
Both oral and written prod. modified
5–6 years n = 12
5–6 years n = 10
Both oral and written prod. unmodified
Moroccan
Chinese
55.56
16.67
12.50
15.28
7–8 years n = 18
47.92
37.50
39.58
4.17
18.75
37.50
Plural condition
18.06 68.06
40.00
30.00
Changes in oral prod. only
0.00
14.58
Both oral and written prod. modified
2.78
20.00
Changes in written prod. only
11.11
10.00
5–6 years n = 12
5–6 years n = 10
7–8 years n = 18
Moroccan
Chinese
Both oral and written prod. unmodified
Singular condition
39.29
62.50
26.79
0.00
10.71
7–8 years n = 14
53.57
0.00
98.08
1.92
0.00
0.00
5–6 years n = 13
Control
100.00
0.00
0.00
0.00
7–8 years n = 17
100.00
1.92 98.08
0.00
0.00
7–8 years n = 17
0.00
0.00
7.14 0.00
5–6 years n = 13
7–8 years n = 14
Control
Table 12.5 Relationship between spoken and written modifications in the number–inflection task (in percentages)
Written representations of nominal morphology 163 ways of understanding. In the situation that we designed for the study, the interviewer provided both spoken and written information—when reading the captions that describe the initial picture—that the child could use in order to complete the task. Examining the way in which children modified or repeated initial utterances to describe the target pictures, we were able to appreciate their sensitivity to the morphological mechanisms of the L2. Assessing the relationships between the modifications produced in spoken utterances and those produced in written representations, we were able to infer children’s degree of awareness of the links between speech and writing in the host linguistic community. We were particularly interested in determining the extent to which typological differences between learners’ L1s, as well as the situations of diglossia from which they proceeded, affect their sensitivity to morphology and their awareness of the links between the WL2 and the L2. Our findings show that there is a subtle influence of learners’ L1 in their mastery of critical aspects of L2 nominal morphology. In line with Dressler’s (1997) hypothesis, growing up in a language with rich morphology like Darija seems to enhance sensitivity to the morphological mechanisms of Catalan—also morphologically rich—more than growing up in a language with poor morphology, such as the language spoken by our Chinese subjects, but only slightly so. Children of Moroccan origin find it easier to reflect changes in number through inflection, though not in a significant way. Conversely, Chinese children were slightly better at solving derivation of locatives. Also, the linguistic features of the L1s seem to influence the deviant strategies children deploy for describing changes in number or derivation of a locative noun from a base noun. In sum, although there were no clear indicators of L1 directly shaping L2 acquisition of number marking and locative derivation, some L1 influences were indeed found, not so much in the assessment of target-like productions, but in the analysis of deviated responses. There are, however, some critical matters that reappear in all studied groups, irrespective of their L1. Chinese, Moroccan, and, to some extent, Catalan children, were all reluctant to produce plural indefinite-article NGs, while being prone to produce singular ones, when required. The cardinal article un (“a”) is considered essential for describing singular objects and, as such, is included in children’s utterances when they are asked to reflect a change from plural to singular. This element, as any cardinal number, has lexical meaning, its use is compulsory, and there are no alternatives to it. To reflect a change from singular to plural the situation is very different. Although un (“a”) alternates with uns (“a-pl.”), the cardinal function of the latter can be fulfilled by any numeral or quantifier—e.g., vuit (“eight”) or molts (“many”). Moreover, uns has no independent lexical meaning, its use is not compulsory and, even in the absence of any modifier, the plural morpheme suffixed to the noun accomplishes the representation of plurality. These reasons converge to explain children’s preference for the use of the indefinite/cardinal article + N for describing
164 L. Tolchinsky and N. Salas singular objects, and just the N(pl) or a quantifier/cardinal + N(pl), for a plurality of objects. The cardinal interpretation of indefinite articles that explains both children’s use of the singular, indefinite article and their reluctance to use the plural, indefinite one in the three languages involved is an example of a core issue that prevails over typological differences. Similarly, the overall difficulty of locative derivation over base derivation is another example of cross-linguistic generality. Data on children’s functional knowledge of literacy indicated that they are quite sensitive to the contexts of use and to the function of social texts (Tolchinsky et al., in press). Nevertheless, many children were still unaware of the basic representational feature of alphabetic writing; namely, that whenever a spoken utterance gets a written representation, the changes this utterance suffers should be reflected in its written representation as well. From the way children related to the written caption in relation to the spoken utterances they had produced, we conclude that many children were not clear about this principle of writing. They copied the written caption while changing the verbal description or, alternatively, they produced a different caption while repeating the verbalization. It could be argued that children’s way of dealing with the written/spoken relation only reflects children’s ignorance of letter/sound correspondences in the WL2 or that it is just a strategy applied by them to save the situation. Against this interpretation we should take into account that most of the participants in the study have already acquired conventional letter-sound correspondences in Catalan. Future research should explore the extent to which the kind of relation between L1 and L2 pointed out in this study is restricted to nominal morphology or can be extended to verbal morphology or other linguistic domains. Moreover, more studies are required to better understand the relations between children’s initial and further literacy learning in a new language.
References Dressler, W. U. (2004). Degrees of grammatical productivity in inflectional morphology. Italian Journal of Linguistics, 15, 31–62. Dressler, W. U. (2007). Introduction. In S. Laaha & S. Gillis (Eds.), Typological perspectives on the acquisition of noun and verb morphology. Antwerp Papers in Linguistics, 112. Durgunoglu, A., Mir, M., & Ariño-Martin, S. (2002) The relationships between bilingual children’s reading and writing in their two languages. In S. Ransdell & M. L. Barbier (Eds.), New directions for research in L2 writing. Dordrecht: Kluwer. Ferreiro, E. (2002). Escritura y Oralidad: unidades, niveles de análisis y conciencia metalingüística. In E. Ferreiro (Ed.), Relaciones de (in)dependencia entre oralidad y escritura. Barcelona: Gedisa. Karmiloff-Smith, A. (1992). Beyond modularity: A developmental perspective on cognitive science. Cambridge, MA: MIT Press/Bradford Books.
Written representations of nominal morphology 165 Liceras, J. M. (1997). La evolución del concepto de parámetro en la teoría lingüística y las consecuencias para la gramática comparada. In P. Fernández Nistal & J. M. Bravo (Eds.), Aproximaciones a los estudios de traducción (pp. 71–113). Valladolid: SAE, University of Valladolid. Perdue, C. (1993). Adult language acquisition. Crosslinguistic perspectives (2 volumes). Cambridge: Cambridge University Press. Rafel i Fontanals, Joaquim (Ed.). (1998). Diccionari de freqüències. 3 Dades globals. Barcelona: Institut d’Estudis Catalans. Smith, N., & Tsimpli, I. M. (1995). The mind of a savant: Language learning and modularity. Oxford: Blackwell. Tolchinsky, L., Salas, N., & Perera, J. (in press). Spoken and written representation of number in L2 Catalan indefinite DPs. Catalan Review. Tolchinsky, L. (2003) The cradle of culture and what children know about writing and numbers before being taught. Hillsdale, NJ: Lawrence Erlbaum. Tsimpli, I. M., & Roussou, A. (1991). Parameter-resetting in L2? UCL Working Papers in Linguistics, 3, 149.
13 Relationships between idea generation and transcription How the act of writing shapes what children write John R. Hayes* and Virginia W. Berninger† *Carnegie Mellon University; †University of Washington
Adult writing researchers have asserted explicitly (Kellogg, 1994, pp. 30–31) or assumed implicitly that transcription skills can be safely ignored in studying writing. For example, Hayes and Flower (1980) and Hayes (1996) identified three cognitive processes in adult writing: Planning (generating ideas and goal setting), Translating, and Revising. Child-writing researchers have observed that handwriting and spelling problems are correlated with composition length and quality (e.g., Graham, 1990). Because many children are referred for clinical assessment of illegible handwriting and/or non-conventional spelling, Berninger and colleagues (Berninger et al., 1992; Berninger, Cartwright, Yates, Swanson, & Abbott, 1994) conducted cross-sectional studies that showed individual differences in handwriting and spelling contribute uniquely to composition length and quality. Their instructional studies showed causal relationships between transcription and composition: training handwriting or spelling transferred to better composing. These findings were replicated by Graham and colleagues for handwriting and spelling and by Christiansen and colleagues for handwriting. (For reviews see Berninger & Amtmann, 2003; Berninger, 2008.) Berninger and Swanson (1994) argued that, because transcription (handwriting and spelling) requires substantial working-memory resources, it plays a critical role in beginning writing. Bourdin and Fayol (1994) demonstrated that working-memory loads affect children’s transcription in general and adults’ transcription when words are unfamiliar or irregular in spelling. Chenoweth and Hayes (2001) added transcription to Hayes’ (1996) model of cognitive processes in writing. Although resource-draining transcription skills play an important role in adult skilled writing (Hayes, in press), relative processing loads for writing activities and their task environment may vary across writing development; see Figure 13.1 for relative loads for young writers (Hayes, 2008). Researchers have investigated whether children produce longer texts of better quality if they dictate text orally, thus bypassing transcription, rather than directly transcribing text in writing. For example, De La Paz and Graham (1995) found that, if primary school children dictated rather than wrote their texts, text quality improved significantly.
Idea generation and transcription 167 Basic writing model Task environment
Writing activity
Assignment
Set goal
Processing load
Generate idea
Translate
Text-so-far
Transcribe
Evaluate
Quit
Figure 13.1 Interaction of writing processes, task environment, and processing load across writing development with relative loads for writers Grades 1–6 (source: Hayes, 2008).
Other research pointed to transfer of practice and instruction within transcription mode—handwriting or typing. Handwriting practice improved children’s writing of high-quality texts by hand (Jones & Christensen, 1999). For eighth and ninth graders with low typing skills, typing practice improved typed text quality, but not length or quality of handwritten texts (Christensen, 2004). Yet other research compared composing across transcription modes. Connelly, Gee, and Walsh (2007) compared fifth and sixth graders’ essays written by hand and by keyboard. Students wrote significantly faster by hand than keyboard. Handwritten essays were significantly superior to typed essays on six analytic scoring categories: Ideas and development, Organization, Unity and coherence, Sentence structure, Grammar, and Punctuation. Collectively, the research findings indicate that transcription is related to composition. However, little research attention has been given to how the idea-generation process, which activates ideas in working memory, might be influenced by the transcription mode (pen or keyboard) used when translating ideas into written language. In our study, we focus on the relation of transcription to idea generation in second-, fourth-, and sixth- grade children, but note children were asked to generate ideas on a specified topic not to generate ideas for the purpose of composing. After an intervening sentence-writing task, they were asked to write an essay about each topic for which oral idea-generation protocols were collected. According to Bereiter and Scardamalia (1987), “By children’s own reports, their main problems in generating text are problems of finding content, not of finding language to express it” (p. 62). They attributed
168 J. R. Hayes and V. W. Berninger Table 13.1 Numbers of ideas generated for two modes and four grades Grade
Dictated orally
Transcribed in writing
2 4 6 8
7.50 9.65 5.57 7.55
8.05 12.45 24.25 27.70
Source: Boscolo (1990).
c hildren’s trouble in finding content to difficulty with metamemorial search, which is answering the question “what do I know about the topic” (p. 66). From this perspective, beginning and developing writers should have difficulty generating ideas even if they do not have to transcribe them. Boscolo’s (1990) study provides a test for this prediction. He asked children in Grades 2, 4, 6, and 8 to dictate ideas on an assigned topic to an adult who transcribed each idea onto a card, which children were then given and asked to use the ideas to write a text. They were told they could add to, delete from, or modify their ideas as they constructed their essays. Boscolo found that children at all four grade levels included more ideas (and in later grades, many more ideas) in their written texts than they had produced while dictating (see Table 13.1). In the current study we also compared the number of ideas children dictated while thinking about a topic with the numbers of ideas included in transcribed texts about the topic, but procedures differed from Boscolo’s studies (see Discussion). We addressed two questions: 1 Do young children express more ideas on a topic when they dictate the ideas than when they transcribe them? 2 Does the transcription mode (handwriting vs. typing) influence either the number of ideas expressed or transcription time when writing an essay?
Method Participants Children were recruited for the longitudinal study by sending letters of invitation to parents of every kindergartener or second grader in a large urban school district in the Pacific Northwest. Children were tested annually at the university for a half-day during which they completed writing and other activities in Grades 1–5 (Cohort 1) or Grades 3–7 (Cohort 2). Children with developmental or medical history indicating developmental problems were excluded. The sample was diverse in ethnicity and parents’ level of education. Parent questionnaires documented children’s prior
Idea generation and transcription 169 experience with computers and keyboards both at home and at school. Average achievement of the sample was above the population mean in written expression (see Berninger et al., 2008 for detailed sample description). Procedures for idea generation by dictating, handwriting, and typing For this study we compared two tasks completed in Grades 2 and 4 (Cohort 1) and Grades 4 and 6 (Cohort 2). First, children orally generated ideas about computers and then wrote an essay about computers by pen. Second, they orally generated ideas about robots and then wrote an essay about robots by keyboard. Topics were varied across transcription modes to avoid effects due to the previously expressed ideas for one mode influencing idea expression for the second mode. However, to keep metamemory search requirements as comparable as possible, two topics were chosen with which children should be equally familiar. Time limits were five minutes for oral idea generation but ten minutes for written idea expression to allow extra time for transcription. Both idea-generation protocols were audiotaped and immediately following the testing session transcribed by examiners into writing for data analyses. For both essays, examiners asked children to identify any word they could not decipher because of illegible handwriting or unconventional spelling and to record intended words for subsequent data analyses. Instructions for dictating ideas about computers: “Here is a laptop (show computer). Tell me all ideas you can about computers. Also use your imagination and tell me your own original ideas about computers. Tell me as many ideas as you can.” Examiners were instructed to prompt children if they paused during idea generation with “Tell me more of your ideas about computers.” Instructions for composing in handwriting: “Explain what a computer is and what it does to someone who has never seen one or used one.” The examiner then gave the child a pen and instructions to write the essay with it and make changes by crossing out and rewriting rather than erasing. Instructions for dictating ideas about robots: “Here is a picture of robots. Tell me all ideas you can about robots. Also use your imagination and tell me your own original ideas about robots. Tell me as many ideas as you can.” Examiners were instructed to prompt children if they paused during idea generation with “Tell me more of your ideas about robots.” Instructions for typing by keyboard: “Explain what a robot is and what it does to someone who has never seen one or used one.” The examiner then instructed the child to write the essay with the computer keyboard and to make changes by backspacing and re-entering. The examiner also recorded children’s revisions. The instructions for dictating ideas were designed to allow us to assess children’s metamemorial search skills and the instructions for writing texts
170 J. R. Hayes and V. W. Berninger were designed to allow us to assess ideas generated while writing texts. The instructions to dictate differ from the instructions to write essays because they do not imply that the child should create a connected text. If the numbers of ideas children express in writing were limited primarily by metamemorial search, then children should express as many ideas in dictation as in writing. However, if children dictate substantially more ideas than they write, that would suggest that the number of ideas that they write is not primarily limited by metamemorial search. Scoring ideas As yet, researchers have not adopted a standard method for counting ideas. Puranik, Lombardino, and Altmann (2008) used number of sentences. Boscolo (1990) counted simple sentences as one idea and counted complex sentences, divided into theme-rheme units, as multiple ideas. Creativity researchers (Marcos & Moran, 1989; Thies & Friedrich, 1977; Turner, 1999; Wallach & Kogan, 1965) typically measured “ideational fluency” by number of answers participants give to questions such as, “How many things can you think of that are square?” or “How many uses can you think of for a piece of paper?” Rijlaarsdam and Van den Berg (2007) used number of writing protocol segments devoted to idea generation as an index of idea generation. Caccamise (1987) developed a scheme based on the premise that ideas correspond to the propositional structure of statements using Kintch and van Dijk’s (1978) methods. For example, “Jeffrey has a new computer” consists of four propositions: 1 2 3 4
Predicate: Has (agent, object) Argument: Agent: Jeffrey Argument: Object: Computer Argument: Modifier: New
Because more complex sentences required more complex analyses that are labor intensive and her data set was large, Caccamise simplified Kintch’s method to analyze her data and provided illustrations but not specific coding rules. For example, her system does not include modifiers as separate ideas. In our coding scheme, which employed a simplified coding scheme similar, but not identical, to that of Caccamise, the following were counted as ideas: 1 Kernel assertions with any modifiers (protocol segments which make an attribution or identify an activity together with associated agents, recipients, and modifiers) were counted as single ideas. For example: Computers are very cool Robots can do a lot of things
Idea generation and transcription 171
Computers can help you with a lot of stuff
The inclusion of “and” typically added an idea. For example, “You can do email and the Internet on computers” was counted as two ideas. 2 Clauses or phrases modifying a kernel assertion were counted as single ideas. For example:
“If you go on the Internet,/you could use Facebook/to meet friends” was counted as three ideas (designated by slashes). Not coded as ideas were: 1 Incomplete assertions. For example: “Computers can help because of . . .” 2 Uninformative language. For example:
In “You can play games with computers and stuff,” for which “and stuff” was not counted as an idea. “A computer is a thing that you can type on” is counted as one idea rather than as an assertion plus a modifying clause. “A computer is a thing” is not counted as a separate assertion because it does provide new information about computers. 3 Final statements, such as “That’s all I know about computers,” were not counted because they are editorial comments rather than new ideas.
Results Figure 13.2 shows the numbers of ideas produced during dictation, handwriting, and typing for both cohorts and all three grade levels. To summarize results, children expressed more ideas by dictation, which does not require transcription, than by handwriting, and more ideas by handwriting than by typing. Furthermore, older children generally expressed more ideas than younger ones. All the differences shown in Figure 13.2 are significant with these exceptions: Students in Cohort 2 did not dictate significantly more ideas in Grade 6 than Grade 4. Thus, oral idea generation may begin to asymptote about this time. No significant differences between the two cohorts occurred at Grade 4. Although students were asked to dictate ideas about both topics, they handwrote only about computers and typed only about robots. For this reason, we separated comparisons between output modes/topics in the following analyses. See Table 13.2, which shows the average number of ideas dictated, and Table 13.3, which shows average number of ideas written by pen and by keyboard. Comparing number of ideas dictated at different grades and on different topics. Repeated measures ANOVA for Cohort 1 showed that main effects for grade (second vs. fourth) (F = 2975.625; df = 1.89; p < 0.001) and topic
172 J. R. Hayes and V. W. Berninger 20 18 16 14 12 10 8 6 4 2 0
Grade 2
Grade 4 Dictated cohort 1 Handwritten cohort 1 Typed cohort 1
Grade 6 Dictated cohort 2 Handwritten cohort 2 Typed cohort 2
Figure 13.2 Numbers of ideas dictated, handwritten, and typed by grade (numbers of ideas dictated are averaged over topics).
(computers vs. robots) (F = 275; df = 1.89; p < 0.015) and grade-by-topic interaction (F = 424.699; df = 1.89; p = 0.003) were significant. Children in Grade 2 dictated slightly, but not significantly, more ideas about computers than robots, but children in Grade 4 dictated significantly more ideas about robots than computers. Repeated measures ANOVA for Cohort 2 indicated no significant main effects or interactions, consistent with a fourth-grade asymptote. Comparing number of ideas dictated vs. handwritten. See Tables 13.2 and 13.3. Repeated measures ANOVA for Cohort 1 showed that main effects for mode (dictation vs. handwriting) (F = 38.12; df = 1.91; p < 0.001) and grade (second vs. fourth) (F = 42.12; df = 1.91; p < 0.001) and mode-bygrade interaction (F = 7.48; df = 1.91; p = 0.007) were significant. The increase in ideas was greater from Grade 2 to Grade 4 for handwriting than for dictation about computers. On average, second graders in Cohort 1 generated 10.22 ideas when dictating and 4.35 ideas when handwriting about computers. Fourth graders generated 14.96 ideas when dictating and 10.93 ideas when handwriting about computers. Repeated measures ANOVA for Cohort 2 showed that main effects for mode (F = 20.92; df = 1.97; p < 0.001) and grade (F = 7.29; df = 1.97; p < 0.008) and mode-by-grade interaction (F = 4.47; df = 1.97; p = 0.037) were significant. The interaction occurred because the increase in ideas
Idea generation and transcription 173 Table 13.2 Numbers of ideas dictated about each topic in three grades Topic
Computer
Cohort Grade 2 Grade 4 Grade 6
1 10.22 14.86
Robot 2 17.72 18.22
1 9.94 18.32
2 16.10 17.95
Table 13.3 Numbers of ideas written about each topic/mode in three grades Topic
Computer by pen
Robot by keyboard
Cohort Grade 2 Grade 4 Grade 6
1 4.35 10.93
1 2.67 7.39
2 10.86 15.25
2 6.74 11.68
from Grade 4 to Grade 6 was greater for handwriting than dictating about computers. On average, fourth graders in Cohort 2 generated 17.72 ideas while dictating and 10.86 ideas while handwriting about computers. Sixth graders in Cohort 2 generated 18.22 ideas while dictating and 15.25 ideas while handwriting about computers. Comparing number of ideas dictated vs. typed. See Tables 13.2 and 13.3. Repeated measures ANOVA indicated that for Cohort 1 main effects for mode (F = 100.68; df = 1.89; p < 0.001) and grade (F = 60.36; df = 1.89; p < 0.001) and mode-by-grade interaction (F = 6.62; df = 1.89; p = 0.012) were significant. The increase in ideas was greater from Grade 2 to Grade 4 for dictating than for typing about robots. Repeated measures ANOVA indicated that for Cohort 2 the main effects for mode (F = 51.09; df = 1.96; p < 0.001) and grade (F = 13.00; df = 1.96; p < 0.001) were significant, but not their interaction (F = 2.86; df = 1.96; p = 0.094). Students dictated more ideas than they typed about robots. In both modes, the numbers of ideas increased from Grades 4 to 6. Comparing ideas expressed when handwriting and typing. See Table 13.3. Repeated measures ANOVA for Cohort 1 showed main effects for mode (F = 58.20; df = 1.93; p < 0.001) and grade (F = 160.08; df = 1.93; p < 0.001) and mode-by-grade interaction (F = 9.56; df = 1.93; p = 0.037) were significant. The interaction indicated that increase in ideas was greater from Grade 2 to Grade 4 for handwriting than typing about different topics. Repeated measures ANOVA showed that for Cohort 2 main effects for mode (F = 69.87; df = 1.98; p < 0.001) and grade (F = 63.81; df = 1.98; p < 0.001) were significant, but not their interaction (F = 0.28; df = 1.98; p = 0.600). Matched t-tests showed that, whether handwriting or
174 J. R. Hayes and V. W. Berninger typing, sixth graders generated significantly more ideas than fourth graders and fourth graders generated significantly more ideas than second graders (p < 0.001). Relationship of transcription speed and text generation. As shown in Table 13.4, we also examined the time for integrating idea generation, text generation, and transcription when composing essays. Consistently across each grade for each cohort, children required fewer seconds per word when transcribing by pen than by keyboard, which may have facilitated idea expression.
Discussion Significance of current findings. Developing writers in Grades 2, 4, and 6 included significantly more ideas in handwritten than typed essays and generated words more quickly by pen than by keyboard. These findings replicate those of Connelly et al. (2007) whose sixth graders outscored fifth graders on “ideas and development” and wrote text faster by handwriting than keyboarding. These findings extend those of Berninger and Swanson (1994) who focused on automaticity (amount written in first 15 seconds) or fluency (amount written within time limit) to text-generation time as a function of transcription mode. The finding that oral idea generation may asymptote about fourth grade is consistent with findings of Thies and Fredrich (1977). They also found a comparable number of ideas produced (10–16 per topic). The number of ideas included in compositions by either transcription mode increased significantly from second to fourth or sixth grade. Likewise, Christiansen (2004) found children express more ideas when practiced and familiar with transcription. However, children showed more relative improvement in written expression than oral expression of ideas over this developmental span. Children consistently generated more ideas by dictation than by either handwriting or typing. This result suggests that metamemorial search was not the primary constraint on idea generation in children’s essay writing. Contrasts with Boscolo’s study. Although at first glance this last result appears inconsistent with Boscolo’s (1990) results (Table 13.1), his and our study differed in at least five ways. First, children in Boscolo’s study were provided with the transcript of the ideas they had dictated earlier, but children in our study were not. The availability of the transcript of generated ideas may enable writers to elaborate further on ideas. Second, in our study, physical representations of the topics were in the child’s presence during brainstorming (a laptop for the computer topic and pictures of robots for the robot topic) but not in Boscolo’s study. Some children used these physical representations, which facilitate metamemorial search challenges described by Bereiter and Scardamalia (1987), to generate some of their ideas. For example, they described the properties of the
By pen
15.52 7.19
6.67 4.60
(msec/w)
Cohort 1 Grade 2 Grade 4
Cohort 2 Grade 4 Grade 6
4.69 3.91
20.99 5.09
SD
12.99 6.10
30.07 13.06
By keyboard
8.56 4.71
16.23 6.44
SD
1,118 1,104
1,85 1,71
df
87.38 12.34
26.22 71.13
F
0.001 0.001
0.001 0.001
p
0.42 0.11
0.24 0.50
Partial eta2
Table 13.4 Mean seconds per word (msec/w) in writing essays by pen and by keyboard for Cohort 1 at Grades 2 and 4 and Cohort 2 at Grades 4 and 6
176 J. R. Hayes and V. W. Berninger robots in the pictures they were shown. Table 13.5 shows the proportion of three categories of ideas brainstormed in our study: Perception—reporting the properties of a visually present object (example: computers have a screen); Personal knowledge—reporting episodic memories or definitions (example: my dad has a computer); Fantasy—reporting personal fantasies, movie plots, etc. (example: robots fight in space). Perception accounted for a smaller proportion of ideas. However, personal knowledge (of personal experience, movie plots, etc.) accounted for the larger proportion of ideas for computers; and fantasy accounted for the larger proportion of ideas for robots. Thus, both topic and transcription may influence idea expression in writing. Third, memory requirements may have interacted with access to knowledge about the topic and explain difference between the two studies. Memory requirements may have been greater for Boscolo’s participants who heard a lecture on “the wind” two weeks before they were asked to brainstorm and write on the topic. Some may have believed their task was to produce information from that lecture, as is often expected in school writing, and not from other sources. Two quotes from the students’ protocols are instructive: “the Bora is destructive” and “Francis Beaufort invented a scale.” Although the first could have come from the students’ personal experience, the second likely did not. Fourth, students in Boscolo’s study may have believed their task during idea generation was to produce coherent content for an essay rather than random ideas. Boscolo found no difference between two conditions in which half the sample were asked to tell all they knew about the wind and the other half were asked to generate ideas for this title: “The wind: How does it form and what effects does it produce?”
Table 13.5 Percent of ideas generated in three categories of ideas
Computers
Robots
Perception
Personal knowledge
Fantasy
Grade 2 Grade 4 Grade 6
3 16 7
88 81 93
9 2 0
Average
9
87
4
Grade 2 Grade 4 Grade 6
9 11 17
43 42 36
48 47 48
Average
12
41
48
Idea generation and transcription 177 Finally, Italian has a transparent orthography that is more predictable, whereas English has a deep orthography that poses more transcribing challenges because its spellings are less predictable (Seymour, Aro, & Erskine in collaboration with COST Action A8 network, 2003).Thus, students in Boscolo’s study could transcribe more than they dictated while those in our study dictated more than they transcribed. Taken together, comparisons across the studies document many vari ables affecting the complex relationships between idea generation and transcription in developing writers, the topic of this chapter which requires further research. Our study provides at least an “existence proof” that, under some circumstances when not burdened by transcription, and for some topics, developing writers readily generate more ideas when transcription is not required. Factors limiting number of ideas expressed in children’s essays. Transcription mode is one factor that may limit number of ideas expressed in writing. As shown in Figure 13.2, consistently from second to fourth to sixth grade, children expressed more ideas in handwriting than by keyboard when composing essays. From Grade 2 to Grade 4 a greater increase in ideas expressed occurred for handwriting than typing (see Table 13.2). Also, children wrote words faster by pen than by keyboard (see Table 13.5). This finding extends prior results showing that they wrote longer essays by pen than keyboard (Berninger et al., 2008). However, four other factors besides transcription could also affect idea expression during composing: choice of topic, constraints on the search space, ability to use text written so far, and motivation. As shown in Table 13.5, the source of ideas varied by topic and could affect the written content. Children’s ideas about computers were drawn largely from personal experience but their ideas about robots came mostly from fantasy. Although children appeared to comply with instructions to say all they knew about the topic, for other writing tasks children may believe that they should produce ideas just for a text they are about to write, which could reduce fluency in generating ideas. Also, as children become more skilled in writing, they should increasingly use text written so far to help them to generate new ideas, as Boscolo found. At some point in writing development, the number of ideas generated orally and in writing and by handwriting and keyboarding might equalize; the results shown in Figure 13.2 suggest that may happen after sixth grade. If writing is perceived as hard work, children may become writing-avoidant because of the effort required to express their ideas in writing. Figure 13.3 depicts a model customized for child writers. The idea generator takes a goal set by the planning process, based on the task instructions and materials, to guide, with cues such as the goal (topic) and previously generated ideas in working memory, the information retrieval from long-term memory for purposes of idea generation. Personal experience or cultural knowledge stored in long-term memory may influence the efficiency of idea generation.
178 J. R. Hayes and V. W. Berninger A model for idea generation in writing Reinstate goal
Goal setting
Task environment
Working memory
• Task instructions and materials • Distractors • Text-so-far
• Goal • Current subgoal • Previous ideas
Filter
No Idea relevant to goal?
Yes
Translating
Long-term memory Knowledge base
Figure 13.3 Idea generation in developing writers (source: Hayes, 2008).
The task environment influences this process in several ways. First, distracters may make it difficult for beginning writers to maintain focus on the topic. Second, task materials themselves may provide cues for ideas and the task instructions (if printed) can be used to reinstate the goal if it is forgotten. Third, beginning writers may vary in their ability to use the text as a source of retrieval cues during idea generation. Finally, the filter, a local editor, may modify the output of the idea generator. If the goal is to generate ideas freely on a topic, then the filter will reject off-topic ideas. However, if the goal is more restrictive, say to select only ideas that contribute to the cohesion or flow of text, then the filter will select only the most relevant ideas from among those proposed. If judged relevant, generated ideas are passed to the translator to be transcribed into written language.
Summary Important findings were that children expressed more ideas when (a) dictating orally than when transcribing, and (b) handwriting with pen than typing with keyboard. They also expressed those ideas in written words faster by handwriting than keyboarding. The results are consistent with the view that transcription processes may interfere with children’s ability to express ideas in written language. Children improved relatively more in written than oral expression of ideas over the developmental span studied, showing the value of teaching transcription to facilitate expression of ideas in writing.
Idea generation and transcription 179
References Bereiter, C., & Scardamalia, M. (1987). The psychology of written composition. Hillsdale, NJ: Lawrence Erlbaum Associates. Berninger, V. (2008). Evidence-based written language instruction during early and middle childhood. In R. Morris & N. Mather (Eds.), Evidence-based interventions for students with learning and behavioral challenges (pp. 215–235). Hillsdale, NJ: Lawrence Erlbaum Associates. Berninger, V., & Amtmann, D. (2003). Preventing written expression disabilities through early and continuing assessment and intervention for handwriting and/ or spelling problems: Research into practice. In H. L. Swanson, K. Harris, & S. Graham (Eds.), Handbook of research on learning disabilities (pp. 345–363). New York: Guilford. Berninger, V., Cartwright, A., Yates, C., Swanson, H. L., & Abbott, R. (1994). Developmental skills related to writing and reading acquisition in the intermediate grades: Shared and unique variance. Reading and Writing: An Interdisciplinary Journal, 6, 161–196. Berninger, V. W., Richards, T. L., Stock, P. S., Abbott, R. D., Trivedi, P. A., Altmeier, L. E., & Hayes, J. R. (2008). From idea generation to idea expression in language by hand. British Journal of Educational Psychology Monograph. Berninger, V. W., & Swanson, H. L. (1994). Modifying Hayes and Flower’s model of skilled writing to explain beginning and developing writing. In E. C. Butterfield & J. Carlson (Eds.), Children’s writing: Toward a process theory of the development of skilled writing (pp. 57–81). London: JAI Press. Berninger, V. W., Yates, C., Cartwright, A., Rutberg, J., Remy, E., & Abbott, R. (1992). Lower-level developmental skills in beginning writing. Reading and Writing: An Interdisciplinary Journal, 4, 257–280. Boscolo, P. (1990). The construction of expository text. First Language, 10, 217–230. Bourdin, B., & Fayol, M. (1994). Is written language production more difficult than oral language production: A working-memory approach. International Journal of Psychology, 29, 591–620. Caccamise, D. J. (1987). Idea generation in writing. In A. Matsuhashi (Ed.), Writing in real time? (pp. 224–253). Norwood, NJ: Ablex. Chenoweth, N., & Hayes, J. R. (2001). Fluency in writing. Generating text in L1 and L2. Written Communication, 18, 80–98. Christensen, C. A. (2004). Relationship between orthographic-motor integration and computer use for the production of creative and well-structured text. British Journal of Educational Psychology, 74, 551–564. Connelly, V., Gee, D., & Walsh, E. (2007). A comparison of keyboarded and handwritten compositions and the relationship with transcription speed. British Journal of Educational Psychology, 77, 479–492. De La Paz, S., & Graham, S. (1995). Dictation: Applications to writing for students with learning disabilities. In T. Scruggs & M. Mastropieri (Eds.), Advances in learning and behavioral disorders (vol. 9, pp. 227–247). Greenwich, CT: JAI Press. Graham, S. (1990). The role of production factors in learning disabled students’ compositions. Journal of Educational Psychology, 82, 781–791. Hayes, J. R. (1996). A new framework for understanding cognition and affect in writing. In C. M. Levy & S. Randall (Eds.), The science of writing: Theories,
180 J. R. Hayes and V. W. Berninger methods, individual differences, and applications (pp. 1–27). Mahwah, NJ: Lawrence Erlbaum Associates. Hayes, J. R. (2008). A framework for studying writing development. Talk presented at the Writing Across the Borders conference, February, 2008. Santa Barbara, CA. Hayes, J. R. (in press). Translating ideas into text. In R. Beard, D. A. Myhill, & J. Riley (Eds.), International handbook of writing development. London: Sage. Hayes, J. R., & Flower, L. S. (1980). Identifying the organization of writing processes. In L. W. Gregg & E. R. Steinberg (Eds.), Cognitive processes in writing (pp. 3–30). Hillsdale, NJ: Lawrence Erlbaum Associates. Jones, D. A., & Christensen, C. A. (1999). Relationship between automaticity in handwriting and students’ ability to generate written text. Journal of Educational Psychology, 91, 1–6. Kellogg, R. T. (1994). The psychology of writing. New York: Oxford University Press. Kintch, W., & van Dijk, T. A. (1978). Toward a model of text comprehension and production. Psychological Review, 85, 363–394. Marcos, G. G., & Moran, J. D. (1989). Cross-cultural comparisons of ideational fluency in preschool children. School Psychology International, 10, 199–204. Puranik, C. S., Lombardino, L. J., & Altmann, L. J. P. (2008). Assessing the microstructure of written language using a retelling paradigm. American Journal of Speech-Language Pathology, 17, 107–120. Seymour, P., Aro, M., & Erskine, J., in collaboration with COST Action A8 network. (2003). Foundation literacy acquisition in European orthographies. British Journal of Psychology, 94, 143–174. Thies, C. D., & Friedrich, D. D. (1977). Creativity: Ideational fluency and originality at the verbal and nonverbal production and recognition levels. The Creative Child and Adult Quarterly, 2, 213–226. Turner, M. A. (1999). Generating novel ideas: Fluency performance in high- functioning and learning disabled individuals with autism. Journal of Child Psychology and Psychiatry, 40, 189–201. Van den Bergh, H., & Rijlaarsdam, G. (2007). The dynamics of idea generation during writing. In G. Rijlaarsdam (Series Ed.), & M. Torrance, L. van Waes, & D. Galbraith (Volume Eds.), Writing and cognition: Research and applications (Studies in writing, Vol. 20, pp. 125–150). Wallach, M. A., & Kogan, N. (1965). Modes of thinking in young children. New York: Holt, Rinehart, and Winston.
14 Academic writing in Spanish compulsory education Improvements after didactic intervention on sixth graders’ expository texts1 Teodoro Álvarez Angulo and Isabel García Parejo Complutense University, Madrid, Spain
Since 2001, the Didactext research group has explored the relationships between writing and school from the perspective of writing across the curriculum and/or heuristic writing (what some call “writing to learn”). They are particularly interested in relating production contexts and cognitive processes with the linguistic and textual regularities of academic writing. This study focused on children in the final year of elementary education (sixth grade of elementary school). Upon finishing this school year, at age 12, children go on to finish compulsory secondary education at a “high school” (for four years, from age 12 to 16). The Didactext Group, in researching at this final elementary school year, had a double goal: to observe writing practices that were carried out in the classroom, and to design a didactic sequence to improve expository texts, which the PISA (Programme for International Student Assessment of the Organization for European Cooperation and Development 2000, 2003, 2006) reported were the kind of texts that were most in demand at high school, though the least used by the pupils. The Didactext Group used as a starting point the hypothesis that, if the processes that are involved in the production of written texts (planning–textualization–revision) were made explicit and practiced in class, the students’ written competence would improve. This chapter presents first the theoretical foundations of the model of didactic intervention used by the Didactext Group. This model is based on classroom work that is organized into didactic sequences, the activities of which guide the writing process, beginning with planning the writing, all the way through revising and editing for the final draft. Second, it offers details of the objectives and characteristics of the investigation carried out by the group and, third, it provides the most relevant results obtained. The final conclusions point toward several pedagogical implications derivable from the didactic intervention in sixth-grade classrooms.
182 T. Á. Angulo and I. G. Parejo
Didactic sequences for learning to write expository texts The group’s didactic proposal, designed for composing expository written texts (cf. Didactext, 2006; García Parejo, 2007), takes into consideration the contributions of cognitive approaches (Bereiter & Scardamalia, 1987; Flower & Hayes, 1994); social and cultural approaches (Barton, 1994; Gee, 1990; Street, 1994; Wertsch, 1991) and textual approaches (Beaugrande & Dressler, 1997; van Dijk, 1980; Bakhtin, 1975). The aforementioned proposal is built upon five basic principles: 1 Inasmuch as writing is a social skill, instruction in order to gain this skill implies creating real writing situations that make sense. 2 The different stages of the process of written composition should be made explicit, since knowledge is being built as the specific uses of the language are being acquired. 3 Resolving linguistic and rhetorical tasks that arise during the writing process (reformulating, paraphrasing, describing, arguing, regulating connectives, etc.) requires mastering specific activities. 4 The development of writing implies integrating the four skills (speaking, listening, reading, writing) related to functional aspects of language. 5 Writing in academic contexts requires teaching and learning specialized texts in different subjects or areas of knowledge. To guide a classroom intervention based on these principles, the group developed a didactic sequence (DS). A didactic sequence organizes classwork into an internally coherent set of time-sequenced didactic activities related to a central theme—in this case, composing a written text (Nemirovsky, 1999, p. 124). The objective of the proposed DS is to write coherent, appropriate, and correct texts following a certain process of text production (access to knowledge, planning, textualization, revision), framed in a social practice, and with explicit reflection about the text. Thus, the contents for the design of the DS are organized around the explicit study of three components: 1 The knowing of forms and uses of the different text types (or genres), as well as the writing process. 2 The practice of strategies related to the writing process (registering, selecting, organizing, evaluating . . .) 3 The adoption by the students of positive, critical, creative, and realistic points of view toward their own writing and the writing of others.
Objectives of the project Considering that literacy practices are socially situated practices (Barton, 1994; Gee, 1996; Street, 1984), this research was articulated around two
Academic writing in Spanish education 183 central themes: observation and intervention. Thus, the team designed a one-year project with two distinct stages, a descriptive one and an experimental one. In the descriptive stage (from October to March), the project objectives were: • •
to analyze the educational practices of the teachers regarding teaching text writing and their beliefs about writing; to analyze the characteristics of written expository competence of the students within this context and their beliefs about writing.
In the experimental stage (from April to May), the project objectives were: • •
to design a model of a teaching sequence based on the stages of written production (planning–textualization–revision) and on the explicit teaching of the textual features of the expository genre; to compare the degree to which there was an improvement in the features of pupils’ expository texts after the application of the didactic sequence.
Method Subjects The research was carried out at three schools in the vicinity of Madrid. At each school one elementary sixth-grade class (12-year-olds) participated. The three classes selected had the following characteristics: Cantoblanco (public school, north of Madrid) 23 pupils: 7 boys, 16 girls San Sebastián de los Reyes (public school, north of Madrid) 25 pupils: 13 boys, 12 girls Alcorcón (semi-private school,2 south of Madrid) 30 pupils: 11 boys, 19 girls. First stage: study of writing practices and textual competence In order to reach the first objectives the group used both qualitative and quantitative methods of data collection: teacher interviews, student questionnaires, observation, and analysis of the contents of school documents (syllabus, lesson plan, National Curriculum in Primary Schools), and the language textbooks used by the teachers. The observation work and questionnaire focused on social, cultural, pedagogical, and psychological issues related to the literacy practices habitual in the classes. The analysis of textbooks and other didactic materials was focused on the writing activities, and particularly on those related to expository texts. The results obtained
184 T. Á. Angulo and I. G. Parejo were commented on and completed during a semi-structured interview with each of the class teachers that was recorded and later transcribed for study. This interview was carried out by a professor from the Didactext Group. The observation and classroom questionnaires were carried out by students in their practice teaching period. In order to analyze the characteristics of expository text produced within the normal classroom practice, the team asked each teacher to provide them with an expository text produced by their students. This text, which we called T1, had different features at each classroom, both in its content and in its form: Cantoblanco: exam on the central nervous system; San Sebastián: texts about the pupils’ favorite animals; Alcorcón: narration about a camp attended in the past. A template was designed for the text analysis, including 26 blocks of variables to identify (1) the superstructure of the text types, that is, the mental schemata that the pupil had followed in order to organize his/her text (for this purpose the team followed the typologies defined by Adam, 1990 or van Dijk, 1980: description–definition; classification–typology; comparison and contrast; problem, solution; question, answer; cause– consequence); (2) textual–discursive aspects characterizing expository texts in Spanish, as defined for instance in Sanahuja, 1992 or Álvarez Angulo, 2001 (reformulation, exemplifications, enumeration, comparisons, introductory colons, explanatory parentheses, explanatory “que,” explanatory appositions); (3) semantic aspects; and (4) syntactic aspects (see Appendix A). Second stage: didactic intervention During the experimental stage of the project the team measured the effects of a didactic intervention on the textual competence of the students of sixth grade. Based on the five theoretical principles mentioned at the beginning of this chapter, the team designed a didactic sequence (DS), whose objective was to write coherent, appropriate, and correct texts following a certain process of text production (access to knowledge, planning, textualization, revision), framed in a social practice, and with explicit reflection about the text. The DS was organized in five stages that took place during five weeks in each classroom (see Table 14.1). The first stage of the DS intervention began with the negotiation of the topic and context for writing with all of the members of each classroom. The topics were drawn from the lives and experiences of the members of the educational community, so a different topic was chosen in each classroom: Cantoblanco: renewable energies; San Sebastián: animals; Alcorcón: signs of the zodiac.
Academic writing in Spanish education 185 Table 14.1 Stages of the didactic sequence applied at each school DS stages
Activities
Products
1 First week. Contextualize writing process
Registering previous Lists, diagrams, forms . . . knowledge: what do we know about the topic, the writing context, and the expository text
2 Second week. To access the knowledge
Lists, diagrams, forms . . . Registering intentions What do we want/need to know? Selecting and organizing data (on the topic, task, and text type)
3 Third week. Planning
Planning First revision of the task Reading activities based on model text
Mind maps, outlines
4 Fourth week. Textualization
Toward a first draft
First draft, intermediate text
5 Fifth week. Revision-editing
Evaluation of the process and of the product Edition of the final text Evaluation of learning
Edition of the final text
At San Sebasatián, the students decided that they wanted to present the characteristics of mammals to their companions in the third grade; the students of Cantoblanco decided to address their texts to the research group; and the students of Alcorcón organized their writing project as if it were part of a writing workshop. They did not have any particular audience in mind, and wrote their texts for their own classroom. The three student teachers put the intervention into practice after participating in specific preparation and training sessions. During the intervention, they received constant orientation through weekly meetings with the research team in order to revise and organize the didactic work. After applying the DS, the team collected Text 2 (T2). They analyzed it using the same template they used to analyze the Text 1 (T1) they had gathered in each classroom before intervention. With the assistance of the Teaching and Research Support Computing Service at the Complutense University of Madrid, we analyzed statistically the changes of text vari ables between T1 and T2. A statistical significance test (Xi square and McNemar) was also applied (Z less than or equal to 0.05).
186 T. Á. Angulo and I. G. Parejo
Results Three approaches to writing in the classroom At the first descriptive stage the research team aimed to find a relationship between teaching practices and written production. A general overview of a few characteristics of the three classrooms (Tables 14.2 and 14.3) allows us to sketch a profile of each one. The observed pedagogic orientation of each classroom correlated with the classroom characteristics documented in Table 14.3 as well as with the teacher descriptions of their practice summarized in Table 14.4. The most innovative classroom seemed to be Cantoblanco’s (C1), since the teacher organized the work in groups, used active methodology, used different kinds of materials, created their own school manuals within the classroom, and organized specific workshops for learning writing skills. The Alcorcón classroom (C3) was at the other extreme; the teacher and pupils played more traditional roles, with activities based on the teacher’s transmitting his/her knowledge. The San Sebastián classroom (C2) was somewhere in between, as it combined traditional features (in classroom organization and methodology) with more innovative ones (such as the integration of the pupils’ everyday lives and reflections on their practices). Thus, writing Table 14.2 Characteristics of the three classrooms Cantoblanco
San Sebastián
a. Location
Poster about features of written language The posters are changed throughout the school year
Empty walls Poster showing how to organize a written text Posters related to curriculum content
b. Distribution of students
Arranged in mixed Arranged into groups of five or individual lines six
Arranged in lines of pairs
c. Differentiated spaces within the classroom
There are several There is a spaces: library, classroom library play corner, and areas for curricular subjects
There are none
d. Didactic resources available in the classroom
Blackboard, video, Blackboard, signs, Blackboard, signs, computer, noticeboard, sound noticeboard, sound computer system system programs, science practice materials
e. Textbook
No textbook is used
Yes (Lengua 6º, SM)
Alcorcón
Yes (Lengua 6º, SM)
Academic writing in Spanish education 187 Table 14.3 How teachers described writing and their classroom practice
a. What is writing? a means of communication a tool that gives access to knowledge a process that includes several stages (planning–textualization–revision) to produce a text with appropriate spelling and handwriting b. How do you organize your literacy practice? group work I take topics of interest to the students into account I use a topic and its explanation as a basis I point out the task objectives I use textual models I give specific instructions on punctuation and spelling I read the texts in groups and comment on them I correct the texts c. Other support for writing weekly writing workshop use of new technologies
C1
C2
X X X
X
C3
X X
X X X X
X X
X X X
X
X
X X
work at Cantoblanco was taken on from a process-based perspective, whereas in Alcorcón it was centered on the product, which was later checked by the teacher. At San Sebastián, the process was also a relevant component, although there were normally no intermediate texts or rough drafts. Some significant improvements found after the didactic intervention The comparative analysis of the textual data was carried out by counting frequencies for each type of variable selected. Generally, there was an increase in the variety of expository textual structures, in the use of explanatory textual organizers, and in the use of paratextual cues (headings and subheadings). Table 14.4 shows the most significant changes related to the pupils’ uses of different textual structures. In both T1 and T2 the predominant subtype in all the schools was that of “Description–definition” (100 percent in San Sebastián and Alcorcón), even though in Alcorcón, the data are, strictly speaking, lost values since they did not fulfill the requirements of the research as T1 was a narrative text. The lower percentage of “Description– definition” in Cantoblanco may be due to the fact that T1 was an exam on the nervous system, and the pupils had to demonstrate their knowledge to the best of their abilities, both in relation to the content and to the means of expression of that content. Thus, organizing the text into a single paragraph that described the function and form of the nervous system was the
188 T. Á. Angulo and I. G. Parejo most widespread structure, perhaps due to the means by which the knowledge on the topic had been accessed. Subtype (a) description–definition appears in T2 in all schools, together with subtype (b) classification–typology, but the difference between T1 and T2 was especially significant in C2 (4 > 11). After the didactic intervention the appearance in Cantoblanco of the subtypes (c) comparison and contrast and (e) question–answer was noted, but only the latter variable was significant (Z = 0.046). Subtypes (c) and (e) were not observed in San Sebastián or Alcorcón in either text. Subtypes (d) problem–solution and (f) cause–consequence were not observed in any class in either text. The results suggest that, on one hand, the students’ prototypical mental schemata for expository texts basically corresponds to description– definition, which they often rely on when they need to explain a topic, such as in example 1. On the other hand, they also suggest that the use and variety of text models before and after the DS were reflected in the features of T1 and T2. The Spanish-language textbook used at San Sebastián and Alcorcón included, at the end of each unit, reading and analyzing each subtype of expository text in a section called “study techniques.” However, the study and practice of this section had not been made part of the normal class routine. During the DS the model texts chosen by the student teachers for analyzing the features of expository texts in San Sebastián and Alcorcón basically corresponded to subtypes (a) description–definition and (b) classification–typology, so that the didactic intervention reinforced pupils’ tendency to use a single macrostructure as their reference. Table 14.5 shows the textual organizers appearing in student texts in each school before and after the intervention. Students in both San Sebastian and Alcorcón increase their use of different explicative organizers, whereas students in Cantoblanco show a mixed pattern evidencing little change, though the results may be confounded because the initial text was for an exam. The pupils, overall, demonstrated a mastery of a great variety of explanatory resources. The variations in San Sebastián were especially significant with respect to the use of reformulation (1 > 17 students); exemplification (15 > 21 students); enumeration (8 > 17 students); and clarificatory parentheses (0 > 4 students). These results may be due to the time spent during the DS on the analysis of the linguistic features of expository texts and on the superficial organization of the text that the pupils had to create. It is somewhat more complex to compare the use of explanatory resources in two different types of texts (narrative and expository). At Alcorcón, it was observed that the pupils were aware of these linguistic resources and used them in their narrative texts. Within the framework of the DS, the use of introductory colons increased significantly, but the use of explanatory que decreased, which might be contradictory: the first fact would confirm the efficiency of the intervention, while the second would seem to lead to the opposite conclusion. What all of these data seem to confirm is that these explicative resources are already available to a lesser or greater degree to students in sixth grade
17 4 0 0 0 0
8 6 3 0 4 0
0.007 0.317 0.083 1.000 0.046 1.000
24 4 0 0 0 0
Note * C1: the total sample includes 21 texts; C2 and C3: the total sample includes 24 texts.
a. Description-definition b. Classification-typology c. Comparison and contrast d. Problem-solution e. Question-answer f. Cause-consequence
24 11 0 0 0 0
T2
T1
*p < 0.05
T1
T2
San Sebastián
Cantoblanco
Table 14.4 Textual macroestructure chosen: predominant subclase (absolute numbers)
1.000 0.020 1.000 1.000 1.000 1.000
*p < 0.05
0 0 0 0 0 0
T1
Alcorcón
24 7 0 0 0 0
T2
0.00 0.04 1.00 1.00 1.00 1.00
*p < 0.05
7 13 15 1 12 10 2 1
4 14 14 10 17
10 6 2
1.000 0.564 0.102
0,257 0.763 0.763 0.007 0.059 0 0 2
1 15 8 7 3
Note * C1: the total sample includes 21 texts; C2 and C3: the total sample includes 24 texts.
a. Reformulation b. Exemplification c. Enumeration d. Comparisons e. Introductory, explanatory colons f. Explanatory parentheses g. Explanatory que h. Explanatory apposition 4 1 3
17 21 17 12 5
T2
T1
*p < 0.05
T1
T2
San Sebastián
Cantoblanco
Table 14.5 Explanation techniques (absolute numbers)
0.046 0.317 0.564
0.000 0.034 0.013 0.132 0.414
*p < 0.05
2 14 4
1 2 5 2 4
T1
Alcorcón
2 6 5
1 1 7 2 14
T2
1.000 0.021 0.739
1.000 0.564 0.157 1.000 0.012
*p < 0.05
Academic writing in Spanish education 191 in elementary school. However, with the didactic intervention in which the use of these marks in different textual structures was made explicit, the increase in certain organizers can be observed, some as complex as parentheses or colons. Table 14.6 shows statistically significant changes in the use of paratextual features between T1 and T2, especially in relation to the organization of the text through titles and headings. All of the students gave their text a title, and most of them organized the text into sections or parts, as shown in Table 14.6. This can be related to the greater variety of textual subtypes that the students in Cantoblanco used, and, more specifically to the time spent during the DS on analyzing textual and paratextual features of expository texts, to planning and organizing the texts. Comparative study of two texts. The results seem to suggest that, independently from the writing practices habitually carried out in the classroom, the sixth-grade pupils at the three classrooms know how to adjust their texts to the communicative task required (narration or exposition). This entailed the use of prototypical mental structures of organization and basic forms of cohesion. As regards the features of their expository texts, in T1 (with the exception of Alcorcón), 100 percent of the students chose the “definition–description” organization format, with at least some sequences including a “classification–typology” format. We can get a more concrete sense of the cumulative effect of these changes by examining the pre- and post-intervention texts of one student. This student from San Sebastián organized the T1 text according to the prototypical description–definition structure, and at the end added a sentence that belongs to the classification–typology explanatory subtype. The student seemed to follow a mental schema including what a dog is like, what it likes, and what kinds of dogs there are. However, the text is organized into a single paragraph, where only one kind of punctuation mark— the comma—is used, and only one connective: the conjunction “y” (“and”) is used. There are no headings or subheadings. And in addition, there are a couple of spelling mistakes: se [sé], pekines [pekinés] with no accent mark. (1) El perro es el mejor amigo del hombre, el perro anda a cuatro patas y le gusta jugar con pelotas se unas cuantas razas, Pastor Aleman, Dowerman, Boxer, Galgo, Pequines, calie, Floxterrier . . . Dogo, caniche, afgano. [The dog is man’s best friend, dogs walk on four legs and like playing with balls I know a few breeds, German Shepherd, Doberman, Boxer, Greyhound, Pekinese, calie, Floxterrier [sic] . . . Dogo, poodle, Afghan.] Although this text presents the features of expository texts, the experiential tone stands out due to the presence of the first person (sé—know— 1st-sg.), and to the lack of descriptive resources, since it is focused on the
17 9 3 2 0
21 16 1 0 1
0.046 0.020 0.317 0.157 0.317
15 1 1 0 0
Note * C1 the total sample includes 21 texts; C2 and C3: the total sample includes 24 texts.
a. Headings b. Subheadings and sections c. Pictures-illustrations d. Graphs e. Tables-diagrams
24 23 1 0 0
T2
T1
*p < 0.05
T1
T2
C2 San Sebastián
C1 Cantoblanco
Table 14.6 Use of paratextual elements (absolute numbers)
0.003 0.000 1.000 1.000 1.000
*p < 0.05
21 0 0 0 0
T1
22 24 2 1 1
T2
C3 Alcorcón
0.564 0.000 1.000 1.000 1.000
*p < 0.05
Academic writing in Spanish education 193 behavior of the animal associated to games that the pupil probably often plays with a dog he or she knows. In the analysis of this pupil’s production after the didactic intervention, the change was clear. There had been a progression from a 35-word text to a 183-word one, and from a text organized into a single paragraph into one with six paragraphs and six headings. The basic organizational structure was the same, description–definition, but it had gained in complexity. (2) Introducción: Los mamíferos son seres vivos, es decir, cumplen las funciones, son animales vertebrados (tienen columna vertebral), su cuerpo esta dividido en cabeza, tronco y extremidades, hay individuos machos y hembras, son vivíparos; es decir nacen del vientre de su madre. Características: Los mamíferos se llaman asi porque (maman del vientre) nacen del vientre de su madre, su cuerpo esta cubierto de pelo, respiran por pulmones y son animales de sangre caliente es decir la temperatura del cuerpo no varía. Alimentación En cuanto a la alimentación pueden ser carnívoros (comen carne) ejemplos: (el león), herbívoros (comen plantas) ejemplo (la vaca), Omnívoros (comen de todo ejemplos (el oso). ¿Dónde viven? Los mamíferos pueden vivir en el medio terrestre (el elefante) en el medio acuático (el delfín, la ballena), o en el medio aéreo, en este caso solo hay uno el (murciélago). Reproducción Son vivíparos es decir las crías nacen del vientre de su madre, su fecundación es interna. Curiosidades La ballena es el mamífero más pesado, el canguro puede llegar a alcanzar los 100 km por hora, el leon es el depredador más feroz. [Introduction: Mammals are living beings, that is, they carry out the functions, they are vertebrate animals (they have spinal columns), their bodies are divided into head, trunk, and extremities, there are male and female individuals, they are viviparous; that is they are born from their mother’s womb. Characteristics: Mammals are called as such because (they suckle from the womb) they are born from their mother’s womb, their body is covered
194 T. Á. Angulo and I. G. Parejo with hair, they breathe through lungs and they are warm-blooded animals that is their body temperature does not vary. Nutrition: As for nutrition they can be carnivores (they eat meat) examples: (lions), herbivores (they eat plants) example (cows), Omnivores (they eat everything examples (bears). Where do they live? Mammals can live on land (elephants) in an aquatic environment (dolphins, whales) or in the air, in this case there is only one the (bat). Reproduction: They are viviparous that is the babies are born from their mother’s womb, their conception is internal. Interesting facts: The whale is the heaviest mammal, the kangaroo can reach up to 100 km per hour, the lion is the most ferocious predator.] With the exception of some errors of accent marking (related to the use of está) and several punctuation errors (the use of the comma after connectives), the other main change in this pupil’s text was related to the linguistic resources used. On one hand, the experiential tone of the discourse had disappeared: the text maintained the characteristic third person of the expository genre, making use of more specialized and technical vocabulary throughout (vivíparo, omnívoro, vientre . . .). On the other hand, the textual coherence and cohesion mechanisms had gained complexity and variety. The connection between paragraphs was mainly brought about using the full stop, which the pupil sometimes forgot to use, but in addition there were connectives with diverse semantic functions (porque, en cuanto a), and the advancement on the topic, mammals, was attained by repetition, references, and the organization of the text itself. Although both the organization and the internal references could be improved, the initial objective set, to write a text explaining to the third-grade schoolmates what mammals were, in the pupil’s opinion, had been met.
Conclusions The didactic intervention reported on here offers encouraging results for the improvement of student writing. After the intervention students became more aware of the expository type of text they were creating; their texts were better organized, with headings and sections; and they increased and appropriately used textual cohesion devices and explanatory resources, which they were already familiar with. Although the results are only partial, the initial pedagogic principles were supported, even in contexts where writing practices turned out to be very diverse. The key factor in the improvement was the intervention model used by way of constructing an expository text guided writing project through a five-stage didactic sequence.
Academic writing in Spanish education 195 Finally, the framework used throughout the research offers very interesting data in two fundamental areas. On one hand, there are challenges to be met in the field of teaching and learning academic writing that can be addressed through the design and elaboration of specific didactic materials for writing academic texts throughout the different levels of compulsory education (cf. García Parejo, forthcoming). On the other hand, these challenges also require teachers that are trained in this particular aspect, teachers who motivate and orient personal and group work, and create a climate that favors learning.
Appendix A Template for statistical analysis of expository texts T1 and T23 School________Text 1 ( ) Text 2 ( ) No. of paragraphs______ No. of words________ Mark the corresponding section for each case in the ExcelTM template. 1 Predominant subtype: a Description–definition b Classification–typology c Comparison and contrast d Problem–solution e Question–answer f Cause–consequence 2 Is there an adjusment to the communicative purpose? 3 Is the register appropriate to the communicative intention? 4 Is the topic contextualized? 5 Does the information contained in the text follow a general to particular organization, or vice versa? 6 Is there a predominance of data or facts over opinion? 7 Use of paratextual cues. Write how many times each cue appears. a Headings b Subheadings/sections c Pictures/illustrations d Graphs e Diagrams 8 Are metatextual organizers used (i.e., typographic resources that contribute to overall text organization)? 9 What is presented: a one paragraph with several ideas? b several paragraphs with the same idea? c several paragraphs with different ideas? 10 Are all of the paragraphs interconnected? In what way? 11 Type and number of predominant sentences
196 T. Á. Angulo and I. G. Parejo 12 13 14 15 16 17 18 19 20
21 22 23 24 25 26
Type and number of verbs in the text Type and number of subordinate clauses The connection between sentences is made by way of Is repetition avoided? How? Non-temporal present tense is predominant Specific adjectivization is predominant Textual modalizers are present Are any deictic expressions used? Explanations are carried out by way of: a Intradiscursive reformulation b Exemplifications c Enumeration d Comparisons e Introductory, explanatory colons f Explanatory parentheses g Explanatory “que” h Explanatory apposition Is the closure of the text made obvious in any way (summary, concluding section)? Are there agreement mistakes? Type Disfunctions are observed in Are there spelling mistakes? Point out the inappropriate uses of the following punctuation signs Are there any explicit references to sources of the documentation used by the student?
Notes 1 This chapter discusses the most relevant results of the project carried out by the Didactext Group Writing processes and the expository text in the improvement of written competence in pupils of 6th grade of elementary school. The group’s website can be seen at www.didactext.net. Members of the Didactext Group are Silvia Agosto Riera, Teodoro Álvarez Angulo, Teresa Chamorro García, Pilar Fernández Martínez, Isabel García Parejo, Carmen González Landa, Miguel Pérez Milans, Roberto Ramírez Bravo, Zahyra Camargo Martínez y Graciela Uribe Álvarez. We would like to thank Professor Aoife Ahern for her help with the English version of this document. 2 Semi-private schools are privately run schools, generally religious, which by way of an agreement with the State receive funding in exchange for offering free schooling. However, the school itself maintains the capacity to choose its own teaching staff and organize the students’ registration. 3 The entire template can be consulted in Didactext (2005, p. 139) and following. Here we only show the variables that have been analyzed within the text: 1, 7 and 20.
References Adam, J. M. (1990). Eléments de linguistique textuelle. Liége: Madaga. Álvarez, T. (2001). Textos expositivo-explicativos y argumentativos. Barcelona: Octaedro.
Academic writing in Spanish education 197 Álvarez, T. (2005). Didáctica del texto en la formación del profesorado. Madrid: Síntesis. Bakhtin, M. (1975/1989). Teoría y estética de la novela, Madrid: Taurus. Barton, D. (1994/1999). Literacy. An introduction to the ecology of written language. Oxford: Blackwell. Bazerman, C. (2006). Écriture, organization sociale et cognition. Pratiques. Université de Metz, pp. 95–115, 131–132. Beaugrande, R. de. (1984). Text production. Toward a science of composition, Norwood, NJ: Ablex. Beaugrande, R. de, & Dressler, W. (1997). Introducción a la lingüística del texto. Barcelona: Ariel. Bereiter, C., & Scardamalia, M. (1987). The psychology of written composition. Hillsdale, NJ: Erlbaum. Didactext, Grupo. (2003). Modelo sociocognitivo, pragmalingüístico y didáctico para la producción de textos escritos. Didáctica (Lengua y Literatura), 15, 75–102. Didactext, Grupo. (2005). Los procesos de escritura y el texto expositivo en la mejora de la competencia escrita de los escolares de sexto de Educación Primaria. Madrid: Editorial Complutense. Didactext, Grupo. (2006). Secuencia didáctica para la escritura de textos expositivos. Textos de Didáctica de la Lengua y la Literatura, 43, 97–106. Dijk, T. A. van. (1980). Estructuras y funciones del discurso. México: Siglo XXI. Flower, L., & Hayes, J. (1994). La teoría de la redacción como proceso cognitivo. Textos en contexto I. Los procesos de lectura y escritura. Buenos Aires: Lectura y Vida, 1996, 73–110. García Parejo, I. (2007). Del caos al orden: de cómo monitorizar el proceso de escritura en el aula. In Teodoro Álvarez Angulo (Ed.), La magia de las letras. El desarrollo de la lectura y la escritura en la educación infantil y primaria (pp. 155–197). Madrid: Ministerio de Educación y Ciencia. García Parejo, I. (Ed.). (forthcoming). Escribir textos expositivos en las aulas. Gee, J. (1990). Social linguistics and literacies (2nd ed., 1996). London: Falmer Press. Grabe, W., & Kaplan, R. B. (1996/1998). Theory and practice of writing. London: Longman. Hayes, J. R. (1996). A New framework for understanding cognition and effect in writing. In C. M. Levy & S. Randell (Eds.), The science of writing. Theories, methods, individual differences, and applications (pp. 1–27). Hillsdale, NJ: Lawrence Erlbaum. Nemirovski, M. (1999). Sobre la enseñanza del lenguaje escrito y temas aledaños. Maestros y enseñanza. (pp. 117–129). Barcelona: Paidós. Nystrand, M. (Ed.). (1982). What writers know: The language, process, and structure of written discourse. New York: Academic Press. Sanahuja Yll, E. (1992). El text expositiu a l’ensenyament secundari. Condicions de producció i suggerments didàctics. In Battaner & Sanahuja (Eds.), Saber de letra I (El texto escrito con finalidades académicas y comerciales en la enseñanza secundaria) (pp. 133–155). Barcelona: Universidad. Street, B. V. (1994). Cross cultural perspectives on literacy. In L. Verhoeven (Ed.), Functional literacy. Theoretical issues and educational implications (pp. 95–112). Amsterdam: John Benjamins. Wertsch, J. (1991). Voces de la mente. Madrid: Visor. Wray, D., & Medwell, J. (1991). Literacy and language in the primary years. London: Routledge.
15 Caught in the middle Improving writing in the middle and upper primary years Val Faulkner, Judith Rivalland, and Janet Hunter Edith Cowan University
Introduction Over the past 25 years in Australia, pedagogy and research about writing in the primary years have mainly focused on process writing and genre methodologies. Recent national benchmark data have raised concerns about the low quality of writing in middle and upper primary schools, which has led to ongoing problems in the secondary years. These data suggest, in fact, that many students reach a plateau in their writing development either before, or as, they make the transition from upper primary to lower secondary school. Given that much of the success of students in secondary school is measured by their ability to write effectively, it is imperative that there is further inquiry into improving the quality of student writing in Years 3–8. Recent writing literacy benchmark data demonstrated that whereas around 84 percent of children in Years 3 and 5 were meeting the benchmark standard, the levels flattened out, and in the case of Year 5 students, even fell slightly. Year 7 percentages have continued to fall steadily in the four years the data have been documented (Department of Education and Training, 2007). This chapter reports on a project that aims to build teacher capacity in assessing and teaching the linguistic, textual, and contextual levels of writing to students in Years 3–8, who are not meeting the benchmark standard. It has built on a pilot study funded by the Fogarty Learning Centre at Edith Cowan University. An extension of the pilot study throughout 2007 resulted in a collaborative arrangement between the Fogarty Learning Centre and the Association of Independent Schools of Western Australia (AISWA). This collaboration illustrates the power of productive partnerships in research with the education sectors and professional associations of Western Australia. We used a Formative Experimental Methodology (Jacob, 1992; Reinking & Bradley, 2004; Ivey & Broaddus, 2007), with the teachers as co-researchers, to develop a model of writing that provided teachers essential knowledge about what to assess in writing in order to support the further development of underperforming writers. Through the adoption of this experimental design, we began to develop an assessment model that helped teachers to more skillfully
Caught in the middle 199 analyze areas of weakness in student writing. As a result, we were able to link the assessment and teaching processes associated with writing in a way that supported a more targeted approach when working with students who were not meeting benchmark standards. This study became known as the Writing Project. The Writing Project sought to improve the effectiveness of the teaching of writing to middle to upper primary and early secondary students. The Western Australian Literacy and Numeracy Assessments (WALNA) data (Department of Education and Training, 2007) and National Benchmark data (MCEETYA, 2006) have indicated that the quality of writing in Australian schools is lower than it should be. When children enter secondary school without adequate written literacy skills, there can be a serious impact on the outcomes of their learning in all discipline areas and this, in turn, is highly likely to affect their life trajectories. Therefore, the children targeted in this project were those who: • • • •
would not write—the avoiders; had serious difficulties (such as an undiagnosed learning difficulty that manifests in bizarre spelling); did not like writing and did not make an effort (not engaged); could write but were not developing as writers when engaged in more complex tasks.
Theory These concerns around written literacy have been reinforced by current debates about the quality of writing demonstrated by students entering tertiary institutions, as reported in the National Literacy Inquiry (2005). Research has shown that writing is a complex cognitive activity that requires sustained, systematic, conscious, and ongoing effective teaching throughout the school years (Britton, Burgess, Martin, McLeod, & Rosen, 1975; Goodman & Goodman, 1979; Cambourne, 1988; Kellogg, 1994; Hayes & Flower, 1994; Deriewianka, 1990; Hammond & Deriewianka, 2001). As a result of the disappointing student outcomes reflected by state and national testing, and the increasingly complicated cognitive complexity of writing as students progress through school, it has become apparent that there is a need for more structured, intentional, and systematic written literacy interventions that can be sustained across the middle years of schooling (Years 3–8). The literature that has informed the teaching of writing in the Australian educational context has focused on a number of key theories. These theories can be grouped under three broad epistemological paradigms—the psychological (Vygotsky, 1978; Kellogg, 1994; Berninger, 1999; Hayes & Flower, 1994; Bereiter & Scamadalia, 1983; Britton et al., 1975), sociocultural (Rogoff, 1991; Smith, 1973; Clay, 1973; Graves, 1984; Goodman & Goodman, 1979; Cambourne, 1988; Lankshear, 1997; Street, 1995; Gee,
200 V. Faulkner et al. 1996; Kress, 1997), and linguistic (Myhill, 1999; Kress, 1994; Halliday & Hasan, 1984; Lankshear, 1997) perspectives. Different ways of thinking about the teaching of writing have arisen from these perspectives—expressive to transactional (Britton, 1970), the whole language or naturalistic approach (Goodman & Goodman, 1979; Cambourne, 1984), process writing (Graves, 1984) and genre (Halliday & Hasan, 1984; Deriewianka, 1990). Most of the ways that teachers have approached the teaching and learning of writing within the Australian context have been informed by these different pedagogical views, drawing predominantly from psychological and sociocultural paradigms and, in particular, the process writing and genre methodologies.
The Writing Project The Writing Project was based on the outcomes of a pilot study (Rivalland & Wooller, 2006), resulting in a focus on building teacher knowledge as a way of supporting teachers in their attempts at assessing, teaching, and improving student writing in Years 3–8. The pilot study was funded by the Fogarty Learning Centre and involved key researchers from the center together with four classroom teachers, from different Western Australian schools,1 who acted as co-researchers. The teachers involved in this initial study taught children aged 9–12 years. The main research activities undertaken with the teacher-participants involved the observation of the teaching of writing in their classrooms, as well as conversations about their planning and thinking when teaching writing. As a way of analyzing these data, a full-day discussion was held with both the teacher-participants and the researchers, where we looked at the issue of assessment of writing, and explored those areas of essential knowledge about writing that require explicit teaching. The data gathered from this small study suggested that teachers use a diverse set of practices which lead to a range of outcomes for students as writers. These discussions affirmed the need for the development of more effective assessment, planning, and teaching processes to support the effective teaching of writing. Emerging from the Pilot Study was a clear research question—How can teachers help lower-performing children to improve their writing so they will be able to meet the needs of secondary schooling? This question highlighted two areas that required further investigation and these were the drivers for the larger study conducted throughout 2007 and 2008 referred to as the Writing Project. These areas focused on building teachers’ knowledge about written language that is supportive of the student writer, as well as an approach to the assessment of writing that was more educative than that currently being used by the teachers. The second focus resulted in the development and trialling of an assessment proforma that emerged from the Composition Model of Writing.
Caught in the middle 201
Methodology The Writing Project developed around a Formative Experimental Methodology; a methodology designed to improve interventions/innovations (Patton, 1990; Jacob, 1992; Reinking & Bradley, 2004; Ivey & Broaddus, 2007), where the researchers make adjustments and changes to practice using the information gained from the field. In this case, the innovations and interventions were aimed at improving written literacy for children who were experiencing writing difficulties in the years of schooling spanning Years 3 to 8. The Formative Experimental Methodology entailed asking such questions as: “What is the problem?”; “What needs to be improved?”; “How can it be improved?”; “Has the improvement helped?” Integral to this methodology is that people can, and will, use information to improve practice and this was evident when conducting interviews with teachers who participated in the project over the life of the study. The methodology included a number of qualitative methods. These methods were teacher interviews, analysis of teacher discussion about teaching interventions linked to assessment, writing samples, and information gathered from the Assessment Proforma developed by researchers. This approach valued researcher–teacher collaboration, with its goal being to maximize educational benefits rather than to understand the current status quo. It emerged from the initial exploration of student writing samples and teacher responses that more attention needed to be given to linguistic understandings and knowledge about the consciousness of action necessary for writers to be able to develop the effective structuring of sentences, paragraphs, and whole texts, in order to compose complex texts. Linguistic understanding could include such things as sentence structure, the function of words, or an understanding of how to build cohesion within paragraphs or across the whole text. A greater consciousness of action would suggest that students need to be supported in their ability to deconstruct and reconstruct both exemplar texts, as well as their own writing. Through the deconstruction students and teachers need to be able to “zoom-in” (Anderson, 2006) and focus on one aspect of the text that requires targeted support. The larger study (2007/2008) recognized the need to find ways to engage and inform teachers in the middle years of primary school about such knowledge and skills. It also acknowledged that student capacity to write more complex texts is something that is required by the secondary curriculum and therefore must be supported by teachers across this phase of schooling. Research design The way in which the project was designed offered a logical timeframe, allowing researchers time to collect data and to present and discuss the findings with research participants and partners. Critical components of
202 V. Faulkner et al. the research process were: identifying a pedagogical goal, determining a range of instructional interventions that had potential to meet the pedagogical goal, identifying factors that inhibited or advanced the effectiveness of the interventions, and making modifications where necessary. Participants Participants in the project were 24 classroom teachers from 12 AISWA schools2 plus five classroom teachers from two Department of Education and Training (WA) schools.3 These teachers were concerned about the teaching of writing to children across the middle years of schooling and willingly chose to participate in the research. Additionally, some of these schools in this larger project were having more success than others in the results their students had achieved in state and national writing assessments.
Building teacher knowledge to support student writers Through a close analysis of the literature conducted by the researchers, and linked to discussion generated among teacher co-researchers, five grammatical dimensions were identified and considered to be areas of essential knowledge that support the teaching of writing. It became clear to both teachers and researchers that when composing text, an author or successful writer relies on the interplay of these grammatical dimensions. The dimensions are: • • • • •
topic knowledge, awareness of audience and purpose; genre; text coherence and cohesion; sentence construction (inclusive of Standard Australian English); vocabulary development and spelling fluency.
This interaction of grammatical dimensions helps the writer to make grammatical choices that will shape meanings to ensure that texts are coherent, purposeful, and engaging to read. This relationship has been illustrated through the use of a number of interlocking “cogs” that are in continual motion. The metaphor of “cogs in motion” is representative of an “expert writer” moving between, and within, the different dimensions of the model. It encourages teachers to understand the cognitive and interactive complexity of writing development. The model shows the importance of student writers playing with word choice, sentence structures, text organization, and content knowledge. When composing, writers adjust ideas by reshaping, cutting, pasting, adding, deleting, and experimenting with language. The interplay of the different dimensions represents the process that the writer engages in when crafting text. Within each of the grammatical dimensions there is a range of essential knowledge all novice writers should be exposed to. Figure 15.1 is a visual representation of the interlocking
Caught in the middle 203 Grammatical choices
Text coherence and cohesion
Vocabulary development and spelling fluency
Sentence construction and fluency
Topic knowledge, awareness of audience and purpose Genre
Grammatical choices
Figure 15.1 Composition model of writing.
“cogs” that underpin the Composition Model of Writing. The analogy of the “cogs” has emerged from a combination of an exploration of the literature, working with teacher-participants, and analysis of student writing sample data.
Model explanation Dimension 1: Topic knowledge and awareness of audience and purpose are central to the way writers use all of the other skills and processes when successfully composing written text. The three areas that make up this dimension act as a fulcrum for all the other aspects of composing. Dimension 2: Genre represents the way in which texts are organized to meet their social purpose and focuses on text organization and top-level structures. Dimension 3: Focuses on coherence and cohesion, that is, the way writing holds together to ensure that the whole text meets the genre requirements and has the logic needed to make it complete and comprehensible by the reader. Text coherence holds the whole text together to
204 V. Faulkner et al. provide connected meaning. Cohesion is created by language-linking devices that occur both between and within paragraphs, as appropriate for the text structure. These include: pronouns, ellipsis (leaving something out), substitution, lexical cohesion, building lexical chains (words that add to the meaning of the text and maintain a theme), conjunctions, paragraphs, and grammar. Dimension 4: Sentence construction and fluency is: marked by logic, creative phrasing, parallel construction, alliteration, and word order that makes reading feel natural. Fluent writing is free of awkward word patterns that slow the reader’s progress; instead, the language underscores the overall meaning of the piece, provoking the reader with a subtle road map. (Culham, 2003, p. 178) Sentence construction and fluency is underpinned by those foundational aspects of standard English usage that can help writers to be conscious avoiders of error (Emmitt & Pollock, 1999). Dimension 5: Vocabulary development and spelling fluency reflects the importance and power of words when composing text, and the capacity to use words without conscious effort to spell them correctly. Word choice is inextricably connected to “voice”: that which connects the reader to the text and establishes a relationship between the reader and the author (Spandel, 2005). Selecting the right words can draw the reader into the text by creating interest, building tension, generating excitement, or using humor. Control of words allows a writer to create visual images, stir emotions, and convey ideas with clarity and precision. Being able to draw on an extensive vocabulary allows writers to bring their text to life using strong verbs, powerfully descriptive adjectives and adverbs, and by using devices such as simile and metaphor. When readers have both an extensive vocabulary and can spell these words without conscious effort, they are able to concentrate most effectively on composing.
Linking the Composition Model to the assessment of writing The teachers in the larger study (2007/2008) felt they needed to know more about how to assess the more complicated grammatical dimensions of student texts represented through the Composition Model. This led to the development of an Assessment Proforma. The proforma was presented to the teacher participants for trialing in their classrooms and it became a primary tool for teacher reflection throughout the life of the project. Figure 15.2 is an example of the proforma that teachers were asked to use and in doing so, were able to follow the writing development of a number of their students. The rationale behind the Assessment Proforma was to help teacher-participants analyze student writing in far greater depth and, at the
Teacher Comments
Teacher Comments
Teacher Action/ Recommended Strategies
Teacher Action/Activities
Teacher Comments
Teacher Action/ Recommended Strategies
Question
Teacher Comments
Figure 15.2 Assessment proforma.
Using spelling analysis from Words their way: Which error patterns are evident? • Phonetic • Incorrect choice pattern • Incorrect use of conventions • Morphemic error • Severe problems
Are words specific, accurate and appropriate, according to needs of audience, purpose and topic?
Teacher Action/ Recommended Strategies
Step 3b: What grammatical choices have been made, and how effectively have they been used? Vocabulary Development and Spelling Fluency
Is cohesion maintained effectively by the use of lexical chains and reference files?
Are paragraph breaks effective and appropriate?
Do details seem to fit where they are placed through logical and effective sequencing?
Question
Step 3a: What grammatical choices have been made, and how effectively have they been used? Text Coherence and Cohesion
Is the text structured to meet its purpose, e.g. recount, narrative, report, description, explanation, exposition?
Question
Step 2: Has the appropriate genre been selected?
In narratives, is the reader’s interest maintained through selective disclosure, tension or humour?
Is there evidence that the writer has sufficient knowledge and/or experience about the topic?
Does the text match the purpose, audience and topic, e.g. describing, explaining, instructing, arguing or narrating?
Question
Step 1: Has the purpose/audience/topic been appropriately identified and demonstrated?
Assessment Proforma
Teacher Comments
Teacher Action/ Recommended Strategies
Teacher Comments
• Main teaching focus:
Step 4: Zoom in/zoom out according to steps 1, 2 and 3
Is there agreement between/consistency of– • Tense • Noun–pronoun • Noun–verb • Singular–plural • First, second and third person • Active/passive voice
Has punctuation (comma, apostrophe, colon, exclamation mark, quotation marks, semi-colon, hyphen, dash) been used appropriately and effectively?
Question
Teacher Action/ Recommended Strategies
Step 3d: What grammatical choices have been made, and how effectively have they been used? Standard Australian English Usage
Is there evidence that the writer is aware of how the language sounds?
Are sentences varied in length and structure in a way that is appropriate to the genre? • Simple • Compound • Complex • Compound-complex
Question
Step 3c: What grammatical choices have been made, and how effectively have they been used? Sentence Construction
Are nouns and modifiers used effectively to convey meaning?
Are verbs used effectively to convey meaning?
206 V. Faulkner et al. same time, facilitate the teachers’ understandings of the Composition Model. It was also considered important to develop a tool that promoted and encouraged teacher reflection and closer teacher scrutiny of student writing with the result that teachers’ actions were far more targeted to student needs. In fact it was hoped that the Assessment Proforma could become a tool for moving teachers’ thinking about their practices when supporting the student writer forward and for helping them to reframe, and even develop, new practices. The following section presents three case studies which share the experiences of teacher participants who trialed the draft project materials throughout 2007 and 2008. These cases demonstrate the impact of building teacher knowledge around the Composition Model and the value of more targeted assessment of student writing. They also begin to illustrate the impact of teaching that is more closely aligned to students writing “needs.” When the teachers are more aware of which grammatical dimensions require greater support, together with a heightened awareness of how this support should be developed, there is an impact on writing outcomes. It was evident from the feedback of the teachers involved in this study that they found the link between the Assessment Proforma and the Composition Model of Writing critical when endeavoring to support their students as writers. The following comments were made by the case study teachers: Catherine: “Useful as an evaluation tool.” Von: “Broke down different aspects of language.” Catherine: “Encouraged me to make suggestions of strategies that were more closely linked to the targeted areas of need.” Tessa: “Easily highlights areas of need and allows me to see the progression of student knowledge over time.”
Case study 1: Von Von was an early-career teacher working in an inner-city school in Perth, Western Australia. The school caters for a diverse student population, many of whom come from low socioeconomic standing. There were a large number of Indigenous students in her class. When joining this project she expressed three major concerns: How can I motivate the student who is reluctant to write? How can my teaching be more explicit when supporting student writers? What linguistic knowledge do I need in order that my teaching is far more effective? Von’s students were in Year 4 and 54 and a number of the children were reluctant to write at all. Compounding this problem were three students who were non-writers. Von found the Assessment Proforma extremely useful. It encouraged her to refer to the Composition Model when analyz-
Caught in the middle 207 ing student work and informed her choices when deciding on the teaching focus needed by her selected students. Eventually the Assessment Proforma became an integral component of Von’s analysis of student writing. After assessing the writing of the children in her class, Von decided that sentence construction was a major issue. As a result of exposure in the workshops to alternative ways to build linguistic knowledge with student writers, Von devised a number of teaching and learning “games”/activities that focused on engaging reluctant writers in sentence construction. While visiting her classroom, we saw a sentence-construction activity borrowed from the UK Literacy Strategy materials.5 The concretizing of writing tasks was a powerful pedagogical approach, as was the “talk” that surrounded these games, and both helped to build the metalanguage needed when discussing the composition of text. It was within this activity that we witnessed a number of obviously reluctant and challenging students actively and enthusiastically discussing word choices and word functions, constructing sentences, and then transcribing these sentences. We overheard children talking about needing a “verb” and a “conjunction” as well as “some punctuation to finish it off.” It was very obvious that Von had helped the children to develop a metalanguage for talking about the function of words and the ways in which sentences are constructed. It was also highlighted that often writing is seen as a quiet, individualized activity that shunned excessive talk, and the children in this classroom needed a more active approach when composing text. When discussing the writing outcomes with Von it was apparent that engaging in the assessment of writing that was systematic and linked to the areas of essential knowledge helped to increase her awareness of which dimensions required greater support. This resulted in a teaching emphasis that was far more focused, paying attention to the role of words in sentences, and the impact of these word choices on sentence construction.
Case study 2: Catherine Catherine taught a group of Year 4 and 5 students6 in a northern suburban school in Perth, Western Australia. Catherine’s school also had a very diverse student population. Her major concerns were: How can I motivate the students to write much more interesting sentences? What are a number of ways to encourage an interest in words? How can my assessment of writing inform my teaching for individual students? This teacher also used the Assessment Proforma to acknowledge what the students could do, as well as discovering which grammatical dimensions required additional support. The increased focus enabled her to make far more informed choices about what to teach. Information gained from the
208 V. Faulkner et al. targeted assessment encouraged Catherine to pay attention to the dimensions that dealt with sentence construction and fluency, as well as vocabulary development and spelling. However, like Von she also explored the function of words and how authors employed their knowledge of these word functions to achieve the greatest effects when engaging their audience. As a way of targeting the different levels of focus, Catherine adopted the pedagogical approach of “zooming-in and zooming-out” (Anderson, 2006) of exemplar texts. Exemplar texts are well-written and carefully crafted examples of writing by published authors. Catherine paid particular attention to the literary genre of “thriller.” Through the deconstruction of exemplar texts she was able to explain the impact of effective sentences on the reader, and how word choice had a bearing on that effectiveness. She demonstrated to the students what a well-written, attention-grabbing first sentence “looked like.” This led to the children exploring the construction of “fantastic” sentences and then generating their own. Another area of concern for Catherine was text coherence and cohesion. In the students’ writing of the “thriller,” the focus was on their ability to build tension as a way of engaging the reader. One technique she employed was to ask children to write shorter pieces of text and, in particular, deconstruct and reconstruct first paragraphs. Catherine used the knowledge gained through exposure to the different dimensions of the Composition Model to support her teaching which resulted in her refining and refocusing her teaching. In doing so she was able to improve the writing of the “thriller,” the narrative that the students were being asked to compose.
Case study 3: Tessa Tessa taught Year 7 in a large northern metropolitan school in Perth, Western Australia. Her school achieved relative success in the state literacy assessments, but there were concerns about the writing skills of the Year 7 students. When embarking on the project Tessa posed two focus questions: What makes a sentence a sentence? What do we mean by “powerful words” and why is it that they create a strong response and clear image for the reader? Tessa used the Assessment Proforma as a diagnostic tool. She found that sentence construction and fluency, as well as vocabulary development and spelling, appeared to be problematic, a common theme that was consistent across many of the participant schools. These findings reiterated that the students in a majority of the case study classrooms worked relatively successfully with whole-text structures, but needed additional support when working at the word, sentence, and paragraph levels. As a result of the analysis Tessa found that many of her students required constant scaffolding when constructing paragraphs and that this became more problematic when they were required to write far more extended text. Furthermore, she
Caught in the middle 209 was conscious that these children would soon be moving on to secondary school where the complexity of written tasks tended to increase. Through “zooming-in” (Anderson, 2006) and focusing on the construction of the paragraph, the students in her class became far more aware of the importance of this device as a way of building cohesion (Myhill, 2008). Prior to the emphasis on constructing paragraphs more carefully, many of the children did not use paragraphs at all, or only tended to use them as “graphic” organizers (Myhill, 2008). This ineffectual use demonstrated a limited understanding of the function of paragraphs, where they were often seen as a collection of white spaces between “chunks” of text. Tessa found that using the Composition Model as a way of understanding the Assessment Proforma ensured a more strategic approach in her teaching and this enabled her to begin to improve the writing outcomes of her students. It became evident that through “zooming-in” on the paragraph she encouraged her students to craft their writing of this aspect of their text far more carefully.
Conclusion This Writing Project research has highlighted a need to build teacher knowledge about language that supports the teaching of writing, as well as the assessment of writing. The chapter argues that it is necessary for teachers to develop a deeper understanding of a number of grammatical dimensions which are articulated through the Composition Model of Writing. Teachers need to build knowledge about words, sentences, and paragraphing as well as to improve their understanding of those linguistic devices that aid coherence and cohesion. In so doing, they will be able to support the struggling student writer in the middle and upper primary years of schooling with greater confidence. A second and equally important consideration is the role assessment plays in student writing. The Assessment Proforma became a tool for moving teachers’ thinking about their practices forward and for helping them to reframe, and even develop, new practices when supporting student writers. Its use reinforced that assessment needs to be far more targeted, enabling teachers to be more specific/ strategic/careful when deciding on their teaching focus. The outcomes of this research have begun to provide schools with an evidence-based research approach to teaching children in the middle and upper primary years who are underperforming in writing.
Notes 1 Perth, Western Australia. 2 Association of Independent Schools of Western Australia. 3 DET (Department of Education and Training), WA schools were invited to participate by the Fogarty Learning Centre researchers and were welcomed by Mr Ron Gorman, Literacy Educational Consultant, AISWA. 4 Children aged between 9 and 10 years old.
210 V. Faulkner et al. 5 Website: www.standards.dfes.gov.uk/primary/publications/literacy/63317. 6 Children aged between 9 and 10 years old.
References Anderson, J. (2006). Zooming in and zooming out: Putting grammar in context into context. English Journal, 95(5), 28–34. Bereiter, C., & Scardamalia, M. (1987). The psychology of written composition. Hillsdale, NJ: Lawrence Erlbaum Associates. Berninger, V. W. (1999). Coordinating transcription and text generation in working memory during composing: Automatic and constructive processes. Learning Disability Quarterly, 22(2), 99–112. Britton, J. (1970). Language and learning. Harmondsworth: Penguin. Britton, J., Burgess, T., Martin, N., McLeod, A., & Rosen, H. (1975). The development of writing abilities. London: Schools Council Publications. Cambourne, B. (1988). The whole story: Natural learning and the acquisition of literacy. Auckland: Ashton-Scholastic. Clay, M. (1972). Reading: The patterning of complex behaviour. Auckland: Heinemann. Culham, R. (2003). 6 + 1 traits of writing: The complete guide grades 3 and up. New York: Scholastic Inc. Department of Education, Science and Training. (2005). Teaching reading: Report and recommendation. National inquiry into the teaching of literacy. Canberra: DEST. Department of Education and Training. (2007). Performance of Years 3, 5 and 7 students, Western Australia: WALNA 2007. Perth, WA: DET. Derewianka, B. (1990). Exploring how texts work. Sydney, NSW: Primary English Teaching Association. Gee, J. P. (1996). Social linguistics and literacies: Ideologies in discourses. Philadelphia: Falmer. Goodman, K., & Goodman, Y. (1979). A comprehensive–centred whole language curriculum (Occasional Paper No. 1). Tuscon, AZ: University of Arizona. Graves, D. (1984). A researcher learns to write: Selected articles and monographs. Exeter: Heinemann. Halliday, M. A. K., & Hasan, R. (1985). Language, context and text: Aspects of language in a social-semiotic perspective. Burwood, Victoria: Deakin University. Hammond, J., & Derewianka, B. (2001). An introduction to genre. In D. Nunan & R. Carter (Eds.), The ELT handbook. Cambridge: Cambridge University Press. Hayes, J. R., & Flower, L. S. (1980). The dynamics of composing. In L. W. Gregg & E. R. Steinberg (Eds.), Cognitive processes in writing. Hillsdale, NJ: Lawrence Erlbaum Associates. Ivey, G., & Broaddus, K. (2007). A formative experiment investigating literacy engagement among adolescent Latina/o students just beginning to read, write and speak English. Reading Research Quarterly, 42(4), 512–545. Jacob, E. (1992). Culture, context, and cognition. In M. LeCompte, W. Millroy, & J. Preissle (Eds.), The handbook of qualitative research in education. New York: Academic Press, Inc. Kellogg, R. T. (1994). The psychology of writing. Oxford: Oxford University Press.
Caught in the middle 211 Kress, G. (1994). Learning to write. London: Routledge. Kress, G. (1997). Before writing: Rethinking the paths to literacy. London: Routledge. Lankshear, C. (1997). Changing literacies. Buckingham: Open University Press. Martin, J. (1985). Factual writing: Exploring and challenging social reality. Oxford: Oxford University Press. Ministerial Council on Education, Employment, Training and Youth Affairs. (2006). National report on schooling. Preliminary paper. National benchmark results reading, writing and numeracy Years 3, 5 and 7. Canberra: DEST. Myhill, D. (1999). Writing matters. English in Education, 33(3), 70–81. Myhill, D. (2008). Developmental trajectories in mastery of paragraphing: Towards a model of development. Unpublished manuscript, University of Exeter. Patton, M. (1990). Qualitative evaluation and research methods. London: Sage Publications. Rivalland, J., & Wooller, B. (2006). Writing in the middle years: “The more control we have of words, the more power we have in society as a whole.” Literacy Learning: The Middle Years, 14(3), 18–27. Reinking, D., & Bradley, B. A. (2004). Connecting research and practice using formative and design experiments. In N. K. Duke & M. H. Mallette (Eds.), Literacy research methodologies. London: The Guildford Press. Rogoff, B. (1990). Apprenticeship in thinking: Cognitive development in social context. New York: Oxford University Press. Spandel, V. (2005). Creating writers through 6-trait writing assessment and instruction (4th ed.). Boston, MA: Pearson Education Inc. Smith, F. (1982). Writing and the writer. London: Heinemann. Street, B. V. (1995). Social literacies. London: Longman. Vygotsky, L. (1978). Mind in society: The development of higher psychological processes. Cambridge, MA: Harvard University Press.
16 Teachers as mediators of instructional texts Suzie Y. Null Fort Lewis College, Durango, Colorado
Even though there may be debate about the extent, most educational researchers agree that curricular texts shape instruction. There is less exploration concerning how teachers shape instruction as they interact with texts, or concerning the impacts these interactions might have on students. Instructional texts interact within the sociocultural milieu of teachers’ values and decisions, the classroom culture that they co-create with their students, and within the school and community cultures. Within these activity systems, a teacher’s choices about using a composition instructional text can determine whether students learn one approach to composing a text or many, whether composition is a collaborative or independent process, and whether they learn to use organizational essay structures as adaptable heuristics or as rigid forms. In this case study, the two subjects, Ms. LaMotte and Ms. Olsen,1 used the same Step up to Writing composition curriculum but created separate enacted curricula in their classrooms.
Research on instructional texts and teachers’ instructional decisions Research on classroom instruction has often been separated into analyses of curricular content and analyses of teachers’ decisions and behaviors. But the two corpuses of research seldom overlap; even though teachers’ decisions are presumably framed around instructional texts, there is little research about how they use and shape the texts they use in their classrooms. In most classrooms, textbooks and other instructional materials provide the primary source of course content (Daniels & Zemelman, 2004). Textbooks create “action opportunity spaces” (Bazerman, 1999) that shape what is defined as knowledge, what activities students will do to learn this knowledge, and set up the assessment procedures that will keep the teachers and students accountable. Given the importance of texts in classroom activity systems (Bazerman, 2004, p. 319), an analysis of instructional texts’ content and implementation should be a critical element in understanding what is happening in classrooms. Yet currently there is little research on this topic.
Teachers as mediators of instructional texts 213 A separate strand of research has emphasized that teachers’ thoughts and actions often have the strongest effect on what and how students learn. Franzak (2008) argued that teachers are the true arbiters of “policy- in-action” (p. 469), filtering extant policies through their “collective networks of values, beliefs, and habits” (p. 470). Researchers have also found that teachers’ interpretations of their subject matter can be highly variable. Noting that “teachers teach who they are” (p. 121), Hawthorne (1992) described how four middle school teachers taught four different versions of “English.” She suggested that this variation might be more beneficial than forcing teachers to use a standardized curriculum that they were not motivated or comfortable enough to teach confidently. Complicating these findings are the relatively rare studies of teachers’ implementations of specific writing programs or techniques. Bratcher and Stroble (1993) wrote that it took teachers three years to fully integrate the techniques they had learned in their National Writing Project summer program, indicating that true curricular reform may be more time and labor intensive than reformers imagine. Raising a further complication, Whitney et al. (2008) found that teachers would use the same terms to describe the techniques they used to teach writing. “However,” the authors wrote, “teachers differed in how they used those strategies, talked about them, and built opportunities for students” (p. 205). In this sense, teachers’ adaptations of new techniques may have a greater effect on student learning than the technique itself. Given that there is little research on how instructional texts function in classrooms, or on how teachers mediate new texts or programs, this study is designed to investigate teachers’ interpretations and applications of an instructional text within their classrooms. This study focuses on two teachers’ uses of Step up to Writing, a supplementary composition text, rather than on the teachers’ uses of their textbook. The fact that both teachers were enthusiastic about Step up to Writing allowed this study to focus on teachers who were motivated to implement a new program.
Educational setting: Oak Grove Junior High Oak Grove Junior High is located between strawberry fields near the edge of a coastal city in southern California. In 2005–2006, its Grade 7–8 (12–14-year-old) student body of over 800 students consisted of 81 percent “Hispanic or Latino” students; 61 percent of its student body received a free or reduced lunch (an indicator of lower-income status); 28 percent were English Language Learners; and 74 percent of its eighth-grade students tested as below proficient on the state’s Language Arts test (as compared to 61 percent statewide) (Greatschools, 2006). The two teachers in the study, Ms. LaMotte and Ms. Olsen, said that Oak Grove’s students had a reputation for being “under-prepared” when they entered high school, and that the school’s faculty had been working hard to overcome this reputation. Raising test scores also were “in the
214 S. Y. Null back of everyone’s minds,” in Ms. LaMotte’s words, but she reported that the school administration was less focused on this and more interested in improving overall instructional quality. To these ends, the school had adopted several new programs, but it seemed to have left the details of implementation to the teachers.
Instructional text: Step up to Writing Step up to Writing had been adopted as the writing curriculum for the entire K–8 district. The program emphasized helping students develop and support main and subordinate ideas by using a stoplight metaphor in which students color-coded their ideas according to the following color scheme: Green: Go! State your topic. Yellow: Slow down! Add a reason, detail, or fact (R/D/F) to explain your topic. Red: Stop! Explain [and] give examples. Green: Go back! Restate your topic (Auman, 2003, p. 1-4). Most of the binder’s instruction centered on using this system to help students generate paragraphs that contained a green topic sentence, one or more yellow reason, detail, or fact (R/D/F) sentence(s), one or more red sentences with explanations, examples, or evidence (e’s), and a green summary sentence (Auman, 2003, pp. 2-9–2-11). While Step up to Writing focused on this basic form, it included extensions into other types of writing and also left teachers and students some flexibility in determining how many of each kind of sentence to write or how many paragraphs to include.
Method Research subjects Ms. LaMotte and Ms. Olsen were both Caucasian female eighth-grade Language Arts teachers. During the focal research in the 2005–2006 school year, Ms. LaMotte was beginning her seventh year of teaching, and Ms. Olsen was beginning her sixth year of teaching. Data collection While I had been collecting data on both teachers as part of a previous study (Blau, Whitney, & Cabe, 2006) during the 2004–2005 school year, this chapter relies primarily on data I collected independently in 2006– 2007. I did six classroom observations and interviews with Ms. LaMotte and four with Ms. Olsen. All of them followed an open-ended observation
Teachers as mediators of instructional texts 215 and interview protocol that focused on the teachers’ uses and perceptions of Step up to Writing, and on their other types of writing instruction. I used ethnographic field notes (Spradley, 1980) to record the data. I used grounded theory analysis methods (Strauss & Corbin, 1998; Charmaz, 1983) by analyzing my notes to generate categories of important actions, statements, and beliefs, which I continued to refine through further analysis. I continued my analysis by examining the categories and cases between the teachers and in relation to other categories.
Findings Both Ms. LaMotte and Ms. Olsen said that they were teaching Step up to Writing according to the lessons in the binder, without additions or elaborations. But both teachers made several conscious choices about which parts of the program to cover, how to present it to their students, and how to assess their students’ writing, and these choices affected the conceptions of writing and types of writing their students learned. Ms. LaMotte: a form for all purposes Uses of Step up to Writing In Ms. LaMotte’s class, the Step up to Writing paragraph form became the standard form for most writing and idea development throughout the year. Ms. LaMotte had previously emphasized literary analysis essays, which she said had been the focus of her undergraduate English major. As she began to use Step up to Writing, she continued to emphasize literary analysis and occasional personal essays. Ms. LaMotte primarily relied on three methods from Step up to Writing, and used the program’s techniques such as highlighting and paper folding to help teach these methods. In several lessons I observed, Ms. LaMotte used the “Name it/verb it/finish the thought” (Auman, 2003, pp. 3-4–3-7, 3-9) schema (which Auman intended as a topic sentence for summary writing) as a method to help students write opening sentences for essays. Figure 16.1 shows both Auman’s schema and an example of the type of sentence Ms. LaMotte’s students generated. Ms. LaMotte also used the “R/D/F & e” chart (shown in Table 16.1), which she adapted directly from Auman (2003, pp. 2-8–2-10), to help students develop and organize ideas. Students would fold their paper into two columns, write their topic sentences at the top and highlight them with a green marker, then write their “Reasons/details/facts” (R/D/Fs) in the left column and highlight them with a yellow marker, and their e’s—explanations, examples, evidence, or expert opinions (such as quotations)—in the right column and highlight them with a pink highlighter (pink was used in place of red). She had the class do this before every essay I observed. But she also sometimes did this with the class as a way of reviewing a text,
216 S. Y. Null Auman’s schema (3-4–3-9): Identify
Verb
Finish the thought
Ms. LaMotte’s use from Lowry’s The Giver : The book The Giver
describes
one boy’s experiences.
Figure 16.1 Ms. LaMotte’s uses of “Name It/Verb It/Finish the Thought.”
Table 16.1 An example of Ms. LaMotte’s “R/D/F & e” chart. Topic: Advice for Parents R/D/F (first) Parents should trust us more
e = to stay in the house by ourselves e = to get things done without being nagged e = believe us when we tell you something
R/D/F (second) They should never ignore what we think
e = give REAL consideration to our ideas e = sometimes do what we want e = admit that we might be right sometimes
R/D/F (third) They should respect our tastes and interests
e = don’t say bad things about our friends e = respect our music e = don’t tell us what we can or can’t wear
Note Throughout the chapter sentences that students would highlight in green are in bold, in yellow are underlined, and in pink are in italics.
even when it was not going to lead to a final essay, so students had extra practice using the chart to review, develop, and organize ideas. Usually, Ms. LaMotte would put the same chart on her overhead projector, and fill it out using ideas that she generated with input from her students. Early in the year, she would do the whole chart with her students. As the year progressed, she would do part of the chart with the class and have students finish it on their own. By the time students did personal essays at the end of the year she reviewed the chart and showed students how to adapt it to write personal essays, but allowed students to develop the entire chart on their own. Finally, students learned to transfer the information from their “R/D/F & e” chart into an “accordion paragraph” (Auman, 2003, p. 2-8). The following is an example of the type of paragraph her second-language learner students developed from the R/D/F and list of “e’s” from Table 16.1:2
Teachers as mediators of instructional texts 217 Parents should trust us more, even when they worry. They should allow us to stay in the house by ourselves. They should trust that we will get things done, and they shouldn’t nag us. When we tell them something, they should believe us. Any good relationship begins with trust. As the above example demonstrates, writing the paragraph often consisted of copying phrases from the “R/D/F & e” chart, putting them into sentence form, and listing the sentences. Ms. LaMotte also started to expand beyond the basic form, both by adapting it to other course assignments, and by teaching other techniques from the Step up to Writing binder. In the fall, she taught students to use the Step up to Writing “R/D/F & e” chart and paragraph form to write papers about mandela posters they had made on the themes, setting, characters, and important elements of books that they had read in small groups. On another occasion, Ms. LaMotte gave an overview of Auman’s “thirteen methods for writing a topic sentence” (2003, pp. 2-20–2-33) and had students write and share the different kinds of sentences. In the spring, students used the essay form more independently when they wrote personal essays. Instructional mediation Ms. LaMotte further mediated the curriculum through her teaching presentations and interactions with students. She made transparencies that were based on the program but were specific to her own lessons, created opportunities for guided practice, and sometimes even introduced props that she had devised. Verbally, she constantly added clarifications to the Step up to Writing format, asked the class questions, polled the class by asking them to raise their hands, used humor, shared personal examples, made analogies, and interacted with individual students both in front of class and as students worked independently. When students individually worked on finishing a chart or writing essays from their charts, they were allowed to help each other while she walked around the room and helped individual students. Student learning opportunities Ms. LaMotte wanted to teach students to use and adapt the formula in as many settings as possible, and to use it to develop ideas. Her grading emphasized having a main idea that was supported by details, with little focus on the particular order of the “yellow” R/D/F and “red” “e” sentences. As the year went on, students went from writing ideas from a class “R/D/F & e” chart, to adding their own ideas to the class chart, to developing their own personal essays. In this sense they had opportunities to expand their skills for developing ideas. On the other hand, students were
218 S. Y. Null not exposed to an incredibly diverse set of writing styles or techniques: the class mostly remained firmly grounded in essay writing, particularly in literary analysis, that relied on the Step up to Writing form. Effects on students Ms. LaMotte felt that Step up to Writing provided a concrete system and vocabulary for teaching students to organize and support their ideas, and that “it gave them knowledge about whether they were doing the basics right.” She cited an example of a student who had recently come from a group home in another city, who was quickly able to pick up the Step up to Writing outline concept and then join the class in the writing assignment. Ms. LaMotte also showed me examples of paragraphs, particularly from her second-language learners, in which their pre-Step up to Writing paragraph consisted of two or three sentences. After doing the “R/D/F & e” chart, the same students’ new paragraphs were longer and had more detail and support about the topic. My findings on what students learned are tentative in that I was not able to collect student writing and had to rely on my observations of students doing assignments. Her students appeared to be engaged during instruction and seemed to know how to do the assignment when they worked on their own. But on a few occasions, I saw students struggle with differentiating between a reason/detail/fact (R/D/F) and an example (e). For example, when I helped two students with their personal essays at the end of the year, they had written down their three explanations/examples (e’s), but then had difficulty writing a more general sentence that encompassed all of their details. Ms. LaMotte admitted that Step up to Writing did not have as much to offer her students who were already proficient writers, but said it gave them a framework they could use to organize an essay quickly. She felt that the program was beneficial for the majority of her students, and that she could help the minority who were ready to move ahead by working with them individually. Also, her interpretation of the program and flex ibility in assessment seemed to give proficient writers enough space to work within the form. Ms. Olsen: different forms for different genres Uses of Step up to Writing While Ms. LaMotte used Step up to Writing as the main prewriting and organizational form for most of her writing instruction that year, Ms. Olsen used the program as one of several forms that students applied to differing writing tasks. Ms. Olsen primarily used Step up to Writing to teach academic writing. To teach summary writing, she used the “Name it/ Verb it/Finish the thought” schema (see Figure 16.1). She used the Step up
Teachers as mediators of instructional texts 219 to Writing outline and accordion paragraph forms (see Table 16.1) to teach students to write a compare and contrast essay comparing themselves to either Anne or Peter in The Diary of Anne Frank or comparing Anne and Peter to each other. For this paper, which had student samples posted on the wall, she had required students to follow the accordion paragraph pattern of green topic sentence, and three yellow R/D/F sentences that were each followed by a pink explanation or example sentence, and ending with a green concluding sentence. The following is an example compiled largely from the verbal examples she gave her students as she taught the paper:3 A nne [ Frank] and Peter [V an Daan ] were very different from each other
One difference between Anne and Peter is their age. Anne is thirteen and Peter is almost sixteen. Also, Anne and Peter like to do different things. Anne likes to talk and Peter likes to spend time by himself. Anne and Peter have different personalities. Anne is loud and Peter is quiet. There are clearly many differences between Anne and Peter. The essays were graded according to how well the students had followed this form and highlighted with the appropriate colors. Ms. Olsen said that after the compare and contrast essay, they did an additional essay in which they used the accordion paragraph technique and then “stretched” it into a full paper. Uses of other forms Ms. Olsen used a variety of other writing styles and forms as she taught other writing genres. On most days I observed, she used “freewrites,” which usually involved students taking a few minutes to answer a specific question that reviewed the previous day’s lesson or previewed that day’s lesson. Early in the year, she had students write about their personal goals for the year. In March, I observed students writing notes from a section in their textbooks. In the spring, Ms. Olsen taught a mini-unit that she had designed on technical writing. In this unit, the class analyzed product manuals; then students collaborated in table groups, but individually wrote product manuals with four sections on their group’s household item. At the end of the year, she taught a poetry unit on ballads, odes, and elegies, and provided separate forms for these types of poems. For example, for the ballad assignment, students could lose points for leaving out verses, leaving out vocabulary words, not having a rhyme scheme, or labeling it incorrectly. In addition, Ms. Olsen assigned independent project choices such as bibliographies, persuasive essays, or independent choice projects to students who she felt needed more challenging work.
220 S. Y. Null Instructional mediation The type and feel of Ms. Olsen’s instructional mediation seemed to change somewhat with the writing assigned. Her method of presentation was fairly direct, in that she would often ask questions in order to prompt students to give the correct answer. Even more open-ended freewriting prompts often had questions designed to elicit specific answers, such as “What have you learned so far this year?” She similarly used the Step up to Writing model to direct student work: when discussing the compare and contrast essay on The Diary of Anne Frank, Ms. Olsen asked questions such as “what comes after [the] green [sentence]?” Since I observed the class where she had grouped several of the students who struggled academically, she also spent class time giving students opportunities to get help with revising their work for a higher grade; this involved working with smaller groups of students who were behind. The feel of the class changed when Ms. Olsen taught the technical writing unit that she had designed herself; students seemed to participate more readily, work cooperatively more productively, and seemed more engaged in critical thinking in response to more open-ended questions. Student learning opportunities Partially because of the grading system that often emphasized adherence to the prescribed form, there were not always significant opportunities for students to contribute and develop many of their own ideas, or for students to modify the forms they had learned. But by clearly structuring each assignment and providing examples, Ms. Olsen made writing expectations clear to her students and gave students examples that they could use to start writing. Ms. Olsen reported that when students got their papers back, they had a clearer understanding about why they lost credit. Effects on students Ms. Olsen’s uses of multiple forms of writing, while more diverse than Ms. LaMotte’s instruction, may have added confusion for some students, in that they may have learned that there were several non-interchangeable and apparently unrelated “correct” ways to write. Students did not seem to have opportunities to independently choose the most appropriate writing form for a new task. This may have created less self-efficacy among Ms. Olsen’s students who weren’t doing independent projects, in that they needed to wait for the teacher to tell them the correct form for a given assignment, and in that the form was not something they could readily guess or even easily remember. This possibly left them without a tool that they would be confident enough to apply and adapt on their own. My observations of students’ conceptual understanding of Step up to Writing were mixed. Many students seemed to be able to use and follow
Teachers as mediators of instructional texts 221 the program’s form, and for some it may have been reassuring to know the “correct” way to do the assignment. Others seemed to have difficulty separating main from subordinate ideas. On one occasion when I helped students who were supposed to be rewriting their compare and contrast essays on The Diary of Anne Frank, it appeared that some of the ones who were struggling had coded their sentences in the correct order of green, yellow, and pink without any apparent understanding of the difference between a main idea (green topic sentences), explanation of the main idea (yellow R/D/Fs), or subordinate detail (pink e’s). For example, Juan had listed several details (pink e’s), but had highlighted some of them in yellow in order to make them fit the correct order. When I explained that he needed a sentence that said how his “e” sentence related to his main idea, he seemed to understand somewhat, but still be confused about how to write a sentence of that type. Another student, Oscar, told me that he needed to rewrite. When I asked him what he needed to rewrite, he said, “the whole thing.” Even though I could see that he only needed to change the colors of a few sentences and write a few new sentences, he started recopying his outline. Ms. Olsen said some of the students who she considered to be the most proficient writers had lower than usual scores when she graded their essays on The Diary of Anne Frank. Ms. Olsen’s interpretation was that they hadn’t correctly followed the green–yellow–red sentence pattern. But the clarity of the grading system based rigidly on the Step up to Writing form also made it easier for her students to see why they had lost credit and to make the necessary revisions to raise their grades. Since she viewed her proficient students as fluent writers who still needed to work on organization, she felt that the program helped them see how and where their writing was disorganized. In this sense, her most proficient writers had opportunities to learn a specific organizational strategy to use in their writing, even though doing so may have temporarily suspended some of their writing fluency.
Discussion Both Ms. LaMotte and Ms. Olsen were excited that Step up to Writing would standardize their district’s writing curriculum, and they felt the improved consistency would benefit their students. But ironically, Ms. LaMotte and Ms. Olsen ended up teaching very different versions of the same program. Ultimately, their decisions and mediations of the Step up to Writing text had a greater impact on student opportunity than the instructional text itself. As Hawthorne (1992) argued, such instructional differences are not necessarily counterproductive. Ms. LaMotte and Ms. Olsen both made informed instructional decisions based on their conceptions of writing, their students, and their school context; both decided which curricula (out of the overwhelming array they were expected to implement) would be
222 S. Y. Null most useful to their students, and which elements of Step up to Writing would help students progress the most. Both were able to mesh the program with their individual strengths and teaching priorities, in ways that were meaningful within their classrooms. For example, Ms. LaMotte’s students received more class time and instructional depth for using and adapting the Step up to Writing essay form, which they primarily used to write about literature, while Ms. Olsen’s students were exposed to a greater variety of genres and learned to write using several different models. Ms. LaMotte showed her students how a single writing form can be generative and adaptable, while Ms. Olsen showed her students that all writing has a form even if the forms vary, and that students can be successful by learning and following each form. In Ms. LaMotte’s class, writing was a semi-collaborative process in which students often used all or parts of a chart that Ms. LaMotte had generated with the whole class, and then were allowed to help each other when they worked individually. In Ms. Olsen’s class, students also collaborated, but with the exception of the technical writing unit, the goal was more often to give the correct answer or use the correct form. While it is not realistic or desirable to expect these two teachers (or any two teachers) to teach the same strategies in the same manner, these findings still leave us to wonder how the differing conceptions of writing served their students in high school and beyond. Even though these findings are based on a small sample, they challenge some of the conventional assumptions of curricular reform. Those who develop instructional texts put great effort into creating materials to be used in classrooms, and states and districts spend a great deal of time deciding which texts to adopt. Apparently, their assumption is that the instructional texts drive the curriculum, and therefore changing the texts would change the curriculum. While instructional texts certainly provide the foundation of knowledge and activities that will be enacted in most classrooms, teachers’ decisions and actions build the actual structures of classroom life. If the true curriculum arises from the interactions between the teacher and the texts, then materials may not be the true locus of curricular reform. Instead true reform may happen as teachers find meaningful ways to integrate their materials into their classrooms.
Notes 1 All teacher, student, school, and place names have been changed. 2 Note: Sentences that students would highlight in green are in bold, in yellow are underlined, and in pink are in italics. 3 Note: Sentences that students would highlight in green are in bold, in yellow are underlined, and in pink are in italics.
Teachers as mediators of instructional texts 223
References Auman, M. E. (2003). Step up to Writing (2nd ed.). Longmont, CO: Sopris West Educational Services. Bazerman, C. (1999). Disciplined by texts: Textually mediated and textually structured activities in disciplinary classrooms and in disciplinary forms. American Educational Research Association, April. Bazerman, C. (2004). Speech acts, genres, and activity systems: How texts organized activity and people. In C. Bazerman & P. Prior, What writing does and how it does it: An introduction to analyzing texts and textual practices (pp. 309–339). Hillsdale, NJ: Lawrence Erlbaum. Blau, S., Whitney, A., & Cabe, R. H. (2006). Evaluating IIMPaC: Teacher and student outcomes through a professional development program in the teaching of writing. National writing project. Retrieved August 29, 2008, from www. nwp. org/cs/public/print/resource/2599. Bratcher, S., & Stroble, E. J. (1994). Determining the progression from comfort to confidence: A longitudinal evaluation of a National Writing Project site based on multiple data sources. Research in the teaching of English, 28(1), 66–88. Charmaz, K. (1983). The grounded theory method: An explication and interpretation. In R. Emerson (Ed.), Contemporary field research. Boston: Little Brown. Daniels, H., & Zemelman, S. (2004). Subjects matter: Every teacher’s guide to content-area reading. Portsmouth, NH: Heinemann. Franzak, J. K. (2008). On the margins in a high-performing high school: Policy and the struggling reader. Research in the teaching of English, 42(4), 466–505. Greatschools: The parent’s guide to K–12 success. (2006). Retrieved May 24, 2006, from www.greatschools.net/cgi-bin/ca/other/7691. Hawthorne, R. K. (1992). Balancing autonomy and obligation: The constellation of factors influencing curricular choice. In R. K. Hawthorne, Curriculum in the making: Teacher choice and the classroom experience. New York: Teachers College Press. Spradley, J. P. (1980). Participant observation. Stamford, CT: Thomson Learning. Strauss, A., & Corbin, J. (1998). Basics of qualitative research: Techniques and procedures for developing grounded theory, 2nd ed. Thousand Oaks, CA: Sage Publications. Whitney, A., Blau, S., Bright, A., Cabe, R., Dwar, T., Levin, J., et al. (2008). Beyond strategies: Teacher practice, writing process, and the influence of inquiry. English education, 40(3), 201–230.
17 Pushing the boundaries of writing The consequentiality of visualizing voice in bilingual youth radio Deborah Romero and Dana Walker University of Northern Colorado
Broadcaster: Do you see, where you need student: Yeah, from here (.) to here (.)
to take it out?
((points to sound waves on screen)) need to let people breathe, you need half a second, I think you left the right amount of space there.
Broadcaster: You
And so goes the conversation during a Language Arts youth radio program, as one middle school student works with a local broadcaster to edit the interview she had previously conducted and recorded. This dialogue illustrates not only how students were learning to read and see images of voices onscreen, but also how they were developing new literacy practices and ways of seeing (Berger, 1972). In this chapter, we explore how English-language learners (ELLs), through a dialogic process of writing and speaking for youth radio, came to visualize voice, both literally on screen and symbolically, as related to language practices and identity. We begin with an overview of the complementary theories on learning, language, and literacy that inform our conceptualization of the issues and research approach used to study youth radio as an educational intervention. Then, we introduce the classroom radio project, its specific design features, and pedagogical goals. Next, we describe our qualitative research methodology, followed by an analysis of collaborative activities in youth radio. We present examples of oral and written texts mainly from two students, Maria and Andres, to illustrate how students drafted scripts for radio, how voice came to be viewed as an artifact and object of activity, and to consider the consequentiality of viewing voice. The discussion explores further the significance of multimodal literacies, and implications that can be derived by pushing the boundaries of what counts as literacy in the classroom.
Pushing the boundaries of school with new literacies Our research on language and literacy learning is conceptualized through mutually informing theoretical perspectives. We adopt a constructivist view on learning, drawing on sociocultural and activity theoretical
Pushing the boundaries of writing 225 erspectives (Cole, 1996; Engeström, Miettinen, & Punamaki, 1999; p Wertsch, 1998), to underscore the significance of language, physical objects, and symbolic representations in mediating human activity and cognition. In line with social accounts of learning, we draw also on the work of the New Literacy Studies (Coiro, Knobel, Lankshear, & Leu, 2008; Barton, Hamilton, & Ivanicˇ, 2000; Street, 2003) to conceptualize the multiplicity of literacies that operate in the youth radio program and to acknowledge their socially and locally accomplished nature. These perspectives account for language and literacy as multifaceted, complex phenomena, in which written or text-based literacies are just one form of symbolic representation and, potentially, participants in any activity context can engage simultaneously with multiple modes of communication. Accordingly, in classrooms and in society, new modes of representation result in hybrid, nonlinear texts, embedded with digital images, and the page is often replaced with the screen and hypertexts (Kress, 2003). Ultimately, all communication involves some degree of multimodality, as is exemplified in much of today’s broadcast media. Not surprisingly, research in the area of youth media (de Block, Buckingham, Holzwarth, & Niesyto, 2004; Rubinstein Avila, 2006) often centers on these aspects, as well as issues of identity, power, and resource inequities (Hull & Zacher, 2004). In contrast, our research examines how students’ use of particular radio-based practices such as drafting and editing of written and oral texts reflected and shaped their language and literacy development. Specifically, we examine the process of editing and visualizing voice. There is, currently, scant research on the use of voice recording and editing software in language learning and teaching. Rather, that which exists centers on technologies associated with second- or foreign-language instruction or research (Chaudron, 2001; Salaberry, 2001), automated speech-recognition systems (Derwing, Munro, & Carbonaro, 2000), and students with learning disabilities (De la Paz, 1999). Closer to our focus of study, Emme (2001) has examined “the role visuality plays in critical thinking” (p. 57) (see also Berger, 1972), but of still more significance is a growing body of research and pedagogy dedicated to the use of digital technologies and the composition process (Costanzo, 2008; Kajder, 2004; Rance-Roney, 2008). This includes the aforementioned New Literacies and research on minority students’ literacies and access to technologies (Duran, 2008). Our research aims to contribute to these perspectives by exploring further the “complex relationships between the verbal and visual in communication and representation . . . by examining new literacy practices in real contexts” (Synder & Bulfin, 2008, p. 829). The youth radio project examined here required students to engage in a range of multimodal literacies, which often challenged the standard forms of school literacy (Gallego & Hollingsworth, 2000). The radio activities were embedded in a larger cycle of activity related to students’ research on their chosen topic of interest and a year-end research symposium (see Table 17.1). While different activities privileged different modalities,
226 D. Romero and D. Walker Table 17.1 Modalities, actions, and goals in the youth radio activity cycle Modality
Actions and Goals
1 Reading:
Research a topic of interest for final presentation – Abortion, immigration, runaways, the death penalty
2 Writing:
Draft questions to ask an “expert” – Work with high school mentors and radio personnel
3 Speaking:
Interview and record the interview with an expert – Police, immigration rights advocate, clinic counselor
4 Hearing:
Upload and review the interview on computer – Listen and analyze interviews with trainers, mentors, peers
5 Seeing:
Edit the audio recording of interview on computer – Understand and use media software
6 Realizing:
Record/burn the oral texts to CD – Broadcast and listen to each other’s interviews
7 Re-presenting:
Create multimodal research poster for schoolwide symposium – Students interviewed as knowledgeable experts on their topics
Note Bold indicates modalities that are of interest to literacies in this chapter.
reading to research or writing to draft questions, many stages drew on multiple modalities and representations. In an effort to rethink the traditional emphasis on reading and writing, we examine the interplay between speaking, hearing, and seeing that occurred in youth radio and the impact on students’ developing understandings of voice. Our analysis is focused on the physicality and consequences of seeing one’s voice as a visual and dynamic image. The findings suggest that the collaborative and multimodal activities afforded new literacy practices and contexts for meaning making and identity formation (Archer, 2006; Kress & van Leeuwen, 2001).
Methodology Research site and participants The youth radio project was developed in a bilingual middle school during school hours by the eighth-grade Language Arts and ESL teachers, in conjunction with the university researchers. This chapter presents findings from one academic year, 2006–2007. Participants included: two volunteers from a community radio station; 19 Mexican immigrant students (Grade 8, average age 14) designated as “struggling readers” and classified as English-language learners. In addition, eight students and their teacher from a local high school class, all trained in the use of the radio recording equipment and editing software, served as mentors to the middle school students. Guest speakers and community organization “experts” provided
Pushing the boundaries of writing 227 authentic community voices whom the middle school students interviewed and recorded, in their roles as reporters on their chosen topics. Data collection and analytic procedures We adopted a microethnographic perspective to study language and literacy as contextualized practices (Bloome, Carter, Christian, Otto & Shuart- Faris, 2005; Green & Wallat, 1981). We were especially interested in understanding how new types of collaborative activity (Engeström, 1987) that resulted from the intersection of classroom and community organizations redefined the relationships between the oral, written, and visual modalities. Qualitative data were gathered over the academic year, from weekly 90-minute sessions, using ethnographic field notes, two video cameras, and eight voice recorders placed strategically in the classroom. Two researchers participated in complementary roles. The first author, Romero, maintained largely an outsider’s perspective, made occasional site visits and provided feedback on activities and unfolding events in regular research meetings and conducted data analysis. The second author, Walker, was the site contact, having worked previously at the school; she developed mainly an insider’s perspective as participant observer, collecting data and analyzing the data along with the first author. Both researchers independently coded the data and compared their coding to enhance interrater reliability and the ongoing analysis. To identify key events, as well as transitions between activities, running records, data logs, and draft transcriptions were developed from over 25 hours of video and audio recordings of various literacy events (Barton et al., 2000) that involved interactions with written or visual representations, on page, on screen, or otherwise. Detailed transcriptions were developed, following conversation analysis and Jefferson’s transcription system (see Lerner, 2004) to represent the sequential aspects of ongoing interactions, including turns at talk, speech delivery, pauses and intonation, as well as nonverbal features and gestures. Although only brief examples of talk are presented here, for us, transcribing constituted part of the analytic process and revealed events as co-constructed phases of activity (Kelly, Crawford, & Green, 2001), to which participants oriented their actions. By analyzing the oral and written discourses, social actions, and the mediating artifacts employed in these literacy events, including the visual representations of voice, we explore the ways in which multiple literacies related to language and identities (Gee, 2001; Ivanicˇ, 1998). Additional data included over 50 artifacts: student drafts, written questions, prompt cards, audio recordings of oral texts, visual images of voice waves, and student evaluations and comments. These were purposefully sampled and analyzed in relation to the event maps, transcripts, and field notes, to inform our understanding of students’ visualization of voice and what counted as literacies in this context.
228 D. Romero and D. Walker
Findings Our presentation of findings elaborate on two main themes: drafting as a form of rehearsal and visualizing voice and its consequences, each connected by a subtheme, collaborative activity. As part of the larger cycle of activity of youth radio, students identified and researched a topic (examples included abortion, immigration, runaways, the death penalty, tattoos, and gangs), before planning an interview with a community expert. Subsequently, they recorded the interviews and uploaded them to a computer for editing. Once edited, the final oral texts were burned to a CD, played for the class, and for audiences at the research symposium. Drafting as rehearsal for self and other In preparing the interview, students had to define key phases and to draft possible questions to ask the expert on their chosen topics. To facilitate this process, teachers and the local radio broadcasters modeled how to conduct an interview, while students observed. Then in small group activities, high school mentors helped students to develop a draft script by asking them to consider what exactly they wanted to know from their expert. When students were not sure of the kinds of questions they might ask, as was often the case, the mentors would ask them questions about the topic, thus probing their recently acquired knowledge, while also modeling potential questions for them to consider including in the interview. Through this rehearsal, modeling, and dialogue, students developed draft scripts for the interviews. By writing down an interview script and questions students focused not only on the required language forms, but also on the functions that their turns at talk would accomplish. The scripts became mediating artifacts that were the basis for rehearsing the interview with radio personnel and high school mentors. For many students, the interview was a unique opportunity to take on new identities. Figure 17.1 illustrates Maria’s script for her interview with an immigration expert. Although seemingly elementary for an eighth-grade student, this was one of the more comprehensive and well-developed scripts. Analysis of the text’s formal features show how it constituted a dynamic, shifting representation of ideas that were to guide her interview; the deletions, scratches, arrows, and sequential organization on the page capture this, and also suggest that for Maria, like for many of the English- language learners in class, committing her ideas to writing was no simple task. Analyses of the discursive and functional aspects of the text reveal it to be a carefully designed tool, a written prompt, for mediating subsequent oral interactions. In this way, the draft represented an understanding of the interview as a discourse genre (Wells, 2007): how a radio interview ought to be structured, planning for the opening and closing turns, and attention to introductions—“Hi my name is Maria . . .”—and closure—“Thank you for coming.”
Pushing the boundaries of writing 229
Figure 17.1 Maria’s draft text for immigration interview.
In terms of identity work, the draft questions can be seen as representing the multiple identities that Maria was simultaneously negotiating. By drafting the written questions, Maria was not only “being a student” and embodying the academic task, but she also revealed—with some risk to her and her family’s security—her social identity as an English-language learner and immigrant: “I chose this topic because my family immigrated here to have a better life.”
230 D. Romero and D. Walker Just as the scripts were developed in collaboration with others, the interviews were, frequently, collaborative endeavors. Maria’s interview was conducted with another classmate, Andres, and together they carefully orchestrated the opening, seamlessly moving between English and Spanish, with only a slight adaptation of the original script: Maria: Hi, my name is Maria. andres: Y, mi nombre es Andres. Maria: And, our topic is immigration.
Thus, scripts served as a material pivot; a linguistic and procedural scaffold, which visibly transitioned students from the conceptualization of ideas about their topic into the realization of a face-to-face interview with an expert. Moreover, the written texts were one modality among many symbolic and technological artifacts—including microphones, headphones, audio recorders, and prompt cards—that mediated the interview process. In turn, the interview yielded another product, a recorded audio file, which became the focus of still subsequent interactions. Visualizing voice as a shared object of activity While the drafting of written texts, a somewhat familiar and school-like activity, helped students prepare and transition into the interview process, the subsequent focus on oral language and the simultaneous attention to the mediating technologies was for many students a new and stimulating experience. In order to conduct and record the interviews, students worked in small groups. Typically, one student wore the headphones and held the microphone, another would hold a voice recorder and help with the prompt cards and scripts, and a high school student would monitor the equipment and sound and voice quality during recording. All middle school students would pose and answer questions. Speaking and hearing one’s voice, especially as it fed back through the headphones, provided unique opportunities for students to rehearse new identities as novice reporters. In addition to the multimodal technologies implicated in youth radio, voice became a primary artifact, mediating the interactions. The resulting attention to voice instigated new challenges and many students, who were self-conscious of their limited English and their developing language skills, were reluctant to listen to themselves and for others to listen to them. Students commented, “It was embarrassing for me, for people to hear me talking” or “It was difficult to record when you [can hear yourself] talk.” However, an important turning point emerged as students transitioned, from using their individual voice in the interviews, to collaboratively working on their voices in the editing process. In contrast to speaking and hearing their voice during the interview, the editing process and the use of specialized audio software provided students with a means for hearing, and more significantly, for seeing their voice
Pushing the boundaries of writing 231 after the event, as a visual representation on screen. In this sense, voice shifted from being a primary object, directly mediating the interactions, to a secondary artifact that became the object of their actions and a tool in subsequent interactions. As students learned to manage the voice-editing software, through instruction and collaboration with the radio broadcasters, the high school teacher, and mentors, they also learned to see, to talk about, and to manipulate the image of their voice on screen. Voice was no longer something personal or intangible, but literally it was transformed into a publicly visual artifact, assuming physical and spatial properties. Consequently, this onscreen objectification of voice in the spectrograph image, much like the written drafts on a page, constituted a socially shared object; one which could be located, paused in time and pointed to, and which was available for discussion and deliberate, collaborative activity. Moreover, this visualization of voice was consequential for both literal and figurative transformations of self and identity. Through instructional conversations (Wells, 1999) with the radio personnel and mentors, students soon learned to recognize and “read” how the sound waves visually reflected volume; larger waves represented louder voice, small waves for soft voice, and no waves represented silences. However, unlike their handwritten drafts, in which the grapho-tactile representations seemingly embodied their personal identities and style, the sound waves on screen transposed and repositioned the individual’s identity and voice, in both form and functionality, into a digital depiction. Accordingly, this functionality enabled students to edit their voices and those of the experts they interviewed. As students became more proficient in “reading” the sound waves, they also developed an ability to reflect and critically act upon prior oral productions. The consequentiality of viewing voice By viewing voice in images onscreen, students came to realize that the spoken interview, like the scripts and written texts, was also a draft of sorts, a work in progress. Although instructional time constraints and availability of experts meant that there were limited opportunities to redo the interviews, some students did choose to redo small segments, or restate questions during the interview. But for the most part, students conducted an uninterrupted interview, knowing that it could, and probably would, be manually edited post hoc. During the process of editing the audio and visual images, students realized that speech is not perfect, that even experts and first-language speakers make mistakes; false starts, repetitions, overlap, and pauses were all identified. From our vantage point, when students were able to comprehend how they could modify their voice and others’, even after the speaking had happened, they were afforded the possibility of taking risks and creating new, redefined representations of themselves and their oral language, vis-à-vis their peers and other speakers. Several students commented positively about the editing process, saying they enjoyed
232 D. Romero and D. Walker being able to “delete things you don’t like” and “that if you make a mistake [speaking] you can actually erase it!” Others confirmed that they learned that “you can record and edit yourself.” Consequently, students were able to reconstruct and realize their voice in new ways: stretching the waves on screen to make it as loud as other voices, or deleting sections of the image to remove errors and clean up the informalities of spoken language. Furthermore, by attentively and repeatedly listening to and viewing their own oral performances, students heightened their awareness of various linguistic and meaning-making practices. For example, they reflected upon their successes and identified the moments in which they effectively posed follow-up questions, or offered critical answers to questions experts asked of them. As Maria and Andres listened and edited together their immigration interview (based on Figure 17.1), Andres’ voice was heard asking: “Y qué pueden hacer los inmigrantes para seguir viviendo en los Estados Unidos?” (What can immigrants do to continue living in the United States?). The recording continued in Spanish, and Andres suddenly interjected, “Oh, remember that question?” “Which one?” asked Maria, still in Spanish. And, he proudly replied, “the one that I did well!” Through the collaborative analysis and action upon oral language, mediated by the audiovisual representations, students came to appreciate the subtleties of how meaning is made and understood. Furthermore, by working on both form and function, students were able to assume new identities as competent speakers, as reporters and editors: they came to realize the power of voice. In closing, we consider further the consequentiality of this realization for language and literacy development.
Discussion In this chapter we examined students drafting, viewing, and editing voice and considered how these resulted from meaningful engagement in a school-based youth radio program. We presented examples of multimodal and collaborative literacies, and the kinds of artifacts that were invoked, developed, and taken up, and reflected upon the consequences of working with multimodal texts and the realization of voice. In this section, we discuss, briefly, the theoretical and practical implications that can be derived from this work. Manifest in these activities is the power of voice and its unique status and varying functions, both literally, as the physical sounds and expression of oral language, and symbolically, as one’s right to an opinion. Thus, for the radio broadcasters voice was the vehicle for communicating and sharing important issues. Voice quality was important, but since radio technologies meant it could be modified, the emphasis when doing radio was on the broadcast message. An example from a collaborative editing session emphasizes this point, when the broadcaster explained, “I’m gonna use the zoom function, I’m gonna make this louder ’cuz that question is important.”
Pushing the boundaries of writing 233 For the teachers, voice was one of the goals of the ESL classroom and was linked both to improving students’ oral proficiency and their competence while speaking, and to raising awareness of certain social issues, in this sense, as a form of empowerment through critical literacy. One teacher commented to a group of girls, after their interview on abortion, “You took—to me—a real leadership role! So, I saw something in you . . . I’ve seen it before . . . tomaste responsabilidad y: it was good to see. Como un lider.” For the students, voice and its transformation into a physical object in youth radio generated a new potential for meaning making, and for acknowledging their voices (and silences) as manifestations of their identities and membership in various discourse communities (Walker & Romero, 2008). In closing, we consider how these new literacies and youth radio practices afforded participants innovative possibilities for empowerment. We attempt to illustrate this dialogic process as the realization of voice, in Figure 17.2. The interlocking circles depict the reflexive nature and interaction between the different modalities in youth radio. Similar to Wells’ (1999) “the spiral of knowing,” knowing is reached through the individual’s “personal experiences, which, amplified by information, is transformed through knowledge building into understanding” (Wells, 1997, p. 85). Thus, students experienced and realized their voice by speaking and hearing, but they also reached a heightened sense of awareness and
Speaking one’s voice
Hearing one’s voice
Realizing one’s voice
Realizing one’s voice Editing one’s voice
Seeing one’s voice
Figure 17.2 The realization of voice through multimodal interactions.
234 D. Romero and D. Walker understanding about voice and identity, by seeing and editing voice, amplified through technology as an onscreen object. This ability to read and edit the words of their scripts and interviews became part of the knowledge- building process, shaping their ability to read the world, the new tools and representations of voice that emerged, and ultimately, invoking new ways of knowing. Reading and editing the images of voice seemingly repositioned speech and writing. In youth radio practices, writing preceded speaking, marking a contrast to historical, pedagogical, and often common-sense notions, where traditionally orality precedes literacy (Goody, 1968; Ong, 1982). Even though students wrote a range of written texts (drafts, note cards, scripts on pages and onscreen), these served, primarily, as functional organizers, precursors for developing oral language. Although an interrelationship between oral and written language has long been acknowledged (Ferreiro, 1986; Olson, 1977, 1991; Scribner & Cole, 1981), it seems that secondary classrooms have been slow to incorporate both modalities into meaningful learning. We hope the findings presented here illustrate what can be accomplished, when integrated multimodal learning occurs in classrooms. The reflexivity between modalities was consequential, not just for the interview and the assignment, but also because the realization of voice contributed to an ongoing process of identity crafting. While youth radio obliged the speaker as interviewer to hear him/herself speaking, it was the necessary editing of oral texts that provided for a focus on voice. Doing so in such a way repositioned the speaker as both listener and as editor, and enabled students to re-present an alternate or modified voice. Consequently, students were able to contest the traditional aspects of oral language and its ephemeral deployment. In fact, when voice was represented, both physically and symbolically via the radio technologies, on screen and in the air (over the school PA system), the potential for transformation was immense. In line with Wartofsky (1979) and others, we contend that the ability to represent an action by other symbolic means, in this case visualizing one’s voice in an image onscreen, generates a distinctive class of artifacts or representations, which in turn leads to new possibilities and subsequent interactions. We hope to have shown that when classrooms dare to push the boundaries of what counts as writing, literacies, and meaning making, to include new modalities and forms of expression, students are also able to push their own boundaries of language, literacy, identity, and voice.
Acknowledgment The authors wish to acknowledge partial support for this research by grants received through the Diversity in Education Research Initiative (DERI), University of Northern Colorado. Earlier analyses based on this research were presented at the British Association of Applied Linguistics
Pushing the boundaries of writing 235 Conference, University of Edinburgh, 2007 and Writing Research Across Borders, University of California, Santa Barbara, 2008.
References Archer, A. (2006). A multimodal approach to academic “literacies”: Problematizing the visual/verbal divide. Language and Education, 20(6), 449–462. Barton, D., Hamilton, M., & Ivanič, R. (2000). Situated literacies. Reading & writing in context. London: Routledge. Berger, J. (1972). Ways of seeing. London: BBC and Penguin Books. Bloome, D., Carter, S. P., Christian, B. M., Otto, S., & Shuart-Faris, N. (2005). Discourse analysis and the study of classroom language and literacy events: A microethnographic perspective. Mahwah, NJ: Lawrence Erlbaum. Chaudron, C. (2001) Progress in language classroom research: Evidence from The Modern Language Journal, 1916–2000. The Modern Language Journal, 85(1), Special issue: A century of language teaching and research: looking back and looking ahead, part 2 (Spring, 2001), 57–76. Coiro, J., Knobel, M., Lankshear, C., & Leu, D. (Eds.). (2008). Handbook of research on new literacies. New York: Lawrence Erlbaum Associates. Cole, M. (1996). Cultural psychology: A once and future discipline. Cambridge: Cambridge University Press. Costanzo, W. (2008). The writer’s eye: Composition in the multimedia age. Boston, MA: McGraw-Hill. de Block, L., Buckingham, D., Holzwarth, P., & Niesyto, H. (2004). Visions across cultures: Migrant children using visual images to communicate. Brussels: European Commission. De La Paz, S. (1999). Composing via dictation and speech recognition systems: Compensatory technology for students with learning disabilities. Learning Disability Quarterly, 22(3), 173–182. Derwing, T. M., Munro, M. J., & Carbonaro, M. (2001). Does popular speech recognition software work with ESL speech? TESOL Quarterly, 34(3), 592–603. Durán, R. (2008). Technology and literacy development of Latino youth. In L. Parker (Ed.), Technology mediated learning environments for young English learners. New York: Lawrence Erlbaum. Emme, M. (2001). Visuality in teaching and research: Activist art education. Studies in Art Education, 43(1), 57–74. Engeström, Y., Miettinen, R., & Punamaki, R.-L. (Eds.). (1999). Perspectives on activity theory. Cambridge: Cambridge University Press. Ferreiro, E. (1986). Literacy before schooling. Oxford: Heinemann Press. Gallego, M., & Hollingsworth, S. (2000). What counts as school literacy. Challenging the school standard. New York: Teachers College Press, Columbia University. González, N., Moll, L. C., & Amanti, C. (Eds.). (2005). Funds of knowledge: Theorizing practices in households, communities, and classrooms. Mahwah, NJ: Lawrence Erlbaum. Goody, J. (1968). Literacy in traditional societies. Cambridge: Cambridge University Press. Green, J., & Wallat, C. (1981). Ethnography and language in educational settings. Norwood, NJ: Ablex. Gutierrez, K. (2001). What’s new in English language arts: Challenging policies and practices, y qué? Language arts, 78(6), 564–569.
236 D. Romero and D. Walker Hull, G., & Zacher, J. (2004). What is after-school worth? Developing literacy and identity out of school. Voices in Urban Education, 3. Kajder, S. (2004). Enter here: Personal narrative and digital storytelling. The English Journal, 93(3), 64–68. Kelly, G., Crawford, T., & Green, J. (2001). Common task and uncommon knowledge: Dissenting voices in the discursive construction of physics across small laboratory groups. Linguistics and Education, 12(2), 134–175. Kress, G. (2003). Literacy in the new media age. London: Routledge. Kress, G., & vanLeeuwen, T. (2001). Multimodal discourse. The modes of media and contemporary communication. New York: Oxford University Press. Lerner, G. H. (Ed.). (2004). Conversation analysis. Studies from the first generation. Amsterdam/Philadelphia: John Benjamin. Olson, D. (1977). From utterance to text: The bias of language in speech and writing. Harvard Educational Review, 47. Olson, D. (1991). The world on paper. Cambridge: Cambridge University Press. Ong, W. (1982). Orality and literacy: Technologizing the word. London: Routledge. Rance-Roney, J. (2008). Digital storytelling for language and culture learning. ET Essential Teacher, 5(1), 29–31. Rubinstein Avila, E. (2006). Publishing “equinox”: Broadening notions of urban youth development after school. Anthropology & Education Quarterly, 37(3), 255–272. Salaberry, M. R. (2001). The use of technology for second language learning and teaching: A retrospective. The Modern Language Journal, 85(1), 39–56. Scribner, S., & Cole, M. (1981). The psychology of literacy. Cambridge, MA: Harvard University Press. Snyder, I., & Bulfin, S. (2008). Using new media in the secondary English classroom. In J. Coiro, M. Knobel, C. Lankshear, & D. Leu (Eds.), Handbook of research on new literacies (pp. 805–838). New York: Lawrence Erlbaum. Street, B. V. (2003). What’s “new” in New Literacy Studies?: Critical approaches to literacy in theory and practice. Current Issues in Comparative Education, 5(2), 77–91. Walker, D., & Romero, D. (2008). When “literacy is a bennie”: Researching contested literacies in bilingual youth radio. Ethnography and Education, 3(3), 283–296. Wartofsky, M. (1979). Perception, representation, and the forms of action: Towards an historical epistemology. In R. S. Cohen, D. Davidson, G. Nuchelmans, & W. C. Salmon (Eds.), Models: Representation and the scientific understanding (Vol. XLVII, pp. 188–210). Dordrecht: D. Reidel Publishing. Wells, G. (1999). Dialogic inquiry: Towards a sociocultural practice and theory of education. Cambridge: Cambridge University Press. Wells, G. (2007). The mediating role of discoursing in activity. Mind, Culture and Activity, 14(3), 160–177. Wenger, E. (1998). Communities of practice: Learning, meaning, and identity. Cambridge: Cambridge University Press. Wertsch, J. V. (1998). Mind as Action. New York: Oxford University Press.
18 Classroom teachers as authors of the professional article National Writing Project influence on teachers who publish Anne Whitney Pennsylvania State University
Since at least the early 1990s, when action research conducted by teachers began to become widespread in the United States as a form of professional development for elementary and secondary school teachers, we have had in the fields connected to field of Language Arts education (such as composition, reading, and English education) a sense that classroom teachers can and perhaps should be active contributors to the published literature of the field. While barriers to teacher publication surely exist, classroom teachers are indeed writing about their classroom practice and publishing articles about their work in US literacy-education journals like English Journal or Language Arts. Further, the consensus among scholars working in teacher education in language and literacy is that teachers benefit from engaging in writing for publication and perhaps that it is not just an outcome of their professional development but also a factor in that development. In other words, the field takes some pride in having the voices of classroom teachers mixed into its scholarly literature and is interested in encouraging those voices. We also know that, in many cases, the publication activities of teacher-authors are deeply influenced by their participation in teacher networks (such as the National Writing Project, the Bread Loaf Teacher Network, local critical friends groups, or other networks). However, there is only a little empirical evidence about how those voices develop and are situated among teaching peers and among the range of other voices contributing to that literature, and even less evidence is available about the role teacher networks might play in the publication activities of teachers. This chapter seeks to add to the available empirical information about teacher publication in Language Arts through a study of published teacher-authors in the context of their participation in a professional development network, the National Writing Project. I explore what such teachers perceive as the challenges to their efforts in writing for publication and how the NWP context, while not explicitly designed for the purpose of promoting teacher publication, offers teachers resources for mitigating those obstacles. The National Writing Project (NWP) describes itself as “a professional development network that serves teachers of writing at all grade levels, primary through university, and in all subjects” and is dedicated to
238 A. Whitney “improving the teaching of writing and improving learning in the nation’s schools” (National Writing Project, 2008). NWP is hosted in over 200 local sites at colleges and universities in all 50 US states, the District of Columbia, Puerto Rico, and the US Virgin Islands. While site programs vary locally, two central activities for all sites include (a) invitational summer institutes where teachers share successful practices, write, and share writing with colleagues, and (b) partnerships with local schools to provide professional development programs for educators. All of the NWP’s activities are based on a teachers-teaching-teachers model and a core principle that teachers of writing need to have the experience of writing themselves. The NWP is the only federally funded program in the United States addressing the teaching of writing, and since its genesis in 1974 it has enjoyed immense popularity. Its influence on writing pedagogy has been tremendous, particularly in the US context: in its early years, it was central to the development and spread of process-oriented views of writing pedagogy; its teachers-teaching-teachers model of professional development has been taken up widely across disciplines; and today, NWP directly reaches approximately 100,000 teachers per year and influences many more through inservice work conducted by those teachers in local schools. Participation in NWP programs has also led to a large number of publications by classroom teachers, both in the NWP’s own in-house publications and in publications of the field such as English Journal or other journals for Language Arts teachers and scholars. These publications constitute part of a larger body of journal literature written for teachers, by teachers, the authors of which include not only NWP teachers, of course, but also teachers working in teacher-research groups, those enrolled in graduate programs, and a range of other teachers who for one reason or another have elected to frame their classroom work for the wider audience of the field. This is perhaps particularly pertinent in the field of Language Arts, where an ethic of the teacher as writer has pervaded public professional discourse for decades (see, for example, the discussion in English Journal in response to 1990 article by Karen Jost, who argued against the notion that teachers should write and in so doing inspired a heated and prolonged exchange). And in fact, “teachers of writing should also write” is a “basic assumption” of the NWP that informs all of its programs. Thus it appears that NWP-affiliated teachers would be especially well positioned to take up the task of writing for publication. As such, I focus on that group here, investigating both their activities in publishing and the connections between those activities and the NWP context. This research is contextualized in a set of related literatures. First, there is a practical literature for teachers who would like to write for publication (e.g., Dahl, 1992; Henson, 1984, 1986, 1988, 1990, 1993, 1995, 1997, 1999a, 1999b, 2001, 2003, 2005) and for those who would hold writing workshops for teachers (e.g., Root & Steinberg, 1996). Second, there has been a small amount of empirical work focusing mainly on showing how
Classroom teachers as authors 239 the conventions of the academic article can be barriers to would-be teacher-authors (Burton, 2005; Casanave & Vandrick, 2003; Fecho, 2003). My work also draws on a small but growing body of research looking at the effects that writing in professional development might have on classroom practice, extending studies of the National Writing Project and other writing-based professional development sites that have highlighted the recursive relationship between writing and its teaching (Lieberman & Wood, 2003; Perl & Wilson, 1986; Sunstein, 1994; Whitney, in press). This research is also in dialogue with the large body of work on the teacher-as-researcher or as a producer of disciplinary knowledge (Bissex & Bullock, 1987; Cochran-Smith & Lytle, 1993; Goswami & Stillman, 1987). In recent years, a slice of the scholarship on teacher research has begun to focus on the representation of classroom practice as a form of publication, whether through a presentation or a website or an article, building on and looking at the implications of developing such a representation for the teacher—both for its potential influence on the practice being represented but also for how a teacher positions him/herself and is positioned relative to colleagues and the field (Hatch, 2005; Stock, 2001, 2005). These scholars reason that writing (or drafting a presentation or other public representation of practice) affords opportunities for teachers to become aware of, reflect on, and refine the nuances of teaching practice. The data I present here were collected as part of a larger study undertaken by NWP known as the Legacy Study, in which 2,114 teacher- consultants who had participated in an invitational institute between the years of the first institute in 1974 and 1994 were contacted for a survey; of the 1,848 cases, a subset of 110 randomly selected participated in in-depth interviews. Of those, eight participants met the criteria of having continued in classroom positions for a substantial period of time and having published articles about their teaching while working in classroom positions. The present study drew from that dataset of eight interviewees who were classroom teachers at the time they wrote a published professional article, defined here as a peer-reviewed or non-peer-reviewed article about teaching. That number excludes teachers who might have published fiction or poetry or memoir, or articles on non-educational topics (of which there are many examples around NWP and who themselves represent a fine tradition of teacher-writers, albeit a different kind of teacher-writer than I am focusing on here). The eight teachers in the sample were asked directly about their experiences in writing for publication. Questions in that section of the interview included a request for an overall account of the publication experience, any aspects of it that were challenging, and what resources teachers drew upon. It should be noted that teachers were not directly asked how NWP in particular offered resources; however, the entire interview of which these conversations were a part centered on NWP influence, broadly defined, on teachers’ overall work and lives. I worked with the eight transcripts, engaging first in open coding and then developing a more structured coding dictionary through a series of
240 A. Whitney passes within-case and cross-case. For the purposes of this chapter, analyses focused on references to contexts, particularly the challenges that were salient for teachers as they worked toward publication and how they drew upon the NWP in facing those challenges. Ultimately, as these data show, the NWP context served as “counternormative setting” in which teachers could work around or through the challenges they faced. The challenges cited by teachers who had published ranged in scope from immediate, practical ones to broader rhetorical ones. These were tied to one another in interesting ways. The challenges are depicted in Figure 18.1. The categories of challenges cited by my informants fall into three rough sites of concerns ranging from practical to rhetorical to self-related, represented here by the overlapping circles of the Venn diagram. The circle on the lower left addressed the most practical concerns, things like finding time to write amid family and teaching duties. These concerns are by no means inconsequential, and in fact I suspect they must be dealt with before any of the rest can even come fully into focus, but they are primarily physical and practical challenges. On the lower right, in what I am loosely terming the rhetorical circle, are the concerns most tied to the audience for the work; these tend to be challenges tied to power, to one’s authority to SELF
Maybe “not really a writer”
Afraid of feedback
Habits/ self-discipline
Time constraints/ other commitments
No reader/ lack of support Troublesome relationships with collaborators How to speak to this audience
Physical challenges
Unfamiliar with article conventions
PRACTICAL
Figure 18.1 Challenges encountered in writing for publication.
RHETORICAL
Classroom teachers as authors 241 speak. In the upper, center circle are a set of self-challenges, challenges related to self or ego. The overlapping areas depicted in the Venn diagram reflect that each of these areas, while helpful distinctions for analysis, actually helps to construct and is constructed by all the others as well. Practical concerns included things like kids, carpools, finding time, computer trouble, etc. These challenges start with even issues as basic as the toll a writing day can take on the body; as one teacher remarked, writing can strain the eyes and cause the ankles to swell from hours in a chair, for example. It also includes the challenge of finding time to write; teachers are busy with student needs, preparing for classes, grading, and outside duties in addition to family and community involvement; one teacher recalled, “I really felt like I was stretched pretty thin.” The practical category also includes things like habits and self-discipline. For instance, one teacher commented that she must not really be a writer because “real writers find time to write and they take time to write,” something she had not succeeded in doing. To the extent that this last challenge can be overcome by establishing a routine and rearranging other duties, I categorize it as a practical challenge, but as Figure 18.1 indicates I also place that challenge within the self domain because as teachers come to define themselves as writers, they get better at finding ways to overcome such concerns. Part of “being a writer” is adopting “writerly habits” and routines. The next of the three circles, “rhetorical challenges,” is a rough cover term for challenges associated with the audience for the writing or how writing the article represented a move into a professional community broader than the classroom or the local level. They concern the positioning of the author relative to topic and audience along with the constraints and affordances of the academic article as a genre. Many of these comments had to do with the sense of “writing up” to an audience of more senior, more educated people with more authority. Related to that, for example, is a seemingly small issue, not knowing how to use APA format; this is not solely a practical consideration but instead also speaks to the notion that there are conventions that insiders in academic writing know and partici pate in shaping, and many teachers see themselves as outside that discourse community. One teacher spoke about her ignorance of the process of peer review, saying “I didn’t really know what the standards were. I didn’t really know what the procedures were for scholarly publishing.” Thus when she received reviews or a response from an editor she did not have the understanding of where such responses came from or how she might be expected to respond. Another had to do with relationships between coauthors, or would-be collaborators. One teacher, describing her relationship to some coauthors on her first publication, noted, “I didn’t really have a lot of—authority isn’t the right word—but that is not how they saw me. I was not a collaborator in it. I was a contributor, sort of.” This sense of being less than a full partner in authorship resulted in her withholding some input she saw as important to the article. So while on the one hand another part of this study shows that cross-level collaborators are really
242 A. Whitney important in that often they invite teachers to write for publication for the first time and therefore sponsor them into publication, at the same time these relationships are fraught and can make the writing difficult or even stop people from participating at times. At even the most accomplished end of the spectrum of published teacher-authors, a challenge still arises with regard to how to position oneself and one’s arguments in a piece of writing relative to the field. One teacher described an unexpected response from a peer reviewer at a journal: Part of what I’ve been working on . . . is how do you broach the conversation or create the dialogue without being offensive? . . . one of the reviewers said that my piece was too negative about traditional methods . . . my piece was about creating a critical lens with students, so I wrote the piece with a critical perspective. And the reader took that as negative. Writing for an audience of teachers, broadly conceived, a teacher thus negotiates questions of how to frame arguments in a way the audience can hear and how to position him- or herself in the writing as an insider speaking to colleagues rather than an outsider coming from outside the existing conversations. In the last of the three circles, a final set of challenges cited by teachers in the study concerns the teacher’s identity and sense of self. These include concerns tied to self-presentation, confidence, and sense of connectedness to a group of likeminded others. One teacher, for example, related that she didn’t have anyone to respond to her work. She noted that in fact she did have a friend who was willing, but she rarely took advantage of this, explaining that it challenged her usual persona as a confident, self-assured person: Being confident can have its drawbacks. You think [a piece of writing is] good, and maybe it’s not . . . sometimes I’m not always comfortable saying, “Well, will you read this?” . . . I don’t have that support. And I’m embarrassed to ask . . . I don’t ask for help very well. Thus while this teacher avoided sharing her work with a peer to avoid feeling vulnerable, she acknowledged that the result was increased isolation and the possibility that her finished work would not be as good as might otherwise be possible. Interview participants also noted, both explicitly and implicitly, the problem of feeling like a fraud or not feeling like a writer. One teacher, for example, described how she felt when drafting a piece was difficult, saying that in those moments “I think, well, maybe I’m not really a writer.” This issue and the others I have located in the “self” category spread out to affect all domains. For example, a sense of isolation from others who might respond to the work is in some sense a practical
Classroom teachers as authors 243 problem (simply lacking proximity to colleagues in the daily schedule of classroom teaching) and a rhetorical problem (as isolation from readers both local and global) as well. Similarly, fear of feedback or criticism can be read (as the teacher citing it read it) as related to a lack of self- confidence as a writer, but it can also be read as a rhetorical problem, a symptom of a perceived power imbalance between the author and his/her audience. The teachers in this study, then, cited all these challenges, practical, rhetorical, and self-related. Yet this group of teachers was successful in publishing an article, sometimes many more than one, even while they grappled with these challenges. All eight of the teachers cited NWP influences as resources that they drew upon when writing for publication. Figure 18.2 displays the forms of influence they cited: On the left, they drew on NWP for content. Teachers cited how they developed ideas for publications in the context of NWP activities or how they had published articles born out of concepts they had first encountered through NWP contacts. In the second column, they drew on people—either as motivators (frequently an NWP leader invited them to write for a specific article or book) or as cheerleaders or as responders. One participant, for example, described first being encouraged to try publication by NWP colleagues and then being coached in the development of a piece by an NWP staff member: I wrote a personal narrative at this [Writing Project] writing retreat. And I was encouraged by everyone when I shared it, that I should send it to the [NWP] Voice. So I did . . . and [the editor] wrote back and said that he really enjoyed it, but—and I was getting ready to read “But we can’t use your piece right now”– but instead he said “. . . and we see a larger point to address, and that is student-teacher relationships, and would you write an essay on that and embed your narrative in the essay. Several teachers cited the NWP as influencing them to begin pieces in the first place. I would not be inclined to write if it were not for all that help and love and support. So it is really because of them and all of their efforts that I kept going with it and carried things through to deadline and got Content Ideas from WP end up in a publication’s content
People Process Writing in a community/ having respondents Invitations or exhortations
Habit of daily writing
Figure 18.2 NWP resources used in writing for publication.
Self Thinking of self as a writer
244 A. Whitney involved in projects because if I were to have been my own initiator, I would not have done it. It is notable that the teacher cites the “love” of NWP colleagues here. All but one of the teachers used terms of relationship and affiliation such as “love,” “support,” “encouragement,” or “mentor” to describe the influence of NWP people on their publication activities. Next, the teachers reported drawing on NWP for support in the process of writing the piece itself—for assistance in thinking about forms, or for assistance in generating a discovery draft or journaling, etc., that they could draw from later in building a piece. One teacher described this influence in detail: I think going back to the Writing Project, and actually doing all the writing that we were expected to do and having the luxury of having the time to write, number one, and a community to write in, I just started writing like crazy. I mean I couldn’t really believe how much I was writing and how much benefit I was getting from it. This teacher explains that she acquired a set of habits through the NWP, including daily writing and the fluency to generate a large amount of text, that carried over into writing for publication later. Note also that in Figure 18.2, “response from peers” stretches across both the category of “people” and that of “process,” for the interactions described typically were instances of both. For example, peers are of course people, and so they play a role as readers, but NWP-sponsored opportunities for peer response also support the teacher-writer in developing a process that includes having others read one’s work. Finally, in the far right column, teachers drew on NWP when writing articles for publication in that they credit NWP with helping them to think of themselves as writers. This led to the establishment of “writerly” habits, of an openness to and interest in feedback, and a sense of having something to contribute that made writing for publication seem possible and even perhaps necessary. Note that this sense of oneself as a writer is not necessarily specific to the particular genre of the academic article: as one teacher put it, “I think what the Writing Project allowed me to do, though, was think of myself as a writer more than it encouraged me to be the professional writer.” Ultimately, the NWP network served as a setting in which the practical, rhetorical, and self-related norms of classroom life were suspended or even subverted. As such it was a counternormative space in which new practices and positions became possible. Teachers drew from resources available in the NWP context to overcome challenges present in the normative context of the classroom and school. Figure 18.3 reflects the complex and layered ways in which teachers translated NWP influence into supports that mitigated the specific challenges they had identified. The relationship between
Classroom teachers as authors 245 the challenges teachers grappled with and the way they called on NWP resources is not a one-to-one relationship. In fact each resource gets applied across a range of challenges. Figure 18.3 shows how multiple resources were brought to bear on each of the challenges teachers cited as significant. It is not the case that teachers worried about article conventions, NWP offered assistance with article conventions and then the teacher moved forward. Instead, the set of resources available within the NWP network were cultural ones—groups of collaborators, audiences to write for and respond to, and models (both in people and in texts) to emulate and adapt from. Thus it was not as a set of discrete and concrete tools but as a space for interaction that NWP supported teachers in writing for publication. Seeing oneself as a writer is actually at the center of how NWP influence supported these teachers in meeting the challenges of writing for publicaSELF
Maybe “not really a writer” 4,5
Afraid of feedback 2,5
Habits/ self-discipline 4,5
No reader/ lack of support 2,3 Troublesome relationships with collaborators 1,3,5
Time constraints/ other commitments 4,5
How to speak to this audience 1,2,3,5
Physical challenges 4
Unfamiliar with article conventions
PRACTICAL Content Ideas from WP end up in a publication’s content (1)
RHETORICAL People Process Writing in a community/ having respondents (2) Invitations or exhortations (3)
Habit of daily writing (4)
Figure 18.3 NWP resources applied to challenges.
Self Thinking of self as a writer (5)
246 A. Whitney tion. That teachers in the NWP context are called “writers” and treated as writers was the most salient feature of the NWP context cited by teachers and applied to the challenges with which they had grappled. It is helpful to envision teachers trying to write for publication as navigating an interlinked web of challenges and resources, with “seeing oneself as a writer” positioned as the central resource that itself makes access to many other resources available. For example, when you see yourself as a writer, you creatively work out ways to find time to write. You get yourself a new chair (addressing physical challenges). You don’t mind hearing feedback from a group, even when it is hard feedback calling for drastic revision, because you are a writer and receiving feedback is part of what writers do. And your sense of “writing up” to an audience or of discomfort taking your rightful place among a group of collaborators is mitigated by this new feeling, that you are a writer among writers and as such you have something to say that must find expression. Prior research on teachers’ experiences in the National Writing Project (Whitney, in press) speaks to how the NWP instantiates these resources in the culture and activities of its Summer Institutes, and a key finding of that research is the centrality of interactions in the writing group to transforming sense of self as a writer. Future study should follow the development of this sense of self as a writer from the NWP Summer Institute to continued network involvement to publication activities and beyond. Considering the features of the NWP network as a community and the potential of that community as a place where publication is supported highlights how the NWP can work as a counternormative setting in which teachers find some purchase in climbing over the obstacles that might keep them from writing for publication and therefore from claiming with full voice their place in our field—since even in light of Patti Stock’s (2001, 2005) work on genres of the scholarship of teaching and work like that of the Carnegie Academy for the Scholarship of Teaching and Learning (Carnegie Foundation for the Advancement of Teaching, 2004; Hatch, 2005), the article is still the coin of the realm when it comes to what we consider to be the literature of our field. It is important to note that this counternormative work was accomplished not by directly reforming school organization or the positioning of teachers within the larger culture of the field—surely a massive and daunting task, if a possible one at all—but by providing a concurrent, alternate discourse community through which individual teachers could experiment with new ways of positioning themselves vis-à-vis writing and audiences as well as observe and take up new practices. They then were able to retain connection to the resources of that discourse community even while working in the unaltered school setting. Note also that NWP supported these teachers not in the form of particular programs and logistical supports such as buying out teachers’ time or offering workshops in the journal article as a genre (although that work does happen and is important) but that their involvement in the network made available certain social resources and certain rhetorical spaces in
Classroom teachers as authors 247 which they could reposition themselves in the ways that were necessary to get the work of publication done. This point is most important for those interested in understanding teacher professional development because it is through that understanding, of NWP as an alternate space or a space in which teachers reposition themselves, that we begin to really unpack the potential of NWP for encouraging teacher leadership, and it is there that we derive the lessons of NWP that go beyond writing—that NWP represents a social setting of a particular type (with features perhaps best described in Lieberman & Wood, 2003) with particular affordances that we might want to reproduce (or at least encourage) elsewhere.
References Bissex, G. L., & Bullock, R. H. (Eds.). (1987). Seeing for ourselves: Case-study research by teachers of writing. Portsmouth, NH: Heinemann. Burton, J. (2005). The importance of teachers writing on TESOL. TESL-EJ: Teaching English as a Second or Foreign Language, 9(2). Carnegie Foundation for the Advancement of Teaching. (2004). The gallery of teaching and learning. Retrieved March 6, 2004, from http://gallery.carnegiefoundation.org. Casanave, C. P., & Vandrick, S. (Eds.). (2003). Writing for scholarly publication: Behind the scenes in language education. Mahwah, NJ: Lawrence Erlbaum. Cochran-Smith, M., & Lytle, S. (1993). Inside outside: Teacher research and knowledge. New York: Teachers College Press. Dahl, K. L. E. (1992). Teacher as writer: Entering the professional conversation. Urbana, IL: National Council of Teachers of English. Fecho, B. (2003). Yeki Bood/Yeki Na Bood: Writing and publishing as a teacher researcher. Research in the Teaching of English, 37(3), 281–294. Goswami, D. E., & Stillman, P. R. E. (1987). Reclaiming the classroom: Teacher research as an agency for change. New Jersey: Boynton/Cook. Hatch, T. (2005). Into the classroom: Developing the scholarship of teaching and learning. San Francisco, CA: Jossey-Bass. Henson, K. T. (1984). Writing for professional publication: Ways to increase your success. Phi Delta Kappan, 65(9), 635–637. Henson, K. T. (1986). Writing for publication: Playing to win. Phi Delta Kappan, 67(8), 602–604. Henson, K. T. (1988). Writing for education journals. Phi Delta Kappan, 69(10), 752–754. Henson, K. T. (1990). Writing for education journals. Phi Delta Kappan, 71(10), 800–803. Henson, K. T. (1993). Writing for successful publication: Advice from editors. Phi Delta Kappan, 74(10), 799–802. Henson, K. T. (1995). Writing for publication: Messages from editors. Phi Delta Kappan, 76(10), 801–803. Henson, K. T. (1997). Writing for publication. Phi Delta Kappan, 78(10), 781–784. Henson, K. T. (1999a). So you want to be published? Kappa Delta Pi Record, 35(2), 79–81. Henson, K. T. (1999b). Writing for professional journals. Phi Delta Kappan, 80(10), 780–783.
248 A. Whitney Henson, K. T. (2001). Writing for professional journals: Paradoxes and promises. Phi Delta Kappan, 82(10), 765–768. Henson, K. T. (2003). Writing for professional publication: Some myths and some truths. Phi Delta Kappan, 84(10), 788–791. Henson, K. T. (2005). Writing for publication: A controlled art. Phi Delta Kappan, 86(10), 772. Jost, K. (1990). Why high-school writing teachers should not write (rebuttal). English Journal, 79(3), 65–66. Lieberman, A., & Wood, D. R. (2003). Inside the National Writing Project: Connecting network learning and classroom teaching. New York: Teachers College Press. National Writing Project. (2008). About the National Writing Project. Retrieved February 29, 2008, from www.writingproject.org/About/nwpitour1.html. Perl, S., & Wilson, N. (1986). Through teachers’ eyes: Portraits of writing teachers at work. Portsmouth, NH: Heinemann. Root, R. L., & Steinberg, M. (1996). Those who do, can: Teachers writing, writers teaching. Urbana, IL: National Council of Teachers of English and the National Writing Project. Stock, P. L. (2001). Toward a theory of genre in teacher research: Contributions from a reflective practitioner. English Education, 33(2), 100–114. Stock, P. L. (2005). Practicing the scholarship of teaching: What we do with the knowledge we make. College English, 68(1), 107–121. Sunstein, B. (1994). Composing a culture: Inside a summer writing program with high school teachers. Portsmouth, NH: Heinemann. Whitney, A. (in press). Writing and the transformation of teachers in the National Writing Project. Research in the Teaching of English.
Part IV
Research on higher education practice
19 The international WAC/WID mapping project Objectives, methods, and early results Chris Thaiss University of California at Davis
Background and rationale My interest in an international WAC/WID “mapping project” grew out of my coordination of the National (US) Network of Writing-across-the- Curriculum (WAC) Programs, which began in 1980. The Network became the International Network of WAC Programs (INWAC) as recently as 2005, when our Canadian members asked that the name be changed to more accurately represent the membership. During its 28 years, the Network’s board of consultants have held annual meetings and, with other invited scholars, collaborated on four volumes of essays (1988, 1992, 2001, 2006) to portray the varied goals and best practices that characterized initiatives in WAC and writing in disciplines (WID), primarily in US higher education. In the earliest of these volumes, Strengthening Programs in Writing across the Curriculum (1988), Susan McLeod (who has edited or co-edited all four volumes) included results of a survey she and Susan Shirley had conducted on the prevalence of WAC initiatives in the United States in 1987. Surveying some 2,700 institutions and receiving some 1,100 responses, the McLeod–Shirley survey remained the only broad-based effort to measure the presence of WAC across the United States until the current project. That it was high time to undertake another broad-sweep survey of US WAC activity was clear. One major change was that, in the intervening 20 years, higher education—and our concept of “writing”—had been revolutionized by electronic technology. Electronic mail and multifunction course management systems have made it exponentially easier for students and teachers to write to each other about disciplinary issues and course content, as well as to submit assignments and scholarly papers and to dialogue about drafts (see, for example, Reiss, Selfe, & Young, 1998). Moreover, in those two decades, the scholarly resources to support WAC/WID initiatives have increased and become more easily available. For example, the WAC Clearinghouse website (wac.colostate.edu) housed at Colorado State University makes available (and easily updatable) information about WAC/WID initiatives across the globe. All four of the volumes noted
252 C. Thaiss above were published after 1987 (with the earliest two now available free at the Clearinghouse). The journals Across the Disciplines and The WAC Journal both began in the 1990s as print publications; both are now published electronically at the Clearinghouse. (Across the Disciplines was a result of a merger between the print publication The Journal of Language and Learning across the Disciplines and the online journal academic. writing.) In addition, the first National Conference on Writing across the Curriculum was not held until 1993; these have been held biennially since then, with the most recent three retitled International WAC Conferences, to reflect the cross-national influences that I will explore below. While the spread of readily available materials suggests that the presence of WAC/WID initiatives has intensified in the United States since 1987, popular lore has questioned this optimism. As WAC became the “old idea,” no longer attracting federal, state, and private funders who had stoked WAC program development in the 1970s and 1980s, some programs died. In their report of a ten-year follow-up survey (1997) of the 418 WAC programs identified in the 1987 survey, Miraglia and McLeod noted that two-thirds of the respondents said that their programs were still led by the same director; while this fact might evidence “strong, consistent leadership” (p. 53), they found threatening to WAC the perception by some leaders that programs would die after these initiators left or retired. Russell, in his second edition of the influential WAC history Writing in the Academic Disciplines: A Curricular History, concluded: “The U.S. educational system will have to find new ways of organizing teaching and assessing learning through writing in order to make WAC an expectation rather than an exception” (Russell, 2002, p. 332). Anecdotal evidence was contradictory. At our annual meetings of the National WAC Network at the conventions of the Conference on College Composition and Communication (CCCC), attendance remained high—but many of the self-identified “just starting” programs were described as attempts to restart a dead former initiative. Perhaps the most profound difference between the 1987 and 2006 contexts has been the internationalizing of the ideas of WAC and WID, another effect of the electronic revolution. Ironically, the US notions of “WAC” in the 1970s owed their origin to British researchers (e.g., James Britton and Nancy Martin) who first studied in depth the roles of language activities in learning across the disciplinary spectrum in the 1960s. But between 1966, when the legendary Dartmouth Conference brought together scholars from the United Kingdom and the United States to share practices in the teaching of writing, and roughly 2000, well into the information age, relatively few language-and-learning scholars from the United States shared concepts and techniques across borders of language and geography with their counterparts in Europe, Asia, Africa, South America, and even the rest of North America. Few scholars in US composition studies, focused as they were on emerging US policies and practices in education, had the international perspective of those who taught English
The international WAC/WID mapping project 253 writing, speaking, and reading to non-native speakers, most often outside the United States. The first International WAC Conference occurred in 2004 (it has still not been held outside the United States), and CCCC is still basically a US organization, though it is attracting an increasingly international mix of scholars and teachers. It would have been easy to limit the new survey of WAC/WID activity to the United States. The changes in US contexts have been sufficiently great over 20 years to stir interest in a new survey. But the notable presence of scholars from diverse countries at the 2004 and 2006 International WAC Conferences, as well as the substantial proportion (30 percent) of international attendees at the 2007 meeting of the International WAC Network in New York, encouraged a “mapping” effort relevant to the global scope of interest. Further, such organizations as the European Association of Teachers of Academic Writing (EATAW), the European Writing Centers Association (EWCA), and the European Association for Research on Learning and Instruction (EARLI) have welcomed participation in conferences by scholars and teachers from around the world—thus encouraging an international perspective on questions of writing in and across disciplines. This interchange has been enriched by the EATAW and EWCA listservs that regularly enable cross-national discussions of pedagogical and research issues. The February 2008 Writing Research Across Borders Conference in Santa Barbara, California, is the largest effort to date to enable writing researchers from around the world to come together to share their scholarship, but it builds on an impressive list of international conferences, such as the EATAW conference in 2007 in Bochum, Germany, with its thematic focus writing in the disciplines. Aiding this specific networking of writing researchers and teachers is the primary goal of the International WAC/ WID Mapping Project.
Aims According to the mapping project website (http://mappingproject.ucdavis. edu), the “first objectives” of the Project are to: 1 build a network of teachers and scholars from many countries who will contribute and keep in contact about their work in this field; 2 gather as much data as we can through surveys, interviews, and meetings. “Future objectives of the Mapping Project,” as we state on the website, “include publication and presentation of results by members of the collaborative in a range of venues internationally.”
254 C. Thaiss
Methods of the US/Canada survey The two surveys that thus far constitute the Mapping Project differ significantly. The US/Canada survey, launched in early 2006, is modeled on the McLeod–Shirley survey of 1987, with additional questions and significant rewording to reflect changes in context and the now much longer history of WAC/WID activity. I designed the survey with consultation from McLeod and from Terry Myers Zawacki, with whom I had collaborated on WAC research publications (1997, 2006). The survey depends for its language and emphases on the lengthy tradition of WAC/WID program literature in the United States, as summarized above. It asks a series of multiple-choice and open-response questions on the following themes: • • • • • • • • •
Program leadership/administration; Sources of funding; Goals of the initiative; Components of the program; Opportunities for faculty/staff training; Connections with other campus support services; Importance of electronic technology; Curriculum elements (e.g., courses, centers) devoted to writing; Incentives to program assessment.
Unlike the McLeod–Shirley survey of 20 years earlier, it also invites responses from colleagues at Canadian universities, not because the traditions of writing program design are the same in the two countries (they are not), but because of involvement in INWAC by some Canadian institutions. Graduate student researcher Tara Porter and I sent email invitations to writing program leaders at more than 2,600 institutions beginning in late 2006. Porter worked with the 1987 survey, the directories of INWAC and the WAC Clearinghouse, weblists of US and Canadian institutions, etc., to compose the original email list, which she continually updated with names suggested by respondents, by searching of websites, and through scholarly sources.
Methods of the preliminary survey of higher education worldwide The first call for international responses did not occur until late 2006— after design of the “preliminary” survey expressly for colleges/universities outside the United States and Canada. I asked Christiane Donahue, now of Dartmouth College and of the University of Lille, France, who participates in European organizations of writing teachers and in collaborative research on the teaching of writing in France (Donahue, 2002), for help in reaching European scholars who would comment on the design of an appropriate
The international WAC/WID mapping project 255 survey. Donahue generously contributed both by creating an initial mailing list of potential respondents and by holding small focus groups of scholars while in France in early 2006. Donahue asked these scholars how the US/ Canada survey might be suited to a European audience. Given that some early US respondents had reported trouble in applying the questions to their local situations, despite the commonness of the survey terms in American higher education, we expected that European scholars would find greater incompatibility. For example, the US/Canada survey assumes “WAC” and “WID” as commonly understood terms; it assumes “WAC program” as a recognizable entity, an institutionally organized effort to educate faculty across disciplines to take on some responsibility for helping students improve as writers, as well as to use writing as a tool of learning and critical thinking. Two decades earlier, 38 percent (418) of respondents to the McLeod–Shirley survey had reported having a “program” that fit this description, and the publications and listservs named earlier have helped ensure wide currency of these concepts. Even more deeply, the US/Canada survey—reflecting American “WAC”—assumes an institutional mission in higher education in the United States to continue the “general education” of undergraduate students in written literacy, as well as in political and civil awareness, training in non-native languages, science, mathematics, etc. (Thaiss, 1992). Not only do most US institutions devote a substantial amount of required coursework—for all undergraduate students—to all these areas of thought, but, specifically, virtually every American college or university follows the tradition of requiring coursework in English composition. “WAC” in US parlance usually implies an effort beyond this required investment in the teaching of writing—e.g., “writing intensive” courses within each discipline or faculty-training workshops for teachers in the humanities, sciences, and social sciences to help them aid their students’ writing development. Thus, a US college could respond to the US/Canada survey that “No,” it does not have a “WAC program,” but almost surely that institution will have had for a long time one or more courses in English composition required of all undergraduates. In designing a workable survey on “writing in disciplines” internationally, these assumptions of the one survey would have to be eliminated. As we predicted, the focus-group participants in Europe saw the US/Canada survey as fully inappropriate. As Donahue reported, virtually every term would have to be replaced, the questions redesigned. The current survey on the Mapping Project website is titled “preliminary,” to show that it is still a work in progress. It was designed by Donahue, Zawacki, and me in mid-2006 and was first sent out to a small mailing list of European scholars later that year. Rather than database entries being considered definitive and complete in themselves, they will help in further refinement of the instrument and as a springboard to further study, for example, in-depth interviews with and articles by survey respondents. Nevertheless, many respondents have
256 C. Thaiss rovided detailed information about writing and its teaching in their instip tutions—as well as generalizing about the teaching of writing in the particular national culture—thereby augmenting the cross-national scholarly record. The questions on the preliminary survey are as follows: 1 Where are students writing in your institution, either in a first language of instruction or in English? In what genres and circumstances? 2 Who cares in your institution about the improvement of student writing or student learning through writing? 3 Is improvement in student writing an objective of certain courses in a discipline or of the overall curriculum? How and why? 4 Have any teachers in/across disciplines met to talk about these issues or made an effort to plan curricula in relation to student writing? 5 What is the source of their interest and what models of student writing/learning development (e.g., articles, books, other documents), if any, help guide these discussions? These open-ended questions limit jargon and invite discursive response. Many responses thus far go on at length into nuances of institutional policy and practice, while frequently commenting on the terms of the questions themselves (e.g., “if by ‘caring’ you mean actual attention to student writing, then . . .”). In wording the questions, the three of us attempted to use terms common in discussions on the European listservs, as well as in exemplars of the European cross-national writing research, such as Björk, Bräuer, Rienecker, and Jörgensen’s (2003) Teaching Academic Writing in European Higher Education. We realized that any model we derived would privilege some discursive practices. The most significant shortcoming of the survey, until May 2008, was that it was in English, thereby privileging one international language. To illustrate the need for greater accessibility, a few respondents answered the questions in their first languages (e.g., French, Greek, Spanish), and cooperative colleagues translated. Beginning in May 2008, translations of the survey in French, German, Russian, and Spanish were steadily added to the website, with use of the Spanish translation particularly noteworthy.1 Graduate student researcher Erin Steinke of the University of California at Davis has assisted greatly in building the email list. The EATAW and EWCA lists (with thanks to Dilek Tokay, EWCA board member and past president, in particular) have helped build responses to the international survey. We have augmented the mailing list with names of the invited speakers to the Writing Research Across Borders Conference in Santa Barbara. Most important to building the list has been the generous “spreading the word” by those who have already responded. Scholars from Argentina, Australia, Austria, Canada, Germany, Greece, Hungary, Israel, Japan, South Africa, Turkey, the United Arab Emirates, and the United
The international WAC/WID mapping project 257 Kingdom, to name a few, have asked counterparts at other universities in their nations, regions, and language groups to send responses.
Responses to the US/Canada survey (as of October 2008) In this limited space, I report results from five of the 13 categories on the US/Canada survey. Further results and analysis will be reported at conferences in the United States and Europe in 2008–2009 and on the website. The responses (1,359) to the current survey are more than 50 percent (of approximately 2,600 institutions directly queried), somewhat higher than to the pre-electronic 1987 survey (43 percent). In comparison to the “Yes” answers in 1987 (418; 38 percent), the 47 percent figure in 2008 indicates substantial growth over 20 years. Moreover, of the 721 who report “No” WAC program in 2007–2008, 219 (16 percent of total respondents) report having “plans to begin a program.” Perhaps most noteworthy in these results is that almost 60 percent of PhD-granting institutions report having a WAC/WID program. The research-intensive nature of these institutions had been thought a hindrance to the development of teaching-focused initiatives such as WAC (Strenski, 1988), so this result bears more analysis in comparison with other findings of the survey. The markedly lower percentage of community colleges reporting programs (30 percent) also demands further analysis;
Table 19.1 Reported presence of “an initiative or program dedicated to student writing across disciplines, e.g., WAC or ‘writing in the disciplines’ (WID)” Survey year and total responses
Yes
No
1987 (1,113) 2006–2008 (1,359)
418 638
695 721
Notes Canadian institutions reporting: 71. Yes 26 (36%). No 45.
Table 19.2 WAC program presence by type of institution Type
PhD granting
Number (% of 148 all responding in that category)
MA/MS granting
BA/BS granting
Community college
139
209
102
258 C. Thaiss Table 19.3 Number of years program has existed Just starting
1–5 years
6–10 years
11–15 years
More than 15
77
154
107
89
166
Table 19.4 Length of time director has served (in years) No director Less 1–5 6–10 than years years 1 year
11–15 16–20 Over 20 Other (hiring a years years years director, rotating directorship, etc.)
2
31
65
224
88
27
14
146
while this percentage is far lower than those for other types of institutions, it nearly matches the percentage found in Stout and Magnotto’s 1987 survey of 1,270 community colleges, to which 401 schools responded. While the number of programs older than 15 years (166) is the largest single category, note that this figure is far short of the 418 reported to have existed 20 years ago in the McLeod–Shirley survey. The figure gives credence to the perception that many initial WAC programs faded after the 1980s, due to such factors as loss of a leader or loss of external or internal funding. Directly addressing this perception is the figure, 149, which represents those who answered “Yes” in the current survey to the category “Once had a WAC program.” However, the significant number of programs in each of the other categories testifies to the vibrancy of the WAC concept. Moreover, the high figures in each category also back up the perception that many programs have restarted after a hiatus. What stands out in this chart, particularly when compared to the previous, is the figure 224—directors having served 1–5 years, easily the most prevalent choice. By contrast, more than 60 percent of WAC programs are older than five years, with “more than 20” the largest single category (see above). Clearly, one measure of the success of WAC programs has been their ability to transfer leadership. Note that only 14 directors have served more than 20 years. How WAC programs have discovered ways to transfer leadership will be another subject for further research. Since WAC as an institutional structure works horizontally, across the traditional vertical hierarchy and stability of academic departments, WAC directors would most likely not have the prescribed terms of department chairs, and so leadership would transfer less predictably and institutions will show variety in how this transfer occurs. In comments by respondents to this question, there was occasional uncertainty about how to answer: for example, if a program had more than one current leader, or if there had been a succession of leaders in a relatively short span. Nevertheless, the figure 256—the number marked as “associate/full professor”—jumps out from the list. The figure represents roughly 45 percent
The international WAC/WID mapping project 259 Table 19.5 Academic rank of program leader Administrative faculty
Assistant Professor
Associate/ Full Professor
Full-time, non-tenureline
Part-time, non-tenureline
Other
50
79
256
54
16
83
of institutions responding to the question; that tenured faculty in this proportion lead WAC programs seems to validate the importance of WAC programs to the institutions that have them—and coincides with the increased recognition of composition studies as an academic discipline. We have not yet correlated the rank of the leader with the age of the program; however, given that the average WAC director has been in the job fewer than five years, it is likely that the position of WAC director is seen at almost half the institutions responding as suitable for a tenured faculty member.
Responses to the international preliminary survey As of October 2008, respondents from 275 institutions in 51 countries have contributed to the survey. The nation with the most responding institutions thus far is the United Kingdom, with 43. Responses from Argentina (18), Germany (15), Australia (13), Colombia (12), the Netherlands (10), Switzerland (10), Spain (9), Israel (8), Turkey (7), France (6), Mexico (6), Norway (6), South Africa (6), Venezuela (6), Brazil (5), Austria (5), are at or above the average per country. Also represented thus far are the following: Belgium; Bulgaria; Chile; Costa Rica; Czech Republic; Denmark; Egypt; Finland; Ghana; Greece; Guatemala; Guyana; Hong Kong (People’s Republic of China); Hungary; Ireland; Italy; Japan; Lebanon; Lithuania; Malta; Marshall Islands; Nepal; New Zealand; Portugal; Puerto Rico; Qatar; Russia; Singapore; South Korea; Sri Lanka; Sweden; Taiwan; United Arab Emirates; Uruguay. While almost all data have come from the survey itself, data on 25 institutions have come from other sources: for example Ganobcsik-Williams (2004); Zawacki’s interviews with program leaders in Sweden, Italy, and Hong Kong; my own conversations with and notes on presenters at the 2007 EATAW Conference. Because of space limitations here, I will give only the briefest characterization of the data received thus far in the responses to the five questions. I refrain from naming specific institutions and respondents, because, as explained on the website, At this early stage of the research, all survey responses are confidential, for data-gathering purposes only. Once the project moves to setting up a network of respondents, each respondent will be asked for permission to be included in the network.
260 C. Thaiss
Summary of responses to questions 1 Where are students writing in your institution, either in a first language of instruction or in English? In what genres and circumstances? The great majority of respondents report an extensive amount of writing occurring across the disciplines with which they are familiar and often across all disciplines in an institution—in either undergraduate or graduate courses or both. Depending on the institution, writing occurs in English and/or in other first languages of instruction. Genres typically include those most pertinent to the methods of the particular discipline. Thus, a variety of technical and experimental reports, reviews of literature, and essays are listed. Exam writing is mentioned ubiquitously, and thesis writing is frequent for both undergraduates and graduate students. At the graduate level, preparation to publish in learned journals—most often in English—is frequently mentioned. On the other hand, a few respondents, about 5 percent of the total, assert the lack of writing in disciplinary courses in their institutions and ascribe traditional cultural reasons for this pattern. These few respondents see a disjunction between their own institutional mission to improve students’ language skills and the lack of opportunity for students to use them in classes. 2 Who cares in your institution about the improvement of student writing or student learning through writing? All but a few respondents have interpreted this question as intended: to identify staff, faculty, and administrators who are taking an active role in helping students improve as writers or use writing to improve learning. Answers have varied greatly, but tend to cluster. Those who “care” about student writing development can be grouped as follows: •
•
Language teachers, either of English or another language of instruction, who, because of their training and awareness of improvement models, work in their own disciplines or as part of academic writing and learning centers; however, their awareness of writing’s importance and interest in writing growth are not shared by most others in the institution; an occasional subtheme in comments from this group is that colleagues do not see “learning to write” as occurring in the university—it should have happened before students reach this level—and therefore (to cite a typical comment) “teaching to write is not a university responsibility” (the attitude that Mike Rose called the “myth of transience”); respondents who see themselves in this situation make up the largest group of respondents to this question, roughly 60 percent. By contrast, language professionals, who, because of institutional or national goals and administrative support, have their interests shared
The international WAC/WID mapping project 261
•
•
by members of staffs and faculties across disciplines; enthusiastic responses from the United Kingdom, Australia, and the Netherlands tend to stand out in this group, though there are examples from other countries (roughly 25 percent of respondents). Again, language professionals, who because of inspiration from leaders at other institutions in the region or through networking, have begun to work toward awareness by others of the goal of writing improvement at their institutions (roughly 10 percent). Teachers who are not language professionals, but who represent a discipline in the humanities, social sciences, or sciences and who are committed to writing improvement (roughly 5 percent).
3 Is improvement in student writing an objective of certain courses in a discipline or of the overall curriculum? How and why? Those respondents who report the most involvement by administrators and disciplines see writing improvement as an overall objective and give a wide range of specific examples, often in multiple forms in the same institution: tutoring, workshops for students and teachers, modules in subject areas; even, in a few cases, required sequences of courses. By far the most frequent responses, however, correlate with the first group of respondents to question 2: in these institutions, the language professionals (a) teach writing as part of courses in the English language or in another first language of instruction; or (b) offer workshops in academic writing as part of an academic skills center; or less often (c) offer modules attached to a particular discipline. In the majority of responses, these professionals tend to feel ancillary to the main interests of the institution. Sometimes this feeling is expressed positively in terms of the autonomy of their operation, but more often is expressed less happily in terms of large workloads, meager resources, and lack of respect. 4 Have any teachers in/across disciplines met to talk about these issues or made an effort to plan curricula in relation to student writing? In US parlance, this is the most “WAC” question in the group, as the term “writing across the curriculum” implies a cross-disciplinary initiative with cross-disciplinary involvement. In the responses to this survey, by far the most frequent responses range from “no” to “beginning” or “planning to.” Relatively few responses speak of coordinated cross-disciplinary involvement, and none could be considered well-established, certainly in comparison with most of those programs responding to the US/Canada survey. Nevertheless, from institutions with substantial involvement of faculty and staff across disciplines, there is considerable enthusiasm and optimism. Again, respondents from places where there is either a significant national commitment to improved academic writing, as in Australia and the Netherlands, or where there have grown up cross-institutional
262 C. Thaiss etworks of those who “care” about student writing, as in the United n Kingdom and to some extent in Argentina and Turkey, are likely to report the most involvement by teachers across disciplines. 5 What is the source of their interest and what models of student writing/learning development (e.g., articles, books, other documents), if any, help guide these discussions? This question has drawn the widest variety of responses, which can be roughly sorted into categories: • • • •
• • •
The availability of scholarship and program models on the Internet, via websites (the most prevalent response). Inspiration from theory, most often “process” or “genre” scholarship, to explain teaching methodology. Institutional pressures, such as widespread local complaints about poor writing, as impetus (and sustenance) for such language center offerings as tutoring and workshops. External pressures, such as the Bologna Process in Europe, the pressure of licensing exams, the need for graduate students to publish in English in learned journals, or national educational goals for improved communication skills (e.g., in Australia). As noted above, inspiration from regional leaders and opportunities for networking in countries or regions. In better-established programs, in-house guidebooks and staff- development materials. The work of named scholars, usually without identification of specific publications: e.g., Bazerman, Bräuer, Carlino, Johns, Lea, Street, Swales (and Feak).
Next steps The collection of responses to the US/Canada survey ended on July 1, 2008. A first set of unanalyzed statistical results was published on the website (http://mappingproject.ucdavis.edu) in August. Tara Porter and I have begun analysis of responses in all categories, and these will be published at conferences and in articles beginning in late 2008. Collection of responses to the international “preliminary” survey will continue for the foreseeable future, aided by a growing number of scholars who spread the request. Meanwhile, this portion of the Mapping Project is about to move into its next phase: an international team of editors will identify roughly 40 institutions from 20 or more countries for in-depth interviews about aspects of program structure and growth. We plan to collect and analyze these more substantial pieces within an online and/or print anthology. It remains a basic goal of the project to translate the survey response list into an active international network of fellow teachers, administrators, and
The international WAC/WID mapping project 263 scholars sharing an interest in WAC/WID, broadly defined. The planned anthology is one way of doing this. Also toward this end, the survey itself was modified in March 2008 by our adding a question regarding confidentiality: “Would you be willing to have your name and institutional affiliation be made public in reports of this research?” About 90 percent of respondents (approximately 50) since the change have said “yes” to this request, and so future reports of the project will be able to share some names of respondents and institutions.
Note 1 I wish to thank the following scholars for the translations of the surveys and of responses in those languages: Céline Beaudet, University of Sherbrooke (Canada), French; Annette Verhein, Hochschule für Technik (Switzerland), German; Nina Shevchuk-Murray, University of Nebraska (US), Russian; Paula Carlino, University of Buenos Aires (Argentina), Spanish.
References Björk, Lennart, Bräuer, Gerd, Rienecker, Lotte, & Jörgensen, Peter Stray (Eds.). (2003). Teaching academic writing in European higher education (studies in writing). Dordrecht: Kluwer Academic. Britton, J., Burgess, T., Martin, N., McLeod, & Rosen, H. (1975). The development of writing abilities, 11–18. London: Macmillan Educational. Donahue, Christiane (2002). The lycée-to-university progression in French students’ development as writers. In David Russell & David Foster (Eds.), Writing and learning in cross-national perspective: Transitions from secondary to higher education (pp. 134–191). Urbana, IL: National Council of Teachers of English. Ganobczik-Williams, Lisa (2004). A report on the teaching of academic writing in UK higher education. London: The Royal Literary Fund. Martin, Nancy (1976). Writing and learning across the curriculum, 11–16. London: Ward Lock. McLeod, Susan (Ed.). (1988). Strengthening programs in writing across the curriculum. San Francisco, CA: Jossey Bass; available at: http://wac.colostate.edu. McLeod, Susan, Miraglia, Eric, Soven, Margot, & Thaiss, Christopher (Eds.). (2001). WAC for the new millennium: Strategies for continuing programs in writing across the curriculum. Urbana, IL: National Council of Teachers of English. McLeod, Susan, & Soven, Margot (Eds.). (1992). Writing across the curriculum: A guide to developing programs. Thousand Oaks, CA: Sage; available at: http:// wac.colostate.edu. McLeod, Susan, & Soven, Margot (Eds.). (2006). Composing a community: A history of writing across the curriculum. West Lafayette, IN: Parlor Press. Miraglia, Eric, & McLeod, Susan (1997). Whither WAC? Interpreting the Stories/ Histories of Enduring WAC Programs. WPA: Council of Writing Program Administrators, 20(3), 46–59. Reiss, Donna, Selfe, Dickie, & Young, Art (Eds.). (1998). Electronic communication across the curriculum. Urbana, IL: National Council of Teachers of English. Rose, Mike (1985). The language of exclusion: Writing instruction at the university. College English, 47(4) (April), 341–359.
264 C. Thaiss Russell, David (2002). Writing in the academic disciplines: A curricular history. 2nd ed. Carbondale, IL: Southern Illinois University Press. Stout, Barbara, & Magnotto, Joyce Neff (1988). Writing across the curriculum at community colleges. In Susan McLeod (Ed.), Strengthening programs in writing across the curriculum (pp. 21–30). San Francisco, CA: Jossey Bass. Strenski, Ellen (1988). Writing across the curriculum in the research university. In Susan McLeod (Ed.), Strengthening programs in writing across the curriculum (pp. 31–42). San Francisco, CA: Jossey Bass. Thaiss, Chris (1992). WAC and general education courses. In Susan McLeod & Margot Soven (Eds.), Writing across the curriculum: A guide to developing programs (pp. 87–109). Thousand Oaks, CA: Sage. Thaiss, Chris, & Zawacki, Terry Myers (1997). How portfolios for proficiency help to shape a WAC program. In Kathleen Yancey & Brian Huot (Eds.), Assessing writing across the curriculum: Diverse approaches and practices. Greenwich, CT: Ablex. Thaiss, Chris, & Zawacki, Terry Myers (2006). Engaged writers and dynamic disciplines: Research on the academic writing life. Portsmouth, NH: Heinemann Boynton/Cook.
20 Rhetorical features of student science writing in introductory university oceanography Gregory J. Kelly*, Charles Bazerman†, Audra Skukauskaite‡, and William Prothero† * Pennsylvania State University † University of California, Santa Barbara ‡ University of Texas, Brownsville
Studies of science education have provided evidence for the importance of writing in students coming to understand and use scientific concepts (Keys, 1999; Rivard & Straw, 2000; Wallace, Hand, & Prain, 2004), as well as learning to participate in science as a learning community (Chinn & Hilgers, 1999). These findings are consistent with work over the last three decades in writing across the curriculum, focused both on writing to learn (Britton, Burgess, Martin, McLeod, & Rosen, 1975; Emig, 1977; Fulwiler & Young, 1986), and writing communities in the classroom (Walvoord & McCarthy, 1990; Herrington, 1985). Simultaneously, studies in the rhetoric of science have made visible that writing and argument play important roles in scientists’ and technologists’ thinking and forming knowledge communities. The forms of expression, invention, and knowledge vary with professions and disciplines. The epistemic activity of researchers is saturated with rhetorical concerns of who is to be convinced of what, how others respond to novel work, what the organization of their communicative activity is, and what the goals of community cooperation are (Bazerman, 1988; Blakeslee, 2001; Knorr-Cetina, 1999; Latour, 1987; Swales, 1998; van Nostrand, 1997). The representation and role of evidence in relation to generalizations and claims has been crucial in the development of scientific argument (Bazerman, 1988; Chandler, Davidson, & Harootunian, 1991; Fleck, 1979; Lynch & Woolgar, 1990). This chapter brings together research in science education with research in science studies and in scientific writing to consider evidence formation in student writing.
Science, rhetoric, and education Rhetorical studies of science view knowledge as actively constructed by scientists working individually or collectively on problems and being held accountable to public standards through the reasoning displayed in texts open to criticism and evaluation. This perspective highlights scientists’
266 G. J. Kelly et al. need to refine reasoning, limit theoretical claims, marshal evidence, and understand strengths and limits of their evidence and arguments so as to make credible and creditable knowledge claims within their knowledgeable and critical communities of peers (Latour, 1987; Myers, 1990). To learn the argumentative practices of their fields, students must gain a feel for the communicative forms, forums, and dynamics of their fields. They must learn the kinds of claims people make and how they advance them; what literatures people rely on and how these literatures are invoked; what kind of evidence is needed to warrant arguments and how that evidence can be appropriately developed, analyzed, and interpreted given community standards; what kinds of concepts are appropriately evoked; and what kind of stance authors can appropriately take as contributors to their fields. As students develop their discipline-specific communicative skills, they enter into community practice of empirical investigation and application of communally developed knowledge. In this communal engagement with the material world (Goodwin, 1995) the role of evidence is centrally important. Currently, scientific fields generally endorse and enforce high levels of accountability between detailed findings and general idea claims through review and argumentation processes (Bazerman, 1988; Myers, 1990). Rhetorical analyses of writing in scientific professions have examined the historically emergent forms of argument deployed in professional practice— the genres and the activity systems they are part of (Bazerman, 1988, 1994, 1997; Prior, 1998; Freedman & Medway, 1994; Swales, 1990; van Nostrand, 1997). Related analyses have looked at the rhetorical specifics and strategies of individual cases of argument (Bazerman, 1993; Pera & Shea, 1991). This chapter brings this research to bear on how student writers make local linguistic, argumentative choices within a genre’s organization and the expectations invoked within the activity system—here encapsulated in a school assignment that foregrounds the use of evidence in relation to claims. While argumentation in spoken discourse has been examined extensively in science education (e.g., Erduran & Jimenez-Aleixandre, 2008; Sampson & Clark, 2008), writing provides a potentially useful strategy to engage students in the social and cognitive practices of evidence formation. Writing tasks can be constructed to socialize students to disciplinary knowledge, norms, and practices, providing realistic learning tasks, as in the case examined here. Written argument can also be used to assess students’ engagement with scientific knowledge, norms, and practices.
Educational setting The study was conducted in an introductory oceanography course at the University of California, Santa Barbara. The instructor (fourth author) of the course has been consulting for several years with the first two authors (Bazerman, Kelly) as part of his reflective development of this course and the related software. Because of the reflective, sophisticated design of the
Rhetorical features of student science writing 267 course and the particular character of oceanography, the course has several unusual features that bear on the study. Oceanography is an inherently multidisciplinary science, drawing from physics, geology, chemistry, and several life sciences. It is a subject that university students usually have little prior experience of in secondary school. Most of the 96 students enrolled in this lower-division general education course during the term studied are not geological science majors. The stated objectives in the students’ laboratory manual included giving students the experience of “thinking like a scientist” in addition to “learning basic facts about the earth.” Students were informed that they will learn to “develop some ability to think critically about science and scientific claims,” “gain skills in using the computer,” and to use real Earth data to make their “own scientific judgments and conclusions” (Prothero, 2001, p. 2). Writing was a key instructional component of this course. For the purposes of our analysis in this chapter, we examined the first paper that required students to engage in scientific practices including understanding relevant background knowledge (i.e., the theory of plate tectonics), asking researchable scientific questions, selecting data and making observations relevant to the question posed, interpreting data to support a theory or model, presenting an argument, and evaluating the work of others. The writing was supported by the course lectures, the laboratory sessions, and an interactive CD-ROM developed to provide access to real Earth data sets organized in geographical and conceptual ways (see Kelly, Regev, & Prothero, 2008; Prothero & Kelly, 2008). The students attended three one-hour lectures, offered by the course professor, and one two-hour laboratory session of approximately 20 students each week, led by graduate-student teaching assistants. The laboratory room had 25 Macintosh power PC computers with CD-ROM drives, and an AppleShare file server, all dedicated to the course. Students in this course used an interactive CD-ROM, “Our Dynamic Planet,” which included a variety of instructional resources and activities, including access to Earth data sets as a basis for solving problems associated with plate tectonics (Prothero, 1995). The data sets are displayed on maps of various magnifications, by which students can plot earthquake locations and cross- sections, seafloor elevation cross-sections, cenozoic volcano locations (on land). They can also determine island ages and measure heat flow as well as access movies and still graphics concerning particular locations. Data of this sort allowed students to pose questions, consider relevant evidence, evaluate hypotheses, and illustrate the theory of plate tectonics. Plate boundary types could be identified through earthquake, volcano, elevation, and heat-flow analyses; and plate motion velocities could be identified through consideration of island ages and hot spots. The technical scientific paper examined here asked students to formulate arguments characterizing plate boundaries and motion in terms of the theory of plate tectonics, based upon relevant data. The assignment
268 G. J. Kelly et al. included suggestions about the relationship of this assignment to previous work, uses of evidence for plate tectonics, and arguments based on geophysical evidence. The scientific genre and more detailed aspects of the argument to be produced by students was specified through instructional episodes dedicated to scientific writing, through a detailed set of instructions and examples provided in a laboratory manual, and through peer and instructor feedback on student writing (see Kelly & Takao, 2002; Kelly, Chen, & Prothero, 2000).
Methods and results To test the hypothesis of a pilot study of two papers (Kelly & Bazerman, 2003), we increased the number of student papers by using a random sample of 18 student papers from the total number of 96 students enrolled on the course during the 2001 version of the course. For each of the student papers we entered each sentence verbatim into an ExcelTM spreadsheet from the seven pre-specified sections of the papers: abstract, introduction, methods, observations, interpretations, discussion, and conclusion. We then proceeded with the three initial analyses of the rhetorical, epistemic, and semantic cohesive dimensions. Rhetorical dimension: Our first analysis considered the rhetorical tasks and subtasks of the technical paper as defined by the seven pre-specified sections of the papers. Our pilot study suggested that while there were rhetorical moves that could be identified within paper sections (following Swales, 1990, the theoretically salient differences occurred as students moved across the predefined sections from overview and introduction (abstract, introduction); to stating facts of the matter (methods and observations); to the more theoretically oriented arguments (interpretations, discussion, and conclusion). Two methodological issues surfaced. First, the “discussion” section was newly added to previous sections of the writing assignment and requested that students put their “findings into a broader context” (Prothero, 2001). This was the only section of the paper that was not described with “typical examples” and a “checklist” in the students’ laboratory manual. The discussion sections thus introduced ambiguity into the analytical procedures. Second, two of the 18 students combined the interpretation and discussion sections with no clear demarcation. In these cases, we averaged numerical counts for the combined sections. Epistemic dimension: Our second analysis consisted of identifying the level of generality of claim of each sentence. In those cases where compound sentences made claims at multiple levels we choose to code the sentence at the highest epistemic level. There were six epistemic levels, which represent a continuum from specific data-pointing claims (level I) to more general theoretical claims (level VI). An additional category “PC” refers to personal comment or other metadiscoursal remarks made by the author (e.g., “Considering the plate tectonics that have taken place in the last 40 million years, it would be interesting to see what this region looks like in
Rhetorical features of student science writing 269 another 40 million years.”). The categories of epistemic generality are subject-matter specific, in this case directly derived from geological description and theory. Further, the epistemic categories were developed in relation to this specific assignment and not across a range of geologic arguments and thus must be understood as assignment-specific as well as subject-matter specific. The six categories are as follows: representations of data; identification of topographical features; relational aspects of geological structures; data illustrations of the authors’ geological theories or models; authors’ proposed geological theory or model; general description of geological processes and references to definitions, experts, and textbooks. (Further details of the epistemic analytic levels are available in Kelly & Takao, 2002.) Our analysis procedures consisted of developing initial common agreement among three coders of the respective definitions. The 18 papers were coded sentence by sentence by two coders (Kelly, Skukauskaite), differences discussed and brought to a third coder (Bazerman). All cases of disagreement were resolved through consistent principles that emerged in discussion. Table 20.1 shows the variation of the epistemic level of claim across the sections of the 18 papers as well as identifying the averages for the four highest- and lowest-rated papers. The highest level of epistemic claims (most theoretical) occurred in the conclusion, followed by the interpretations and discussion sections, the abstract and introduction, followed by the observations and methods (least theoretical, most factual). Semantic cohesive dimension: A third analysis concerns ways the students’ arguments cohere lexically, with particular attention to semantically related specifics and generalizations. Based on the work of Halliday and Hasan (1976) and Hoey (1983, 1991), our specific focus was on reiteration of the same word or word root (e.g., volcano, volcanic). In addition we considered collocation, the association of lexical items that regularly co-occur (e.g., plate and tectonic). We treat synonymous terms (e.g., earthquake, tremor, seismic) as distinct lexical items. We also did not consider indexical pronominal references (e.g., “these” referring to specific earthquakes) as lexically cohesive with the original term. Within this corpus no lexical items had dual meanings, with all terms used univocally and explicitly. We have not, however, studied this interesting lexical univocality and how it might be related to the technicality of the writing task or the state of student knowledge. Nonetheless, the convergence of semantic meanings and lexical forms in this case simplifies the analysis. After identifying the specific lexical cohesions in each sentence, we then sought to examine the overall cohesive picture of each student paper through four analyses. First, we plotted the key terms (those identified as having at least one repetition) by first mention (y-axis) against the sentence sequence (x-axis), noting the placement of these terms in the sentences comprising the paper (see Figure 20.1). For example, for the student paper mapped in Figure 20.1, the term “plate” appeared in the following
Abstract Introduction Methods Observations Interpretations Discussion Conclusions
Paper sections
3.88 3.55 1.20 2.59 4.07 4.03 4.23
1.22 1.76 0.56 1.02 1.07 1.43 0.97
3.91 4.16 1.17 2.52 3.91 4.83 4.00
0.32 1.57 0.19 0.48 0.39 1.08 0.38
stdev
average
average
stdev
Highest-rated papers (n = 4)
Total sample (n = 18)
Average epistemic level of claim
3.86 4.42 1.08 2.99 3.94 4.39 4.63
average
0.59 0.89 0.17 0.53 0.27 0.63 0.48
stdev
Lowest-rated papers (n = 4)
Table 20.1. Aggregate distribution of epistemic level of claims across paper sections. Shown are the results for the total number of student papers (n = 18) and then the subset of four highest- and lowest-rated papers by subject-matter expert
Figure 20.1 Sentence-by-sentence analysis of lexical cohesion for highest-ranked paper.
272 G. J. Kelly et al. s entences: 1, 2, 4, 5, 6, 8, 9, 29, 30, 31, 38, 39, 40, 41, 44, 45, 46, 47, 48, 54, 57, 58, and 59. By considering these key terms, their introduction, their place in the overall organization of the paper (in sections, place within and across sections) and their frequency, we were able to get an overview of the text organization. For example, our subsequent cohesion analyses (cohesive terms per sentence, examining key-term use, and section boundary sentences) followed from the cohesion plots like that of Figure 20.1. Our second analysis considered the average number of cohesive terms per sentence, aggregated by paper section. We were interested in examining how the differing rhetorical needs of the paper sections led to variations in lexical cohesion. As shown in Table 20.2, we considered the entire sample as well as the four highest- and lowest-rated papers. Third, we were interested in examining how the cohesion exhibited by first sentence of a paper section compared to the overall cohesion within the specific section. For example, consider sentence 29 in Figure 20.1, which in its original form read: “The topography of the area has resulted from the shifting of the plates which has caused the abundant earthquakes and volcanoes, as well as the trenches and the islands.” Through the analysis presented in Figure 20.1, this first sentence of the interpretations section can be seen to be bringing together a theoretical term (plates) and geological features (trenches and islands) with data-referencing terms (area and topography) and specific data items (earthquakes and volcanoes.) To examine this in detail we considered the number of cohesive terms per sentence for the first sentences of each paper section and compared these values with the overall number of cohesive terms per sentence for the relevant section. A summary of these calculations in presented in Table 20.3. Our fourth consideration of lexical cohesion concerned use of some of the terms across the 18 student papers. From the initial lexical cohesion maps (Figure 20.1) we considered those terms used by the most number of students. Of the terms appearing most often by the student writers, 12 can be classified into four categories: (a) theory terms (plate, subduct-, converg-) refer to words that cannot be read off the data representations and require some understanding of the mechanisms of plate tectonic theory; (b) data terms (volcan-, ocean[ic], [earth]quake) refer to words that can be viewed as icons in the data representations of the CD-ROM database; (c) terms referencing data representations (data, profile(s), figures(s)) locate data and draw readers’ attention to data; (d) directional terms are south, north, west, east, and combinations such as northwest, etc. For these four types of terms we were concerned with the number of students making use of the terms and in which sections they appear. Our focus was on appearance, and not the total number of times a term occurred. Figures 20.2a–d represent the different ways these four types of terms were used by the student writers across the paper sections.
Abstract Introduction Methods Observations Interpretations Discussion Conclusion
Paper sections
8.75 7.49 4.59 6.43 7.20 5.70 8.75
1.55 2.16 2.19 1.33 1.67 1.55 2.59
8.26 7.75 3.88 6.43 6.41 5.29 7.58
2.01 1.96 1.49 0.81 0.89 1.08 0.79
stdev
average
average
stdev
Highest-rated papers (n = 4)
Total sample (n = 18)
Average number of cohesive terms per sentence
Table 20.2 Average number of cohesive terms per sentence in each paper across sections (n = 18)
8.91 6.55 4.91 6.55 7.18 5.25 9.08
average
2.09 2.20 2.60 1.49 1.80 2.37 3.85
stdev
Lowest-rated papers (n = 4)
274 G. J. Kelly et al. Table 20.3 Comparison of cohesive terms used in first sentence in paper sections with total in respective sections Paper sections
Abstract Introduction Methods Observations Interpretations Discussion Conclusion
Comparison of first sentence of paper sections with total section average number of cohesions First sentence higher number of cohesive terms
First sentence lower number of cohesive terms
8 9 7 11 13 11 11
10 9 11 7 5 5 7
Discussion of hypotheses in light of results The previous study of two highly rated papers suggested five hypotheses which we tested against this larger, more varied sample. Hypothesis 1: The arguments showed a hierarchical arrangement within the logic of the genre structure, i.e., the students introduced and maintained use of key conceptual terms (e.g., plate, tecton-, topograph-, boundary(ies)). These terms were combined with specific geographical terms (those locating the areas in question: e.g., California coast, Aleutian Islands) and a set of lower-level terms (often conceptual such as island, trench, depth, mountain). Hypothesis 2: The epistemic status of the claims made varied according to the rhetorical needs of the differing sections, defined by the genre structure. The introduction, interpretations, and conclusions showed the greatest levels of generality. Our first and second hypotheses are closely related. They both examine the relationship between epistemic level and organizational components of the paper arguments. As in the original study we found distinctive differences in epistemic level among the predefined sections of the paper (abstract, introduction, methods, observations, interpretations, discussion, conclusion), and these differences followed the organizational logic of the different sections. These differences followed a high–low–high pattern (see Table 20.1) as in the original study, with abstract and introduction presenting material of a higher epistemic order (3.88 and 3.55 on a six-point scale, respectively) than the methods and observations (1.20 and 2.59, respectively). Then the interpretations, discussion, and conclusion include claims of higher epistemic level and thus theoretical import (4.07. 4.03, and 4.23, respectively). Further within the middle section, observations exhibited higher epistemic levels than the methods.
Theory terms
20
Appearance
15
10
5
0
1
2
3
Plate
4 Paper sections
5
Subduct-
6
7
Converg-
Figure 20.2a Appearances* of theory terms across paper sections.** Notes *“Appearances” refer to number of students using term in respective paper sections. **Paper sections are as follows: 1 = abstract, 2 = introduction, 3 = method, 4 = observation, 5 = interpretation, 6 = discussion, 7 = conclusion.
Data terms
20
Appearance
15
10
5
0
1
2
3
Volcan-
4 Paper sections Ocean(ic)
5
6
(earth)quake
Figure 20.2b Appearances of the data terms across paper sections.
7
Terms referencing data representations
20
Appearance
15
10
5
0
1
2
3
Data
4 Paper sections
5
Profile(s)
6
7
Figure(s)
Figure 20.2c Appearances of data referencing terms across paper sections. Directional terms
Appearance
15
10
5
0
1
2
South
3
4 Paper sections North
5
6
West
Figure 20.2d Appearances of directional terms across paper sections.
7
East
Rhetorical features of student science writing 277 Hypothesis 3: Multiple cohesive links were formed across the majority of the sentences forming the complete argument set in the technical paper. These links often included a set of key conceptual terms, introduced within the first few sentences and maintained throughout the papers. As in the earlier study multiple cohesive links were formed across the majority of the sentences forming the complete argument set in the technical paper. These links often included a set of key conceptual terms, introduced within the first few sentences and maintained throughout the papers. The aggregate results (see Table 20.2) of the analyses of the repeating terms in each paper reflect the same pattern of higher cohesiveness in the first two sections and the interpretation and conclusions. The methods and observations section had on aggregate fewer cohesive terms per sentence. The added and less well-defined discussion section did not conform so clearly to this pattern. Hypothesis 4: Sentences at the boundaries of sections and subsections tended to have denser cohesive links with other sections of the paper and tended to tie together semantic items of multiple epistemic levels. Our fourth initial hypothesis concerning the greater cohesiveness of boundary sentences was not confirmed in a direct form. However, the sample of 18 papers exhibited a more complex pattern. When examining the first sentence of each section we found different patterns in the different sections. In the first two sections (abstract and introduction) the introduction sentences did not differ from the cohesive pattern of the sentences in the whole section. This was measured by seeing whether the introductory sentence had more or fewer cohesive terms than the average of all sentences in that section. The third section (methods), however, reversed our expectation by having more introductory sentences with fewer cohesive terms than the section averages (11 lower, 7 higher). The final four sections, nonetheless, followed our original expectations, with the first sentence of each section having more cohesive terms than the sentences that follow. As shown in Table 20.3, for at least 11 of the 18 cases in each of the four sections, the lead sentence had more cohesive terms than the average number of cohesive terms per sentence in the section. This pattern echoes, in some respects, the pattern of epistemic levels and cohesive density revealed in the results of our tests of the first two hypotheses. Hypothesis 5: Often repeated terms built up cohesive density and thematic saliency as they were associated with other terms in different sections of the paper. In exploration of our fifth hypothesis we identified the terms that were used by the most student authors. Among the most prevalent terms, appearing in almost every paper, we identified four categories of words:
278 G. J. Kelly et al. terms referring to theoretical concepts (plate, subduct-[ing, ion], converg[ing, ence, ent]); terms referring to specific geologic formations directly reported in data (volcan-[o, ic], ocean (oceanic), (earth)quake); terms referring to aggregations of data (data, profile[s], fig[ures]); and terms of direction (south-, north-, -west, -east). When we analyzed the appearance of these terms across each of the sections of all 18 papers, distinctive patterns of distribution emerged for the different kinds of terms (see Figures 20.2a–d). The theoretical terms (see Figure 20.2a) consistently appeared in the first and last three sections. Although there was variation in sections 2–4, each had the fewest appearances in the methods section. The variation may be due to the difference in character among the terms with plate being part of name of the overarching theory (plate tectonics), with the other two terms describing processes identified by the theory. The data terms appeared consistently across all sections (Figure 20.2b), but the terms indicating aggregations of data were less prevalent in the first and last two sections. Terms making reference to data appeared most frequently in the methods and observations sections, although the terms considered showed significant variation in patterns of use (Figure 20.2c). The directional terms have the highest appearance in the observations section, and continue with comparatively high appearances throughout the last three sections (Figure 20.2d). In order to determine if the patterns we observed were consistent across the range of papers, we compared the epistemic and cohesive profiles of the four papers judged of highest quality by the course professor with the four papers rated lowest. We found that both groups fit the pattern equally well (Tables 20.1 and 20.2). Specifically, the average epistemic level of the statements in each section of the four highest and four lowest papers fit the epistemic pattern noted across all papers with roughly equivalent values, well within the standard deviation of the averages of all papers (Table 20.1). Further the average number of cohesive terms per sentence in each section in the four highest papers and the four lowest papers conformed to the general shape and values of all 18 papers, well within the standard deviation of the overall numbers (Table 20.2). We did not pursue for the high and low papers our other two measures (of first sentence cohesion and key-term appearance) because our method of analysis by appearance in each section did not lead to sufficient number of instances to produce meaningful results. Thus it appears that in this class for this assignment, students of both high and low achievement had sufficient genre knowledge of the assigned paper to produce sentences of the epistemic level and degree of cohesion appropriate to each section. The more theoretical sections of both high and low papers showed a higher epistemic level for claims and more cohesive terms per sentence than the methods and observations sections.
Rhetorical features of student science writing 279
General discussion All hypotheses and results reveal the same assignment-appropriate pattern of organization of argument, which we may characterize as the Epistemic U. Specifically the epistemic levels, the degree of cohesion, the cohesiveness of introductory sentences in the later sections, and the appearance of different kinds of repeated terms within the sections of the paper all conform to a single pattern consistent with the following arrangement of the sections of the paper: The overview and introduction present the material to be discussed at a higher theoretical level. The methodological section is more concrete. The observation section begins to report the data in patterned ways that reaches back up toward theory, and the final sections consider the reported data in theoretical terms. Greater theoretical orientation is associated not only with epistemic level, but degree of cohesion, the use of theoretical terms, and in the later sections, use of aggregating statements at the beginning of each section. These language patterns work together to structure the exposition of data in relation to the development of concepts. This rhetorical knowledge is apparently shared by all students sampled. From applied linguistic and rhetorical perspectives, such evidence of section by section structuring of several different kinds of features of language (level of claim, cohesiveness, use of aggregating introductory sentence, and use of particular kinds of lexical terms), points toward how aspects of language form are organized within genres. From the point of view of research in science education, our findings raise questions about how the students came to learn the genre conventions, whether such knowledge contributes to competence in the disciplinary domain, and whether there are other knowledges required for further participation and success in the relevant social-rhetorical practices. The first question regarding how all students came to know and use the same genre knowledge with respect to epistemic levels and cohesion in this case appears to result from explicit instruction in the genre organization and epistemic levels of the paper. As a result of the findings of Kelly and Takao (based on the 1998 iteration of the course) and related investigations and consultations, the instructor added explicit instruction in the laboratory manual concerning epistemic level of claim and provided examples of poorly written and well-written student papers in this regard. Students in the 2001 version of the course studied here were given exercises to assess the epistemic level of each claim in weaker and stronger sample papers (Takao, Prothero, & Kelly, 2002). This material was reinforced in lecture and laboratory section with the teaching assistants. While there was some discussion of developing coherent arguments and using appropriate vocabulary consistently, there was not explicit instruction of lexical cohesion, nor was that term introduced. The patterns of cohesion and terminological appearance may then be to a greater or lesser extent cognitive and/or formal consequences of the scaffold established by the assigned and instructed form.
280 G. J. Kelly et al. The implication would be that students learn what they are taught, but this still leaves open the question if what they were taught was significant and sufficient to lead students to scientific reasoning. The question remains whether instruction in the genre leads students to produce substantive reasoning or merely empty formalisms? If it is of value, are there other crucial elements of the genre to be taught? And what other reasoning skills that are not implicit in the genre also need to be taught? The significance of the genre knowledge for reasoning may be assessed by investigating the relationship between this knowledge and other measures of scientific reasoning. Such a measurement of scientific reasoning would have to be sensitive to the task-specific features of the student writing, which include familiarity and facility with the relevant geologic theory. Furthermore, if, as we argue in this chapter, scientific reasoning is associated with the form of the argument through which it is realized, it is hard to disambiguate tests of reasoning from genre knowledge.
Conclusion The arguments presented in student responses to the assignment studied uniformly exhibit a clear epistemic and cohesive structure that corresponds to the sections of the assigned paper. This public display of structured reasoning through students’ written work provides an important means of knowing how students reason scientifically in a specific subject and problem context. Further, analysis of assigned writing is a non-obtrusive and ecologically valid procedure, which builds upon the already existing concerted activity of the members of the classroom. Researching scientific reasoning in this way allows for recursive change in the instruction, since a major goal of instruction is the successful production of reasoned argument in the form of these assigned papers. In the case of this study, the recursive effect is further facilitated by the inclusion of the instructor in the research team. Increased knowledge of the form and substance of successful argument holds the promise of improving reasoning with scientific evidence and concepts through more informed instruction.
References Bazerman, C. (1988). Shaping written knowledge: The genre and activity of the experimental article in science. Madison, WI: University of Wisconsin Press. Bazerman, C. (1993). Forums of validation and forms of knowledge: The magical rhetoric of Otto von Guericke’s Sulfur Globe. Configurations, 1, 201–228. Bazerman, C. (1994). Systems of genre and the enactment of social intentions. In A. Freedman & P. Medway (Eds.), Genre and the new rhetoric (pp. 79–101). London: Taylor & Francis. Bazerman, C. (1997). Discursively structured activities. Mind, Culture and Activity, 4(4), 296–308. Blakeslee, A. (2001). Interacting with audiences. Mahwah, NJ: Erlbaum.
Rhetorical features of student science writing 281 Britton, J., Burgess, T., Martin, N., McLeod, A., & Rosen, H. (1975). The development of writing abilities (11–18). London: Macmillan. Chandler, J., Davidson, A. I., & Harootunian, H. (1991). Questions of evidence: Proof, practice, and persuasion across the disciplines. Chicago, IL: University of Chicago Press. Chinn, P. W. U., & Hilgers, T. L. (2000). From corrector to collaborator: The range of instructor roles in writing-based natural and applied science classes. Journal of Research in Science Teaching, 37, 3–25. Emig, J. (1977). Writing as a mode of learning. College Composition and Communication, 28, 122–128. Erduran, S., & Jimenez-Aleixandre, M. P. (Eds.). (2008). Argumentation in science education: Recent developments and future directions. New York: Springer. Fleck, L. (1979). Genesis and development of a scientific fact. (F. Bradley & T. J. Trenn, Trans.). Chicago, IL: University of Chicago Press. Freedman, A., & Medway, P. (Eds.). (1994). Genre and the new rhetoric. London: Taylor & Francis. Goodwin, C. (1995). Seeing in depth. Social Studies of Science, 25, 237–274. Halliday, M. A. K. & Hasan, R. (1976). Cohesion in English. Longman: London. Herrington, A. (1985). Writing in academic settings: A study of the contexts for writing in two college chemical engineering courses. Research in the Teaching of English, 19, 331–61. Hoey, M. (1983). On the surface of discourse. London: George Allen & Unwin. Hoey, M. (1991). Patterns of lexis in text. Oxford: Oxford University Press. Kelly, G. J., & Bazerman, C. (2003). How students argue scientific claims: A rhetorical-semantic analysis. Applied Linguistics, 24(1), 28–55. Kelly, G. J., Chen, C., & Prothero, W. (2000). The epistemological framing of a discipline: Writing science in university oceanography. Journal of Research in Science Teaching, 37, 691–718. Kelly, G. J., Regev, J., & Prothero, W. A. (2008). Analysis of lines of reasoning in written argumentation. In S. Erduran & M. P. Jimenez-Aleixandre (Eds.), Argumentation in science education: Recent developments and future directions, (pp. 137–157). New York: Springer. Kelly, G. J., & Takao, A. (2002). Epistemic levels in argument: An analysis of university oceanography students’ use of evidence in writing. Science Education, 86, 314–342. Keys, C. W. (1999). Revitalizing instruction in scientific genres: Connecting knowledge production with writing to learn in science. Science Education, 83, 115–130. Knorr-Cetina, K. (1999). Epistemic cultures: How the sciences make knowledge. Cambridge, MA: Harvard University Press. Latour, B. (1987). Science in action: How to follow scientists and engineers through society. Cambridge, MA: Harvard University Press. Lynch, M., & Woolgar, S. (Eds.). (1990). Representation in science practice. Cambridge, MA: MIT Press. Myers, G. (1990). Writing biology: Texts in the social construction of scientific knowledge. Madison, WI: University of Wisconsin Press. Pera, M., & Shea, W. (Eds.). (1991). Persuading science: The art of scientific rhetoric. Canton, MA: Science History Publications. Prior, P. (1998). Writing/disciplinarity. Mahwah, NJ: Erlbaum. Prothero, W. A. (1995). Taming the large oceanography class. Journal of Geological Education, 43, 497–506.
282 G. J. Kelly et al. Prothero, W. A. (2001). Student manual: Introduction to oceanography. University of California, Santa Barbara. Prothero, W., & Kelly, G. J. (2008). Earth data, science writing, and peer review in a large general education oceanography class. Journal of Geoscience Education, 56(1), 61–72. Rivard, L. P., & Straw, S. B. (2000). The effect of talk and writing on learning science: An exploratory study. Science Education, 84, 566–593. Sampson, V., & Clark, D. B. (2008). Assessment of the ways students generate arguments in science education: Current perspectives and recommendations. Science Education, 92, 447–472. Swales, J. M. (1990). Genre analysis. Cambridge: Cambridge University Press. Swales, J. (1998). Other floors, other voices. Mahwah, NJ: Erlbaum. Takao, A. Y., Prothero, W., & Kelly, G. J. (2002). Applying argumentation analysis to assess the quality of university oceanography students’ scientific writing. Journal of Geoscience Education, 50(1), 40–48. van Nostrand, A. D. (1997). Fundable knowledge: The marketing of defense technology. Mahwah, NJ: Erlbaum. Wallace, C. S., Hand, B., & Prain, V. (2004). Writing and learning in the science classroom. Dordrecht: Kluwer. Walvoord, B. E., & McCarthy, L. P. (1990). Thinking and writing in college: A naturalistic study of students in four disciplines. Urbana, IL: National Council of Teachers of English.
21 Reading and writing in the social sciences in Argentine universities1 Paula Carlino CONICET – University of Buenos Aires, Argentina
The scholarship and teaching of writing in the disciplines are endeavors only recently undertaken in Argentine universities. Most related research tended to focus on undergraduates’ difficulties in reading and writing in college. In contrast, the present study has arisen from the relevance to our context of the contributions of North American “writing across the curriculum” (WAC) (Bazerman et al., 2005; Nelson, 2001; Russell, 1990) and “writing in the disciplines” (WID) (Hjortshoj, 2001; Monroe, 2003), as well as English (Lea & Stree, 1998; Lillis, 1999) and Australian (Chanock, 2004; Vardi, 2000) “academic literacies” research. WAC and WID emphasize college instruction to promote learning, while academic literacies studies direct their attention to the institutional power relationships between what teachers and students do, think, and expect regarding written assignments. A further and congruent theoretical root for the present work is an Argentine constructivist approach, the “didactic of language practices” (Ferreiro, 2001; Kaufman, 2004; Lerner, 2003; Nemirovsky, 1999).
Context of the research Argentina has 39 public universities and 43 private ones, with 1,304,000 and 279,000 undergraduates respectively (in a country with a population of 39 million). They greatly vary in size from 358,000 undergraduates in the University of Buenos Aires to fewer than 1,500 in the smallest and newest institution (Anuario, 2007). The gross schooling rate for higher education in Argentina was 68.6 percent in 2006 (Anuario, 2007). While this enrollment rate is the highest in Latin America, the Argentine tertiary system is said to be increasingly inefficient. Estimates are that the freshman dropout rate is about 50 percent and that only 20 percent of the university students finally graduate (Marquís & Toribio, 2006). Public universities tend to be the most prestigious ones with the 82 percent of the university population. Undergraduate studies are completely free and most departments have not required a placement or admissions test since 1983, the year of the recovery of democracy. While this open tradition of Argentine public higher education has favored the access of many
284 P. Carlino working-class undergraduates, it is also true that this unrestricted entry policy does not guarantee their progress in their studies or their degree completion. Other open-access policies, like student financial aid, orientation programs, or student support services, are rare. Most teachers are part-time teaching assistants and are not well paid; teacher development through university programs tends to be infrequent. In the social sciences and humanities, reading and writing are usually required but academic literacy skills are not taught explicitly. Undergraduates are asked to read from sources, and to write the answers to exam questions during class hours, or to write ill-defined essays (called “monografías”) at home, once or twice a semester, for assessment purposes. Classes greatly differ in size but could reach 50 and even more undergraduates in tutorials (and 300 in lectures). Neither US-style “writing centers,” “writing tutors,” “WAC or WID programs,” nor Australian-style “Teaching and learning units” or “Language and academic skills advisors” exist. Teachers complain that “students can’t write, they don’t understand what they read, they don’t read.” Undergraduates’ reading and writing problems make headlines every year. My previous research Three partially overlapping stages of inquiry led to the current study. The first treated academic writing as a cognitive skill and researched, through draft analysis, how undergraduates’ texts were revised during an exam. The second stage proceeded from the difference found between these students’ revisions compared to those of French and North Americans which had been reported in the literature that inspired my study (Hayes & Flower, 1986; Piolat, Roussey, & Fleury, 1994). I realized that this difference was not cognitive but cultural, and attributed it to the dissimilarities of national instructional experiences regarding writing. This gave rise to a comparative study in which I “discovered” realities previously unknown within Latin American literature, such as the Australian teaching and learning units and teacher-development programs, and the North American writing centers, writing intensive courses, as well as the WAC/WID and academic literacies contributions. The third line of work was a six- year action-research project that tried out several reading and writing tasks in psychology and education courses to increase student participation in class and enhance their cognitive action over the learning material. I published the results of the latter two lines of inquiry to promote the necessity of reading and writing support in any university course (e.g., Carlino, 2005a; Carlino, 2005b). Nevertheless, more data were needed. The research in this chapter aims to provide this kind of data.
Reading and writing in Argentine universities 285
The study Research questions and method What are the undergraduates’ and teachers’ perspectives about literacy practices that take place in social sciences courses? How do teachers respond to students’ written assignments? To answer these and other questions, a team of four researchers besides myself2 have begun a qualitative inquiry, funded by a grant from the National Council of Scientific and Technical Research of Argentina. The study has so far comprised ten social sciences subjects in three public universities through two focus groups with 45 freshmen after which they individually wrote about “What are the usual reading and writing assignments in high school?” and “What do you find new in university literacy tasks?” We also carried out in-depth semi- structured interviews with 15 teachers and 21 undergraduates (see Appendix A). An original device found helpful at the end of some interviews was showing the interviewee a set of cards with written accounts of hypothetical classes showing different kinds of writing to learn and learning to write support. Students and teachers were asked whether they resembled their own classes, whether they found them useful, and, in case they were not frequent in their experiences, why it happened. Alternatively, in other interviews, we requested them to show us an already assessed exam or essay, and inquired about the meanings they gave to the teachers’ written feedback on students’ work. In addition, we examined syllabi searching for what was said about reading and writing for each course. Inspired by a Lea and Street (1998) research design, we have not intended a representative sample of the whole universe, but a corpus of perspectives in which to explore and specify our initial hypothesis about the institutional experiences we were surrounded by. We aimed to apprehend and objectify everyday practices that appeared as transparent, natural, and even necessary, to make them observable by their actors and stakeholders. The ultimate goal of our study was to open them to critique. Findings Reading and writing assignments are ubiquitous in social science courses but tend to go unnoticed: they do not appear in the subjects’ syllabi and they are not explained by the teachers. Instead, they are taken for granted. Teachers’ and students’ perspectives reveal that: I Literacy practices in Argentine universities are new and challenging to undergraduates because they greatly differ from modes of reading and writing required in high school. II In spite of (I), teachers in the disciplines do not make college-level expectations explicit; guidelines are rare and feedback is minimal.
286 P. Carlino III Most of the teachers and students interviewed think that reading and writing in the disciplines should not be an object of instruction within the university. IV Within a small proportion of interviewees, we found a contradiction: while it is generally claimed that literacy instruction is inappropriate for university, at the same time there are a few teachers who do address undergraduates’ reading and writing without acknowledging that this helps them to improve their literacy. V Some of our interviewees attribute teachers’ disregard of literacy practices to institutional limitations. VI While institutional constraints need to be reconsidered by stakeholders, the pervasive assumptions referred to in (III) also hinder teachers’ taking care of writing and reading in their subjects. I will quote some interviewees to illustrate our findings.3
I College literacy practices are challenging to students because they greatly differ from modes of reading and writing required in high school. Undergraduates state that high school reading for writing demands just looking for what questionnaires ask and transcribing literal portions of text. Instead, in college students need to make inferences about the text as a whole and in relation to other texts: Student: . . .
in high school, you don’t have to read, [instead] you are asked to answer questions. You are given a questionnaire and teachers ask you to answer it. Interviewer: And how do you do it? S: Oh, the old story of the questionnaire! It is very silly: “Let’s see . . . this answer is . . . here” [she points at some part of a text] i: And what about college? S: No. Not in college. In college, you are supposed to read. (first-year Education student) I get lost because [in college reading] the inferences you need to make, extract, are not written anywhere [within the text] . . . So, it’s sometimes difficult to know whether they are right. (first-year Psychology student) College writing from sources confronts students with a new way of reading that requires them to compare different points of view about the same issue and to take into account the relationships between authors’ stances. There are no absolute truths, i.e., facts to rehearse like in high school, but several claims and arguments for each topic:
Reading and writing in Argentine universities 287 [A question posed by the teacher] asked, “What does Althusser add to Marx?”. Oh, so, I’ve only just known that I have to study Marx and Althusser together, because they complement each other, but I had seen them separately. With this question, I already know that I will be asked about their relationship, but without it, I don’t know. (first-year Education student) Teachers’ accounts agree in that undergraduates get lost when reading from college texts. What they do not know is that students were used to surface reading in high school assignments and that probably they try to read for university but, without understanding, some give up: students have too many difficulties to see what is important in the readings. They especially find it very difficult to extract what is relevant for the subject. (teaching assistant, first year, Sociology) Interviewer: What do undergraduates do when they read? Professor: Nothing, students do nothing when they read,
nothing and
nothing. Students don’t read. (full professor, first year, Sociology)
II Even though “students’ problems” are recognized, this does not imply that their learning needs are taken into account. On the contrary, teachers in the disciplines tend to ignore the improvement of student literacy. Neither do their classes include reading and writing as tools for conceptual learning. Expectations are not explicit, oral or written guidelines are rare, and feedback is scarce: Interviewer: Does your subject work with any reading guide? Teacher: No! . . . giving them a reading guide, no way! It makes
no sense, [texts] are clear. If they don’t understand them, I want students to tell me “this is not clear, would you explain it to me?” (teaching assistant, first year, Psychology)
In all courses, you are required to structure texts [when you write], to be clear, but this is what you are asked for, but teachers don’t explain anything [about how to achieve this]. . . . Teachers don’t tell you how to include quotes or references; you are supposed to know it already or to find it out by yourself. (first-year Fine Arts student) Undergraduates’ writing is mostly required for assessment purposes but it is not considered a learning tool. This is evident by the scarcity of guidelines and also by the minimal teachers’ feedback that students receive after-
288 P. Carlino wards. Both situations are considered “normal,” although some students complain because they perceive the learning opportunities they are missing: Usually, the exam always comes back with a check mark and a grade. Very rarely does the teacher guide you through her/his assessment. . . . What they do is to underline what is wrong . . . She either makes a check mark or underlines [your work], and [in the latter case] you know that it is wrong but you don’t know why, whether it’s unnecessary, it’s the opposite, or what. (third-year Law student) We don’t receive much feedback but we do get those marginal comments “incomplete” or “concepts missing”. Of course, they don’t specify . . . (second-year Social Work student) Unspecific and ambiguous written feedback of this sort, interchangeable among student papers, serves more to justify the grade (Hjortshoj, 1996; Mosher, 1997; Sommers, 1982) than to help undergraduates’ elaboration of meanings, or understanding mistakes and learning how to overcome them. In the previous quote, the use of “of course” denotes that this kind of feedback is a generally instituted practice that everybody knows (and expects) to happen. In spite of this habitual experience, students’ wording and intonation subtly criticize it as teachers’ carelessness: This is the only [teacher] who clarifies each item [each question asked], how many points it is worth. That’s why I’ve brought it with me, because it makes the grade explicit. But it has nothing. I mean, there’s nothing written [no feedback from the teacher]. (fourth-year Psycho-pedagogy student) Besides the interviews, we examined a corpus of syllabi from social sciences courses. These tend to consist of a list of disciplinary topics paired with the required readings, the course’s objectives and, occasionally, the intended methodology that teachers would implement in their classes. They also specify the number of assessment tasks of the subject. Most syllabi do not mention literacy at all. Nevertheless, student writing is implied when assessment is noted and reading is suggested by the word “Bibliography,” which precedes the reading list. A rare example of a syllabus where writing is explicitly named just says: “The evaluation of the course will be through an individual written exam during class time . . . and an assignment consisting of an individual conceptual synthesis, written at home” (Psychology syllabus).
III Most of the teachers and undergraduates interviewed think that reading and writing in the disciplines should not be an object of instruction. Some common-sense assumptions appear to justify this claim.
Reading and writing in Argentine universities 289 The first one views reading as extracting a pre-given meaning from a text. That would be why there is no apparent need to address it. It is presupposed that students already have this general ability. Instead, if reading were recognized as a process of co-constructing meaning through the interaction between the text and the reader’s disciplinary purposes and knowledge, teachers could make explicit these latter, which are unfamiliar for undergraduates. Likewise, writing appears as a surface medium of communication to convey already made thoughts and does not constitute the elaboration of substantial meaning relevant for a field of study. In this approach, taking care of writing would be emphasizing textual features at the sentence level and correcting errors because writing is viewed “as a textual product rather than an intellectual process” (Carter, Miller, & Penrose, 1998). Within this framework, it is unnecessary to continue learning and teaching to read and to write for college because both activities are regarded as the prolongation of generalizable skills previously “learnt outside a disciplinary matrix” (Russell, 1990). For these reasons, teachers like the one in this transcript made it clear that they did not consider writing instruction to be part of their job: Interviewer: Do
you think it’s your duty to teach them to write [in your discipline]? Teacher: No! I: Why not? T: Because I have to teach them the discipline . . . They should have learnt [to write] better in high school. I: And how do you think they could have learnt to write a text of the quality you have told me that you require? T: I suppose they already know what a good text is. I: And where could they have known it from? T: [He laughs] Very good question. . . . From previous subjects, because this is a second year course . . . I think they should have writing courses . . . with specialized, Literature teachers. (teaching assistant, second year, Work Relations) It is interesting to note that this teacher changes his ideas during the course of the interview. He first asserts that it is not his duty to address students’ writing because he assumes that writing should have been learnt in the previous educational level. Then, the interviewer reminds him about another part of the interview when the teacher had specified the properties of what he considered a good text for his discipline. At this moment, the teacher laughs because he suddenly realizes that nobody has ever taught his students about it. Consequently, he recognizes that undergraduates have some learning needs that he believes he could not address in his instruction and he suggests that other courses, with specialized teachers, do so. The second belief invoked for disregarding students’ literacy considers that undergraduates are or ought to be autonomous (Chanock, 2001).
290 P. Carlino Being adult is equated with being knowledgeable, in an equation that, by virtue of the first assumption, age warrants familiarity with what are supposed to be previously learnt transferable reading and writing skills: I believe they are university students, and that’s why they are responsible for what they decide to do. . . . I don’t guide them [in their reading] because I understand they’re university students and they have to decide for themselves. . . . I leave them alone because I want them to make their own [reading] journey and that they decide for themselves. (teaching assistant, first year, Psychology) Even the students assume that they are old enough not to receive reading guidance: “The teacher goes and lectures . . . She/he says: ‘read these texts’, and that’s all, it’s up to you. [Undergraduates] are mature and, frankly, [the way you read] depends on you” (first-year Fine Arts student). We see a third underlying belief behind the claim that it is not the university teachers’ duty to deal with reading and writing. Both teachers and students sustain a restricted model about the instructional process and its object. Teaching in the social sciences is conceived as merely lecturing to explain concepts. Accordingly, teachers’ role does not consist of scaffolding (guiding and responding to) new activities so that students can progressively acquire them. Learning is seen as passively internalizing a pre-given meaning rather than assuming risks through taking part in literacy tasks. This also means that the object of instruction is looked at as a piece of information or as a body of declarative knowledge. Tacit or procedural knowledge, as implied within unfamiliar disciplinary literacy practices, is not taken into account. Similarly, undergraduates tend to expect that classes be organized around teachers orally communicating some information and undergraduates receiving it. Other class dynamics are frequently seen as a waste of time: [The interviewer shows a card with a written account of a class where students work in pairs with their written drafts.] s: Make a draft and work it with a peer, revise it between the two, and then within the whole group. This would be helpful, yes, at least to discuss about the topic. It would be good if classes were smaller, if there wouldn’t be 80 students in a class . . . i: Do you think this is not done because classes are too large? s: Yes, if we were 20, [there would be] 10 drafts to revise [during the class]. So, when will the class start? (second-year Literature student; italics added) In the previous quote, the student acknowledges that intertwining writing with oral discussions and receiving feedback is very difficult in large classes. But she also demonstrates that she would not consider it a class because what she expects is listening to the teacher:
Reading and writing in Argentine universities 291 I’d love working in small groups with peers, say two, three, and being able to discuss a lot of things for college. Instead, in class, I go to incorporate knowledge. . . . It would be nice [to work in groups], but I’m very used to working alone and . . . I like to work alone . . . I go to class, and I want to take notes, and then I will go to study them. (second-year Literature student; italics added)
IV We found an apparent contradiction between saying and doing in some interviewees. While both teachers and undergraduates generally claim that teaching reading and writing is inappropriate for higher education and should not be an object of instruction, at the same time there are a few teachers who do address students’ literacy, as it appears both in students’ and teachers’ accounts: sometimes, for instance, we work with these steps towards the hypothesis [hypothesis = elaboration of a written idea that unifies an analysis of some arts work]. We do it orally among the whole class, . . . and then we tell students that in groups they write the hypothesis and read them aloud. And perhaps other group justifies them or other group asks them questions or we talk about why that hypothesis is right or wrong . . . They bring a written paper from their home and what we do in class is that: read everybody’s written papers, and discuss them. (teaching assistant, second year, Media History) Students greatly appreciate when they receive this kind of teacher’s support and feedback because it helps them to understand what is expected from them: i: Do
you find that receiving or not receiving written feedback from the teachers makes any difference? s: Yes, it’s quite different. Because . . . if you just receive a mark, but you’re not told what’s wrong in your work, where you have failed or what [the teacher] expects . . . so, you can’t . . ., I mean, it’s like a guide when the teacher writes on your paper and explains. (second-year Social Work student) There are courses where you’re given a reading guide, which is quite helpful because you know a bit more about what you have to pay attention to [in the texts] and what you skip. (first-year Psychology student) Even if students appreciate this infrequent literacy support, they assert it is not writing or reading instruction. Likewise, teachers who offer it do not acknowledge that this helps students to improve their literacy. They just
292 P. Carlino take it as a way of teaching their subject. This apparent contradiction can be understood if we take into account the unsupported assumptions mentioned in III about the nature of literacy, of undergraduates, and of the object of teaching.
V Students and teachers were asked why they believe that most teachers in the disciplines do not address literacy tasks within their subjects. They attribute it to institutional constraints: (a) scarce class time and teachers’ paid time, (b) too many students per class, and (c) lack of teachers’ training.
VI While institutional limitations need to be taken into account by stakeholders, the widespread assumptions referred to in III also prevent teachers from including writing and reading in their subjects. These beliefs “can have motivational force because [they] not only label and describe the world but also set forth goals (both conscious and unconscious) and elicit or include desires” (Strauss, 1992, p. 3, emphasis original, in Curry, 2002). However, they pass unnoticed, because of their common-sense status that has rendered them “natural.”
Discussion This research was born from a need to promote a local debate and justify, with empirical data, the need for literacy teaching across the disciplines in Argentine and Latin American universities. Nonetheless, its theoretical roots, grounded in the North American, Australian, and British contributions, encourage a wider dialogue as well. The present study suggests that Argentine universities neglect undergraduates’ reading and writing to learn the disciplines even more than in the English-language world. Some widespread ideas behind this situation tend to be similar: an unsupported conception of literacy (Bogel & Hjortshoj, 1984; Carter et al., 1998; Creme & Lea, 1998; Lea & Street, 1998; Russell, 1990), and a questionable notion of undergraduates’ autonomy (Chanock, 2001, 2003). Our study offers a further exploration of the prevalent beliefs about the nature of reading and instruction. This set of unsupported assumptions, also labeled discourses (Gee, 1990; Ivanic, 2004), myths (Creme & Lea, 1998), implicit models (Lea & Street, 1998), approaches (Lillis, 2003), and tacit theories (Gee, 1990), “lead to particular forms of social action, . . . decisions, . . . choices, and omissions” (Ivanic, 2004, p. 124). Together with factual institutional constraints, they prevent teachers from responding to students’ educational needs and disempower (Gee, 1990) or handicap “non traditional” students, for whom the confusion from not receiving guidance is maximized (Lillis, 1999). Our results also strengthen the constructivist “didactic of
Reading and writing in Argentine universities 293 language practices” approach (e.g., Lerner et al., 2003), which claims for teaching (as sharing and making explicit) the required literacy to take part in school and society, instead of blaming students for what they do not know yet.
Appendix A: basic questions around which interviews were conducted Students 1 How did you read in high school?/What were the reading assignments you had? 2 How do you read in college?/What are the reading assignments you have now? What do you find difficult while reading in college? Give me an example. 3 Do your teachers support your reading? How? Why? (The same for writing) 1 When you write an assignment, an exam, etc: What is teacher feedback like? Show me an exam or essay already assessed. 2 (With the view of an exam or essay already marked) Why do you think your teacher ticked/underlined/wrote this? Do you find it helpful? Why? 3 Before the exam or written assignment, how did the teacher tell what you had to do? Teachers 1 How do your students write in college?/What are the writing assignments you give them? Do you work with them in class? How? Give me an example. 2 What do they find difficult while writing in college? 3 Do you give them any support for writing? Describe. 4 Do you think taking care of students’ literacy is part of your job as a discipline teacher? Why? (The same for reading) 1 When you assess your students writing: What is your feedback like? Show me an exam or essay already assessed. 2 (With the view of an exam or essay already marked) What have you ticked/underlined/written this for? Do your students find it helpful? Why? 3 Before the exam or written assignment, how did you tell the students what they had to do?
294 P. Carlino
Notes 1 Thanks to Julian Hermida, Assistant Professor of Law at Algoma University (Canada), for his generous help with the editing of the English manuscript. 2 Graciela Fernández, Viviana Estienne, Emilse Diment, and Silvia Di Benedetto. 3 The interviewees’ comments as well as the interview questions have been translated from Spanish into English for this chapter.
References Anuario de Estadísticas Universitarias de la República Argentina 2006. (2007). Coordinación de Investigaciones e Información Estadística, Ministerio de Educación: Buenos Aires. Retrieved May 2, 2008, from www.mcye.gov.ar/spu/ documentos/Anuario-2006.zip. Bazerman, C., Little, J., Bethel, L., Chavkin, T., Fouquette, D., & Garufis, J. (2005). Reference guide to writing across the curriculum. West Lafayette, IN: Parlor Press. Retrieved October 2, 2008, from http://wac.colostate.edu/books/ bazerman_wac. Bogel, F., & Hjortshoj, K. (1984). Composition theory and the curriculum. In F. Bogel & K. Gottschalk (Eds.), Teaching prose. A guide for writing instructors (pp. 1–19). New York: Norton. Carlino, P. (2005a). Escribir, leer, y aprender en la universidad. Una introducción a la alfabetización académica. Buenos Aires: Fondo de Cultura Económica. Carlino, P. (2005b). Representaciones sobre la escritura y formas de enseñarla en universidades de América del Norte. Revista de Educación, 336, 143–168. Retrieved October 2, 2008, from www.revistaeducacion.mec.es/re336_09.htm. Carter, M., Miller, C., & Penrose, A. (1998). Effective composition instruction: What does the research show? Center for Communication in Science, Technology and Management, Publication Series, no. 3, April, North Carolina State University. Retrieved May 2, 2008, from www.chass.ncsu.edu/ccstm/PUBS/no3/ index.html. Chanock, K. (2001). From mystery to mastery. Changing identities: Proceedings of the 2001 Australian Language and Academic Skills Conference, University of Wollongong, New South Wales, November 29–30, 2001. Retrieved October 2, 2008, from http://learning.uow.edu.au/LAS2001/selected/chanock.pdf. Chanock, K. (2003). Autonomy and responsibility: Same or different? Proceedings of the Independent Learning Conference, University of Melbourne, Australia, September 13–14, 2003. Retrieved October 2, 2008, from http://independentlearning.org/ILA/ila03/ila03_chanock.pdf?q=ila03/ila03_chanock.pdf. Chanock, K. (2004). Introducing students to the culture of enquiry in an arts degree. Milperra: HERDSA. Creme, P. H., & Lea, M. (1998). Student writing: Challenging the myths. Proceedings of the fifth annual Writing Development in Higher Education Conference, Centre for Applied Language Studies, University of Reading, April 1–2, 1998. Retrieved July 2, 2008, from www.rdg.ac.uk/AcaDepts/cl/CALS/wdhe98/cremelea.html. Curry, M. J. (2002). Cultural models in the U.S. writing classroom: Matches and mismatches. In M. Graal (Ed.), Writing development in higher education: Changing contexts for teaching and learning (pp. 45–61). Leicester: University of Leicester.
Reading and writing in Argentine universities 295 Ferreiro, E. (2001). Pasado y presente de los verbos leer y escribir. Mexico: Fondo de Cultura Económica. Gee, J. (1990). Social linguistics and literacies: Ideology in discourses. London: Falmer Press. Hayes, J., & Flower, L. (1986). Writing research and the writer. American Psychologist, 41(10), 1106–1113. Hjortshoj, K. (1996). Responding to student writing. New York: Knight Writing Program, Cornell University. Hjortshoj, K. (2001). The transition to college writing. Boston, MA: Bedford/ St. Martin’s. Holm-Nielsen, L., Thorn, K., Brunner, J., & Balán, J. (2005). Regional and international challenges to higher education in Latin America. In H. de Wit, I. Jaramillo, J. Gacel-Ávila, & J. Knight (Eds.), Higher education in Latin America. The international dimension (pp. 39–69). Washington, DC: The World Bank. Retrieved May 2, 2008, from http://siteresources.worldbank.org/EDUCATION/ Resources/278200-1099079877269/547664-1099079956815/LAC_universities_ wps4002.pdf. Ivanic, R. (2004). Discourses of writing and learning to write. Language and Education, 18(3), 220–244. Kaufman, A. M. (2004). La inclusión de los niños en el mundo letrado: Un desafío para la escuela. In S. Gvirtz & M. E. de Podestá (Eds.), Mejorar la escuela. Acerca de la gestión y la enseñanza. Buenos Aires: Granica. Lea, M. & Street, B. (1998). Student writing in higher education: An academic literacies approach. Studies in Higher Education, 23(2), 157–172. Lerner, D. (2003). Leer y escribir en la escuela. Lo real, lo posible y lo necesario. Mexico: Fondo de Cultura Económica. Lillis, T. (1999). Whose “common sense”? Essayist literacy and the institutional practice of mystery. In C. Jones, J. Turner, & B. Street (Eds.), Students writing in the university. Cultural and epistemological issues (pp. 127–147). Amsterdam: John Benjamins. Marquís, C., & Toribio, D. (2006). Informe sobre la educación superior en Iberoamérica, 2006. Capítulo sobre la Argentina. Santiago de Chile: Centro Interuniversitario de Desarrollo. Retrieved June 2, 2008, from www.cinda.cl/pdf/ INFORME%20ARGENTINA.pdf. Monroe, J. (2003). Writing and the disciplines. Peer Review, Autumn, 4–7. Mosher, J. (1997). Responding to students papers. Responses to avoid and productive advice to give. Teaching with Writing, 7(1). Retrieved January 2, 2008, from www.boisestate.edu/wcenter/ww9091.htm. Nelson, N. (2001). Writing to learn: One theory, two rationales. In P. Tynjälä, L. Mason, & K. Lonka (Eds.), Writing as a learning tool: Integrating theory and practice (pp. 23–36). Dordrecht: Kluwer Academic Publishers. Nemirovsky, M. (1999). Sobre la enseñanza del lenguaje escrito . . . y temas aledaños. Mexico: Paidós. Piolat, A., Roussey, J.-Y., & Fleury, P. H. (1994). Brouillons d’étudiants en situation d’examen. Le français aujourd’hui, 108, 39–49. Russell, D. (1990). Writing across the curriculum in historical perspective: Toward a social interpretation. College English, 52(January), 52–73. Sommers, N. (1982). Responding to student writing. College Composition and Communication, 33(2), 148–156.
296 P. Carlino Vardi, I. (2000). What lecturers want: An investigation of lecturers’ expectations in first year essay writing tasks. Proceedings of The Forth Pacific Rim, First Year in Higher Education Conference 2000: Creating Futures for a New Millennium, Queensland University of Technology, Brisbane, July 5–7, 2000. Retrieved April 2, 2008, from www.fyhe.qut.edu.au/past_papers/papers/VardiPaper.doc.
22 Preparing students to write A case study of the role played by student questions in their quest to understand how to write an assignment in economics Barbara Wake University of Adelaide, Australia
Introduction This chapter reports on the introduction of dialogic learning, as demonstrated in classroom discussion, into a final-year economics program at an Australian university. The rationale for using dialogic learning was to ensure the academic success of an increasingly important student cohort in Australian university education, second-language international students. Despite calls for university education in Australia to engage students in more interactive learning environments (Biggs, 1996, 2003; Trigwell, Prosser, & Waterhouse, 1999; among others), any formalized non-virtual dialogic negotiation of meaning often plays a minimal role, if indeed any role at all. It will be shown, by way of a case study, that by using extensive discussion the students were offered an environment which both encouraged their explorations of meaning and was tolerant of their learning difficulties. The particular aim of this approach was to provide a remedy for students’ need to copy model answers. As such, the students were able to explore uncertainties, seek clarifications, and to confirm interpretations. The chapter examines, in particular, the students’ attempts to negotiate the meaning of an economic model via their questions to the economics lecturer and class colleagues as they prepare to write a short-answer assignment. The chapter will show the kinds of linguistic strategies undertaken by the lecturer in response to the students’ questions and the impact these strategies had on the students’ learning and appropriation of meaning. The case study reveals that while the students’ learning was a highly collaborative process, any increments in understanding were overall devolutionary. Rather than moving toward new dimensions of abstract and metaphorical language to explain and exemplify economic phenomena, the lecturer necessarily shifted from theoretical syllogisms to more common-sense narratives to illustrate real-world economic activity. The data indicate these shifts were in response to the students’ confusions evident in their
298 B. Wake uestions. In their responses, the data reveal an interesting transition in the q students’ explanations; that is, they shifted between attempts to explain the theoretical principles and more congruent reasons indicating partial progress in their appropriation of the discourse. The study argues that without the assumed background knowledge, the complex nature of hypothetical causal explanations as a beginning point in the lecturer’s explanations was confounding for the students. The linguistic analysis indicates that the classroom discussion offered the students opportunities, a praxis, to remediate their background knowledge, even partially. In Vygotsky’s (1986, p. 150) terms, the students’ gradual control of meaning meant they were able to move beyond a mere “parrotlike repetition of words to cover up a vacuum.” This is evidenced by the advantages they took to pose questions and doggedly seek reasons. The article concludes by considering the effectiveness, or otherwise, of dialogic learning as a learning methodology for this student cohort. The contributions of this study are, first, to offer critical insights into how “doing economics” by way of discussion occurred. Rather than making intuitive assumptions about students’ background knowledge, or how a student’s cultural and linguistic background will be an automatic impediment to their learning, these findings reveal the realities of these students’ confusions with the meanings of the discourse. And second, the study demonstrates how the students were able to clarify their understanding in guided cooperative learning environments before undertaking their written assignments. In order to have theoretical validity, the analysis of the spoken data uses the multistratal resources of systemic-functional linguistic theory to analyze these strategies, in particular, the analytic categories for asking questions, as developed by Hasan (1991). The advantage of adopting this approach lies in its view of language as constructing meaning. As Halliday (2004, p. 21) explains, all of the components of language are variants of a single motif: the organization of meaning in the grammar. Thus, the examination of the meanings sought in the students’ questions and those provided in the lecturer’s responses are acknowledged as being realized by the reciprocal relationship between lexis and grammar (lexicogrammar), semantics (or system of meaning), and the context (discourse). The discussion will take account, particularly, of the shifts in meaning in the students’ questions and in the lecturer’s responses. The study reported here is part of a more extensive research project on dialogic learning. For the sake of brevity, one student cohort from the original study will be the focus of this discussion. The particular cohort comprised five newly arrived Chinese-Malaysian students enrolled in a business degree program. They were all completing the final stage of a “twinning program,” an arrangement between the Australian university and an educational institution in Malaysia. The students had undertaken their first two years of study in Malaysia and then completed their final year in Australia. To ensure the success of this arrangement economically and
Preparing students to write 299 a cademically, all twinning program students were offered an elective subject for one semester in the form of an academic support program, the Business Communication Program (BCP). The curriculum provided intensive support in two compulsory subjects in their core degree programs: economics and accounting. The study proceeds from the proposition that any confusions these students experienced, and their need to copy model texts, were due to a lack of assumed background understanding of discipline theory, particularly in economics, rather than any intentional plagiarism or “violation of academic integrity.” Their impoverished background knowledge can be examined empirically in their language. The view taken, then, is that copying or mimicry, adopting a Vygotskian (1986, p. 94) perspective, is considered to be a natural transition from students’ intermental to intramental understanding as they acquire “the echoes and reverberations” of historical public discourses (Bakhtin, 1986, p. 91), particularly those considered to be unnegotiable authoritative discourses, such as economics. These experiences are shared by many students, but made more difficult for students whose academic as well as personal experiences are outside the main educational context. As such, this study lies within a social and linguistic rather than a psychological orientation toward tertiary education.
The challenge of academic economics The use of dialogic learning in the BCP rests on a recognition that the conceptual demands placed on students studying economics equate with Vygotsky’s (1986, p. 149) finding that the difficulty with educational knowledge lies in its “abstractness and detachment from reality.” This feature of educational knowledge is particularly apparent in the written discourse of academic economics—a discourse acknowledged as arcane for many students and particularly so for second-language students (Hewings, 1990; Mason, 1990). The complex nature of the discourse is evident when it is described as an a priori science rather than an observational science (Smith, 1989, p. 151). Economic theory, unlike science, has not been derived from the world, according to Wignell (1998, p. 323); the theory is used to interpret the world. The simplicity and precision of the hard sciences cannot be captured in the same way in economics, Marshall (1895) argued, due to the complex nature of human behavior. To illustrate this complexity, linguistic analyses conducted by Wignell (1997, 1998) and Wake (2006) of economic discourse reveal how explanations, unlike the physical sciences, typically begin by theorizing at higher levels of abstraction. Rather than shifts from material activities toward metaphor, generalizations, abstractions, and technicality, such as I’m working → we’re working → work → labor (Wignell, 1998), economic explanations remain at the most abstract level (labor) depending whether they are used to refer to work in general or labor in the abstract. The frequent
300 B. Wake encoding of the field, or subject matter, in visual images and symbolism, such as graphs and tables, further restricts any explicit relationship with material phenomena or physical activities.
The students’ discussion: attempting to use economic theory to explain real-world scenarios The economics assignment required students to write a short answer using equilibrium theory to explain the impact on the price of compact discs (CDs) if the trade restriction, parallel importing, were imposed on their importation into Australia. Parallel importing is a means by governments to protect local industries from overseas competition. The effect is that nobody other than music companies is able to import CDs into Australia, resulting in higher prices. Prior to writing their short answers as homework, students discussed the assignment in small groups in class over 50 minutes. The lecturer moved between each group to answer any questions about the task. The target student cohort discussed the assignment with the lecturer on two occasions, described as Phases 1 and 2 in the data. It is the shifts in meaning between Phases 1 and 2 that is the particular interest of this examination.
The spoken data The spoken data consist of transcripts of recordings of: (a) the students’ questions posed to each other and their responses, and (b) the students’ questions posed to the economics lecturer and her responses. The questions of interest are those in which the students attempt to understand the reasons for parallel importing. The rationale for choosing these particular questions is the significant changes in the kinds of responses given by the lecturer between Phases 1 and 2 as she realizes the students’ confusions. For easier comprehension, when elision occurs ( ) are used; any inaccurate utterances are corrected by using [ ].
Analysis of the students’ questions The analysis of the questions will use, as noted, the categories for asking questions, as developed by Hasan (1991). These categories will provide the linguistic means to investigate: 1 how the students sought to repair their impoverished background knowledge of the topic; 2 how they initially recast their questions to the lecturer to more theoretical construals; 3 how the students’ questions revealed their confusions;
Preparing students to write 301 4 how the lecturer’s responses made critical shifts toward more congruent meanings; 5 how prediction and consequence were reconstrued dialogically in the discussion, i.e., shifts from hypothetical syllogisms to narrative sequences between Phases 1 and 2; and 6 the impact the lecturer’s strategies had on the students’ appropriation of meaning.
Using analytic categories for asking questions According to Hasan’s (1991) analytic categories for asking questions, two kinds of information can be sought: explanations and confirmations, as shown in Figure 22.1. A speaker can elicit some more specific information by choosing the option [apprize], or they can elicit a yes/no response by choosing the option [confirm]. In this data, the students’ questions choosing the [apprize] option were critical in their attempts to define, explain, and illustrate the economic model. For example, the student Cin’s question: why isn’t there direct importing on the CD? selects the option [apprize]. Their questions choosing the [confirm] option clarify the accuracy or otherwise of their interpretations. For example, the student Li’s question: is it manufactured by the local producers is it? selects the [confirm] option.
1 Verify A Confirm Demand
2 Enquire
Asking questions 3 Precise Information
B Apprize
1a Probe
Endorse Validate
1b Reassure 2a Ask 2b Check 3a Explain
Reason
3b Specify
Nucleus
Method
Scope
4 Vague
Figure 22.1 Simplified network for asking questions indicating the focus on the [apprize] option in this data (source: adapted from Hasan, 1991).
302 B. Wake Within the limits of this discussion, the students’ questions of interest are those questions choosing the options [apprize:explain], [apprize:reason], [apprize:method], and [apprize:vague]: For a detailed analysis, including questions choosing the option [confirm], see Wake (2006). Questions choosing the [apprize] option Of the 80 questions posed by the students throughout the discussion, 27 chose the [apprize] option, 41 chose the [confirm] option. In contrast, the lecturer asked only seven questions. This pattern suggests that rather than the key concepts of economic theory being presented didactically, the classroom discussion offered the students opportunities to heuristically explore complex meanings of the discourse. The option [apprize] The relevant features of questions choosing the options [apprize:precise], [apprize:explain], [apprize:reason], [apprize:method], and [apprize:vague], being the particular focus of this discussion, will now be described. The instantiations for these options are shown in Table 22.1. The option [apprize:vague] is realized by ellipsis “how/what about?” The meaning can be retrieved from the co-text, e.g., how about this shortage? This could be expanded from the co-text as What does this shortage of CDs mean in relation to parallel importing? Students’ questions i Seeking to repair their impoverished background knowledge of the topic As noted, the questions the students posed to each other choosing the [apprize:explain], [apprize:reason], and [apprize:method] sought a logical Table 22.1 Semantic realizations for the [apprize] option in this discussion B
Semantic option: [apprize]
Example
3
Precise
Which one are you doing?
3a
Explain
Why does the government apply parallel importing?
Reason
Why does the government restrict imports from overseas?
Method
How does the government protect the local industry?
Vague
Tiff what else?
4
Preparing students to write 303 Table 22.2 Student questions choosing the [apprize] option seeking to define and explain parallel importing Phase
turn
Questions
1 1 1 1 1
16 96 103 96 238
What’s parallel importing? How does it (= parallel importing) protect local industry? Why does the government use it (= parallel importing)? How does it (= parallel importing) protect local industry? What’s the effect (of parallel importing)?
series of inquiries to define and explain parallel importing: what?, how?, why? The particular questions of interest are shown in Table 22.2. The students were able to provide adequate responses for two questions: 1 What’s parallel importing? Response: The government (does) not allow produce (to be) bought from overseas. 2 Why does the government use it (= parallel importing)? Response: The main reason to have this is to protect their own producers. The students, however, experienced difficulties in explaining the consequences and rationale for parallel importing: how does parallel importing protect local industry? and what’s the effects of parallel importing? These difficulties resulted in their request for the lecturer to assist them. ii The students recast questions to the lecturer to more theoretical constructions in Phase 1 Curiously, the question the students then posed to the lecturer why is the world price lower than the equilibrium price? appears to bear no relationship to their discussion or questions to each other up to that point. They had focused on more congruent elements of parallel importing: imports— the government—CDs—the effects of the price rise—the price—the effects of parallel importing—the government, not on world or equilibrium prices. This apparent recasting of their question can possibly be explained by the lecturer’s previous explanations to this group. In two earlier instances the lecturer had explained that the beginning point for explaining parallel importing is equilibrium: • •
First of all you might start here at equilibrium; It assumes you start with your 30-dollar equilibrium here your steep demand curve and your flat supply curve add in a lower overseas price 20 dollars then show the effect of parallel importing pushing that price back up to equilibrium.
304 B. Wake The recast question, and others during the discussion, do not seem, however, to be a desire by the students to seek more theoretical explanations. Rather they suggest a need to indicate to the lecturer that her explanations had been understood, or at least to echo or mimic “how economists speak.” Not to construct meaning to the same metaphorical degree may have revealed the students’ poor control of the discourse and their inadequate understanding of the principles of economics, particularly in the early phase of their discussion. iii Questions revealing the students’ confusions in Phase 2 As the students began to interact more with the lecturer during their second discussion with the lecturer, their questions became less self- conscious, revealing their confusions. The question choosing the [apprize:vague] option how about this shortage is it supplied by the local producer? initiates one of the critical shifts in the overall discussion. This particular question began the process of focusing on more mundane ways to explain the consequences of parallel importing, both by the students and in the responses offered by the lecturer. iv Initiating critical shifts by the lecturer toward more congruent meanings in word choice and logical construction As the lecturer responded to the students’ apparent confusions in the second discussion, she abandoned explanations which had invoked equilibrium and demand theory in her first interaction with the students. The shifts in her explanations are evident linguistically in word choice and in the logical prediction–consequence construction of her explanations. Examples of the shifts in word choice from abstract theory to congruent participants are shown in Table 22.3. The shifts in word choice occur in the latter phase of the discussion. Rather than replicating the written discourse of economic theory, evident in Phase 1, perfectly competitive market structures, equilibrium, the world price, demand and supply, the lecturer construes real-world mundane scenarios involving actual numbers of CDs in Phase 2, the government, producers, and people as consumers. For comprehensive lexico-grammatical analyses of these data, see Wake (2006). v How prediction and consequence were reconstrued dialogically in the discussion: shifting from hypothetical syllogisms to narrative sequences A comparison of the prediction–consequence structures used by the lecturer in Phases 1 and 2 is shown in Table 22.4. The comparison illustrates the lecturer’s contingency strategies as she shifts from hypothetical syllo-
Preparing students to write 305 Table 22.3 Lecturer’s responses showing shifts in word choice to more congruent meanings Phase
Turn
Comparison in lecturer’s responses between Phases 1 and 2 showing shifts from abstract to more congruent word choice
1 1 1 1
139 142 153 255
1 2 2 2
255 274 274 305
2
321
We’ve got to assume perfectively competitive market structures First of all you start at equilibrium It assumes you start with your thirty dollar equilibrium The equilibrium price thirty dollars is set by local demand and supply The world price is set by demand and supply in other countries Local producers tend to supply that much At that price this is how much people want to demand The government says right we’re going to say you’re not allowed to import this many you’re only allowed to import ten Seventy five is the total amount that people will want
Table 22.4 Comparison of explanations by the lecturer between Phases 1 and 2 showing shifts from hypothetical syllogism to theory expressed as a narrative implication sequence Phase 1 and then what parallel importing does is if the world price is lower we’ve got this shortage it’s a shortage of product and that’s imports
Phase 2 we’re ten short so that starts to push the price up and as the price goes up brings more sellers into the and takes buyers out buyers can’t afford this much
gisms expressing demand and supply theory in Phase 1 to narratives expressing mundane scenarios in Phase 2. In the lecturer’s earlier response in Phase 1, the theoretical aspects of the model are explained. The lecturer provides an account of the inherent functions of parallel importing, albeit hypothetically, which closely reflects the goals of written academic economic discourse, i.e., economic predictions construed as generalized principles tested by hypothetical theoretical models. In contrast, the excerpt from Phase 2 shows a more narrative sequence which extends information using and and so. The various predictions and implications of parallel importing unfold as the lecturer describes the consequences of the import ban. vi The impact of the lecturer’s strategies on the students’ appropriation of predictive reasoning The data reveal interesting transitions in the students’ understanding. The exchange below from Phase 2 appears to have been a eureka moment for
306 B. Wake the students. It is here they realized a consequence of parallel importing: local suppliers will supply more—because the price is higher—the producers will produce more—and the buyer maybe can’t afford it. The excerpt also illustrates scaffolded guidance by the lecturer confirming their interpretation, albeit limited to that’s right, See Li Lecturer Li Cin Li Lecturer
on the other hand it’s just um we put this into it ah I mean parallel importing then ah the local supplier will might supply more yes yes because the price is higher that attract = = = = that’s right the producer to = = produce more = = produce more and the price is higher and the buyer maybe can’t afford it = = = = that’s right
As the students continued to discuss the consequences of parallel importing, a transitional appropriation of the discourse appears in their interactions. Indeed, similar strategies adopted by the lecturer appear in their interactions; that is, they also shifted between the theory and more congruent explanations. These shifts are apparent in student Li’s explanation, shown in Table 22.5.
Table 22.5 The lecturer’s construals of prediction–consequence in Phases 1 and 2 and student Li’s prediction-consequence sequence in Phase 2 Economics lecturer Phase 1 and then it (= parallel importing) says four effects
Li Phase 2 we’re ten short
Phase 2 the supplier will (intro) produce more
and those are the same as so that starts to push the a tariff price up
produce more
price goes up
and as the price goes up
and the the price is higher
quantity supplied rises
brings more sellers into the
and the demand is less
quantity demanded falls
and takes buyers out
because the buyer maybe can’t afford
and imports fall to the right amount
buyers can’t afford this much
and [the buyer is] not willing to buy at that higher price
Preparing students to write 307 The transitional shifts in Li’s explanation involve theoretical terms the demand is less as well as more mundane entities and processes, the buyer maybe can’t afford (the higher price). The logical structure is expressed by a prediction–consequence and prediction–reason narrative sequence similar to the lecturer’s more narrative explanations.
Discussion To acknowledge that economic discourse is arcane and complex is to acknowledge that students require learning opportunities to realize essential linguistic and rhetorical functions. A contribution of this case study has been to offer critical insights into the application of dialogic learning in a university classroom and the impact “doing economics” dialogically appeared to have on the students’ appropriation of meanings. In summary, the results from the examination of the spoken data indicate that: • • • •
the questions choosing the [apprize] option seek a logical series of inquiries from a definition of parallel importing to how the model operates and its effects; the students appear to recast their questions to the lecturer using more theoretical language; in the latter phase of the discussion a shift in focus occurs whereby the lecturer’s explanations construe real-world scenarios; transitional appropriation of the discourse occurs as the students begin to tentatively use both theoretical discourse and more congruent prediction–consequence/reason sequences in the later phase of their discussion.
The encouraging finding in relation to the students’ confusions is their desire to understand the economic model and their task. It can be argued that the influence of the lecturer’s contingency strategies on the students’ understanding went beyond mimicry. This is particularly evident in the kinds of interweaving between metaphor and more common-sense knowledge by the students as they attempted to overcome their confusions. These interweavings illustrate an important aspect of Vygotsky’s framework, that is, that the appropriation and construal of educational knowledge depends on an iterative process of mediated interactions. The shifts in the students’ negotiations shown here capture something of this iterative process. The overall deconstrual of meaning by the lecturer, however, was unexpected.
Conclusion In Australian universities, transitions in the learning process are rarely the focus of curricula, nor is the role of language. A degree program is often a kind of passage with the primary focus on the written product. This is
308 B. Wake particularly so given the formal role that assessment plays. Learning scenarios, on the other hand, which are tolerant of the difficulties that many students encounter, such as mimicry and copying, can generate a heightened appreciation by students of the complex meanings they need to acquire in order to produce written texts and so achieve success in their academic studies. The application of Vygotskian notions involving mediated learning in a university context, as in this case study, recognizes that guiding second- language international students in their understanding of complex concepts relies on personalized learning and teaching systems involving the co- constructions of meaning between students as mentees and the lecturer as mentor. The learning environment needs to anticipate and tolerate students’ difficulties as they attempt to understand relationships between arcane theory and their future professional needs.
References Bakhtin, M. M. (1986). Speech genres and other late essays. Austin, TX: University of Texas Press. Halliday, M. A. K. (2004). Introduction to functional grammar. 3rd ed. London: Arnold. Hasan, R. (1991). Questions as a mode of learning in everyday talk. In R. Le & M. McCausland (Eds.), Language education: Interaction and development (pp. 70–119). Launceston: University of Tasmania. Hewings, A. (1990). Aspects of the language of economics textbooks. In A Dudley- Evans & W. L. Henderson (Eds.), The language of economics: The analysis of economics discourse (pp. 29–42). ELT Documents 134. London: Modern English Publications and The British Council. Marshall, A. (1895). Principles of economics. London and New York: Macmillan. Mason, M. (1990). Dancing on air: Analysis of a passage from an economics textbook. In A. Dudley-Evans & W. L. Henderson (Eds.), The language of economics: The analysis of economics discourse (pp. 16–28). ELT Documents 134. London: Modern English Publications and The British Council. Smith, V. L. (1989). Theory, experiment and economics. Journal of Economic Perspectives, 3(1, Winter), 151–169. Vygotsky, L. (1986). Thought and language. Cambridge, MA: MIT Press. Wake, B. (2006). Dialogic learning in tutorial talk: A case study of semiotic mediation as a learning resource for second language international students. Unpublished PhD dissertation, University of Adelaide. Wignell, P. (1997). Making the abstract technical: On the evolution of the discourse of social science. Unpublished PhD dissertation, University of Sydney. Wignell, P. (1998). Technicality and abstraction in social science. In J. R. Martin & R. Veel (Eds.), Reading science: Critical and functional perspectives on discourses of science. London: Routledge.
23 Can archived TV interviews with social sciences scholars enhance the quality of students’ academic writing? Terry Inglese University of California, Santa Barbara
European archives in transition Archives hold valuable and meaningful resources, especially for educational purposes. They contain forgotten and hidden treasures defined under the label of cultural heritage, according to the Digicult study (2002), a research project financed by the European Commission to understand the impact of the digitization on Europe’s memory institutions. Museums, libraries, and archives are re-examining their roles as cultural heritage organizations within the “Information and Communication Society.” They hold treasures not discovered yet, especially by the education sector. They are not accessible by the public, mostly because their mission (at least in the European tradition) has been to preserve artifacts, and to provide “cautious” access. But with technological advances, memory institutions are obliged to change their practices: from storing objects, archives should become more dynamic. What changes over time are the value, relevance, and meaning of these archival artifacts, including how they might be newly interpreted. The old media technologies have allowed us to record and accumulate an incredible amount of media materials over the last 150 years. The next stage is to store, organize, and access these materials with a new approach to culture (Manovich, 2001). The following two case studies represent a joint collaboration between archivists, professors, and educational researchers. Our goal is to recognize the pedagogical and instructional benefits of studying with TV archives, especially TV interviews with social sciences scholars that are usually studied only through written texts. As we will see, these interviews can also be evaluated as learning tools with an impact on students’ academic writing. Our leading research question is: can multimedia and multimodal archives be used to enhance students’ writing quality when learning about social sciences audio-visible scholars?
Availability and affordability In using archives for educational purposes, there are two characteristics: their availability and their affordability (Inglese, Mayer, & Rigotti, 2007).
310 T. Inglese With new media technologies, archival materials produced in the past with analog technologies are very easily “transformed” into digital forms, increasing their availability as “multimedia content.” A joint collaboration between a university and an archive, such as ours, embodies an example worth sharing. The affordability concept supports the multimedia learning hypothesis (Mayer, 2001) that students learn more deeply when they interact with multimedia content, especially if they feel a personal relationship with the author to be studied, if this author is offered through a video interview format. We have chosen three main theories to give a theoretical framework to our case studies: the cognitive theory of multimedia learning developed by Mayer (2001, 2005), because learning is a cognitive activity, which gives us a solid but an incomplete picture. The multimodal learning theory (Jewitt, Kress, Ogborn, & Tsatsarelis, 2001; Jewitt & Kress, 2003) together with Brandt’s intersubjectivity concept in literacy (1990) provide more elements in order to understand how reading and writing are strongly influenced by the situated social environment. We will briefly describe the three theories, followed by the narrative and the results of the case studies.
The cognitive theory of multimedia learning and its social agency interpretation The cognitive theory of multimedia learning shows that students learn better with words and images than with words alone. This theory is consistent with several decades of empirical research on how people learn with multimedia. The focus is on the learner, who is able to use various coding systems to decode the information. Learners have separate information- processing channels for verbal and pictorial information. The dual coding theory of Paivio’s theory (1986), Baddeley’s working-memory theory (1992) and Sweller’s cognitive-load theory (2005) represent important theories of reference. Recently Mayer, Sobko, and Mautone (2003) developed the social agency theory: social cues in multimedia instructional messages lead to social responses in learners with the consequence that the content is studied in a deeper way and with better learning outcomes. The learner interprets the content as a social communication event, instead of an information delivery instance. In addition, there is the personalization effect. For Mayer (2004a, 2004b, 2005), instructional messages result in better transfer when the verbal material is presented in a conversational style, using the first and the second person (you and I), rather than using the third person (it). Research shows that students work harder to make sense of the instructional messages when they feel engaged in a social interaction, and the personalization effect involves converting the formal style into a conversational style. Personalized messages may prime the conversation schema in learners; they assume themselves to be in a human-to-human conversation with the instructional material and try hard to understand
Can TV interviews enhance students’ writing? 311 what “the other person” is saying, relating the material to their prior knowledge.
The theory of multimodal learning In conjunction with the cognitive theory of multimedia learning, we add the multimodal learning theory. According to Jewitt et al. (2001) and Jewitt and Kress (2003), learning is created through an interaction between visual, actional, and linguistic communication. It is multimodal, because it involves the transformation of information across different communicative systems defined as modes. Meanings are made and shaped in interpretation through many representational and communicative modes, not just through language (whether speech or writing). Because teachers in the classroom use different modes to express meanings, learning is a process of selection, adaptation, and transformation motivated by the learner’s interest and the context of learning. The authors criticize the dominant view of learning as a primarily linguistic accomplishment, because there are a variety of meaning-making resources which learners and teachers bring into the classroom. This conception of learning is seen as a dynamic process of sign making, inspired by the social semiotics of Halliday (1994), where communication is interpreted as a social phenomenon composed by different communication modes, like speech, gesture, image, and writing. Under these premises, reading and writing are semiotic processes that can be turned into a sense of what learning is for the learners, that is to say an affective and cognitive process. A multimodal perspective offers the advantages of (a) focusing on a rich range of resources, including talk, speech, image, action, gesture; (b) expanding the notion of meaning beyond speech and writing; and (c) focusing on the active role of teacher and her instructional content, as in our cases, the use of audio-visible authors.
A third perspective: intersubjectivity in literacy According to the constructivist learning theory, texts hold no absolute meaning in themselves. Individuals assemble their own mental representations based on the evaluation of the information, and they make connections with their past understanding. Words on a page have no properties of their own; they are what their readers make of them. The two theories of multimedia learning and multimodality are complementary, if we involve Brandt’s intersubjectivity concept in literacy (1990), which we combined with studies about the “author awareness” (Shananan, 1992) and the “visible author” concepts (Nolen, Crowley, & Wineburg, 1994; Nolen, 1995; Paxton, 1997, 1999, 2002). In reading comprehension, the author awareness concept (Shanahan, 1992) is an important factor in understanding the relationship between readers, texts, and authors, and it is fostered if reading is perceived as a conversation with
312 T. Inglese the text’s author. The same assumption counts with the visible author concept (Nolen et al., 1994). Nolen (1995) and Paxton (1997, 1999, 2002) investigated this phenomenon within two social sciences disciplines. Nolen studied the author’s visibility effects on statistical texts and Paxton on historical texts with adults and K–12 audiences. In both cases, the authors’ visible style of writing helped students in learning. The presence of a visible author, thanks to rhetorical forms defined by Crismore (1984) as metadiscourse, creates a variance in readers’ responses to the textbook. By using metadiscourse, textbooks reinforce the possible relationship between author and reader. Metadiscourse consists of words, phases, clauses, different points of views, such as the use of the first person (I think that) and the use of the second person (Remember, you will read that). A visible author offers a potential benefit to students, because who is presenting the information does matter. Additionally, a visible author can act as a sort of substitute teacher, leading students through the primary information and adding clues about the sources of information. Students reading a textbook written by a visible author are expected to make frequent references to the author, holding mental conversations between themselves and the person they imagine the author to be. This kind of un-abstracted relationship may lead students to a closer association with the material in the text. The consequences for learning the visible author’s material and for literacy are interesting. Brandt’s (1990) conception of intersubjectivity offers an additional point of view, emphasizing the distinction between written and oral language. Speech is natural, universal, and spontaneous and has a quality of being immediate and social. Literacy, on the other hand, is transmitted through formal learning. Social involvement appears as a fundamental basis of orientation for reading and writing, “making literacy not the narrow ability to deal with texts, but the broad ability to deal with other people as a writer or a reader” (Brandt, 1990, p. 14). The oral–literate dichotomy is a relationship of context and text. In speech, the direction of meaning is from context to text; in writing and reading, the direction of meaning is from text to context. The context is the situation, “the possible world,” where time and space define the joint presence of participants. In oral exchanges, context is embodied by the shared backgrounds of the participants. Context is everything that participants in a conversation know and understand, that contributes to how they make sense of what is said. On the other hand, in literate exchanges context is represented only by the text. In both exchanges, literate and oral, intersubjectivity is the mutual recognition of the presence of the other, and is the core of interpretation and meaning. One’s own involvement with other people, rather than with written texts, is the center of literate interpretation and development. For Brandt, literacy is a growing metacommunicative ability, an awareness of and control over the social means by which people sustain discourse, knowledge, and reality. This becomes a condition where social involvement is the key model of literacy and literacy growing, because
Can TV interviews enhance students’ writing? 313 making sense of print requires the fundamental realization that written language is about an involvement of writer and reader; its reference and meaning depend on the inter-subjective bonds established in the acts of writing and reading. Developmentally, that makes knowledge about the acts of reading and writing . . . the key knowledge for literacy development. Learning to read and write depends critically on immediate social involvements with people who read and write and who can show you how the work goes. More broadly, this perspective suggests that we look toward our ordinary social ties . . . as the very means that enables reading and writing. (Brandt, 1990, p. 32) To understand writing and reading as intersubjective processes, and texts as facilitators in those processes, is to characterize literacy as a metacommunicative knowledge, and how people use language to get to meaning. Paxton, Nolen, Shanahan, and Brandt carried out their studies using the social presence of the author through a text-based support. They did not use multimedia learning content. With the use of multimedia authors, it is possible to develop new ways of reading, writing, and learning, thanks to the intermingling of oral and written supports, which recalls the “second orality” of the electronic and digital media, studied by Ong (1998, 2002) and Olson (1989). Summarizing, we chose these three theories to understand how students read and studied texts of audio-visible vs. invisible authors. We argue that the audio-visible authors’ texts are the means through which students can create and maintain an intersubjective bond with these audio-visible authors, because of the multimedia, multimodal, and involvement quality of their texts. The purpose of the research was: (a) to use TV interviews with important social sciences scholars who are usually only studied through written texts as instructional materials, and (b) to make a comparison between them and an invisible author represented only through a written text. As main data we used the written compositions that students produced as the final exam after taking the course. Our working hypothesis is that audio-visible authors would promote a richer learning context than the invisible ones.
The two case studies The RTSI archive Because of the lack of a higher educational institution, such as a university, in the region of Ticino and Grigion (see Toppi & Sassi, 1994), the Swiss Public Radio and Television of the Italian-speaking area of Switzerland (RTSI) aimed to produce from the 1960s to the 1990s educational programs to offer the Italian-speaking population of Switzerland different ways of enhancing education and culture. Throughout the past eight
314 T. Inglese decades, the RTSI has produced high-quality educational programs including video-based interviews with important scholars, today stored in their archives (Inglese, 1998; Inglese et al., 2007). Together with the young University of the Svizzera Italiana, founded in 1996, we negotiated with the RTSI the educational usage of its TV interviews collection. The design of the case studies The scholars portrayed in the TV interviews for the first case study were the two Austrian science philosophers, Karl Popper (1902–1994) and Paul Feyerabend (1924–1994). For the second case study we used Feyerabend again and the French anthropologist Claude Lévi-Strauss (1908–). These one-hour-long interviews were aired on TV and archived by the RTSI for more than 30 years without any additional TV screening. Today, these interviews represent, as we will demonstrate, important multimedia and multimodal instructional content that could be used for different learning activities, opening up new literacy research questions. Definitions of audio-visible and invisible authors We define the audio-visible author presented through two layers of visibility: (1) the audio-visible author is visible through an audio-video format that represents him in a dialogical I–you relationship; (2) the audio-visible author is visible also through the transcript of the video and some chosen theoretical chapters that still present the dialogical I–you relationship. The invisible author is not visible through a video format, and not visible through a visible text, because he writes in a very detached and impersonal third-person form, and his writings lack the metadiscourse quality. Procedure We worked with a professor of a political science course at the faculty of Communication Sciences. We chose excerpts from TV interviews, made textual transcripts and planned instructional activities. One educational objective of this course is to develop in freshmen a critical understanding of abstract concepts such as culture, society, institutions, and media, where “communication” is not neutral, but is a political and social act. The two case studies took place in the naturalist setting of the classroom during the academic years 2003–2004 and 2004–2005. They were quasi- experimental cases. One of the most significant limitations was the absence of a control group; all students were exposed to the same multimedia messages. The frontal teaching, taught in Italian, was enriched by classroom discussions, coordinated by the professor. The classes were composed by native Italian speaker students and non-native ones.
Can TV interviews enhance students’ writing? 315
The first case study The first case study presented the following concepts: democracy, mono culturalism and multiculturalism, liberalism, universalism, cultural relativism, “exportable democracy” into non-democratic countries, collective and individual human rights, and the meaning of citizenship. These concepts were explained through the professor’s lessons using texts from different media, introducing contemporary thinkers who discussed the mentioned concepts and gave arguments for and against monoculturalism and multiculturalism. Some authors were presented through videos (Popper and Feyerabend) and some others only through texts (Semprini). Popper discussed concepts like democracy and reforms as one of the most important political systems and Feyerabend discussed the roles of experts in a democratic system and their “collective” responsibilities. The Italian multiculturalism scholar Semprini focused on the role of a multicultural society compared to a monocultural one. His text was required reading and contained the differences between monoculturalism and multiculturalism and four abstract models he developed for defining multiculturalism. No hypothesis was developed before conducting the first case study, which examined how students reacted to these two types of authors. Procedure Each video (two videos with Popper and one with Feyerabend, for a total of 11 minutes) was shown twice in class and was preceded by the professor’s explanation. The first screening provided students with an introduction of the author. We gave the video transcript, followed by a second screening of the same video. The students could follow the video, reading the text and/or listening to the audio-visible authors. A discussion followed after the second screening. All the instructional materials were stored on a website. The population The participants were 112 freshmen (65 females and 47 males). The class was composed of 78 native Italian speakers and 34 non-native ones. The average age was between 20 and 25 years. The final written exams The exam included seven questions. Important for the analysis were only two related to the audio-visible authors and one to the invisible author. The other questions were devoted to other materials that we did not analyze. Some 105 students took the exam (73 native Italian speakers and 32 non-native ones). The non-native Italian students could write the exam in
316 T. Inglese their own languages: German, French, Spanish, or English. The professor used a 10-to-1 grading scale, with 10 as the highest and 1 as the lowest. Each question was given a separate grade for comparing the performances for each type of author. In order to prevent students from being rewarded for mere verbosity, they were told that they would be graded for the content and not for the length of their answers. The professor was unaware of the type of analysis we intended to conduct. For our analysis, we manually transcribed into a Word document all the exams’ texts, focusing on three dependent variables: (1) the grade for each author’s exam question, (2) the number of written words, and (3) the number of times students quoted or mentioned the two types of authors in their compositions. The independent variables were: the language of the students (Italian native speakers and non-native ones) and the author’s visibility or invisibility. We are aware that the dependent variables might be influenced by other variables, which we could not detect extensively. Word counting and quotation counting Scardamalia and Bereiter (1991) found that the number of words in students’ written compositions correlates substantially with indicators of their writings’ quality. The assumption for the word-counting analysis is that if students feel that the author is socially closer, they might establish an interpersonal relationship with him/her and write more words. We also decided to count how many times students quoted the two types of authors, a research method we adopted from Paxton and Nolen. Descriptive statistics The grade The average grade of the 73 Italian-speaking students for the audio-visible authors’ questions was 7.29 (SD = 1.67) and the average grade for the invisible author one was 6.80 (SD = 2.32). The average grade of the 32 non-native Italian students for the audio-visible authors’ questions was 7.33 (SD = 1.56) and the average grade for the invisible author one was 6.67 (SD = 2.13). Both groups performed better with the audio-visible authors. The number of written words The average number of written words of the 73 Italian-speaking students for the audio-visible authors’ questions was 207 (SD = 114) and the average number of written words for the invisible author was 133 (SD = 72). The average number of written words of the 32 non-Italian-speaking students for the audio-visible authors’ questions was 170 (SD = 81) and the average
Can TV interviews enhance students’ writing? 317 7.4
7.33
7.29
Visible Non-visible
Mean of course grade
7.2 7.0 6.8
6.8
6.67 6.6 6.4 6.2 73 Italians
32 non-Italians
Figure 23.1 Effect of language and author visibility on course grade.
Mean of number of words
250 200
Visible Non-visible
207 170
150
133
143
100 50 0 73 Italians
32 non-Italians
Figure 23.2 Effect of language and author visibility on number of words.
number of written words for the invisible author was 143 (SD = 68). Both groups wrote more words with the audio-visible authors. The number of quotes The average number of quotes for the 73 Italian-speaking students for the audio-visible authors’ questions was 2.02 (SD = 1.54) and the average number of quotes for the invisible author was 0.26 (SD = 0.64). The average number of quotes for the 32 non-Italian-speaking students for the audio-visible authors’ questions was 1.75 (SD = 1.35) and the average
318 T. Inglese
Mean of number of quotes
2.5 2.0
Visible Non-visible
2 1.75
1.5 1.0 0.5
0.2
0.3
0 73 Italians
32 non-Italians
Figure 23.3 Effect of language and author visibility on number of quotes.
number of quotes for the invisible author was 0.31 (SD = 0.59). Both groups quoted the audio-visible authors more times. ANOVA tests We collapsed the data of Popper and Feyerabend into the visibility vari able. We conducted a multivariate ANOVA testing the effects of language (Italian vs. non-Italian speakers) and author visibility on the three dependent variables. We found a main effect of visibility on grade, F(1,314) = 5.59, p < 0.05; of visibility on numbers of words F(1,314) = 16.35, p < 0.001; and there was an effect of visibility on quotes F(1,314) = 93.61, p < 0.001. We found no main effects of language on grade, on numbers of words, and on numbers of quotes. There was a marginally significant interaction between visibility and language on number of words F(1,314) = 3.49, p = 0.63). We conducted a second case study a year later to see if there were similar patterns in the data.
The second case study The second case study was structured as the previous one. This time, instead of Popper, we used videos from a Lévi-Strauss interview. We chose two excerpts where he explains racism and the prejudiced differences between superior and inferior societies. Each video (two for Lévi-Strauss and one for Feyerabend for a total of ten minutes) was shown twice in class and it was preceded by the professor’s explanations.
Can TV interviews enhance students’ writing? 319 The Moodle learning platform All of the course content was uploaded on the Moodle learning platform. Additional course materials were: the course syllabus, the transcriptions of the audio-visible authors’ videos, the weekly slides, the professor’s lectures and copies of her article (Rigotti, 2003), the multiculturalism text written by Semprini (2002), and two chapters extracted from the two audio-visible authors’ books (Lévi-Strauss, 2002; Feyerabend, 1981). The Moodle platform, used as a storage tool, provided a log-file method for collecting students’ behavioral data and their movements in the platform, including how texts and videos were used and how many times the two groups accessed these materials, which we will not report here. Measuring the social presence For this case study we measured the social presence of all the authors’ texts by counting how many times they used the first and the second person singular and plural, rhetorical questions and examples, and the total amount of words in their texts. We adopted the counting methodology from Paxton and Nolen, with the hypothesis that this measure might have an impact on how students perceived the authors’ texts. The Lévi-Strauss text was characterized by a 1.90 percent of social presence, the Feyerabend one by 3.90 percent and the Semprini text by 0.20 percent of social presence. We are aware that the audio-visible authors were perceived as more “present” than Semprini and that the videos’ transcripts were shorter than the invisible author’s text. Therefore, the professor selected additional reading texts from Lévi-Strauss (2002) and Feyerabend (1981) books. At the end, the amount of text that students needed to study for the three authors was approximately the same. The population The class was composed of 108 freshmen (84 females, 24 males): 60 native Italian speakers and 48 non-native ones. The average age was from 20 to 25 years. Table 23.1 Social presence of authors’ texts
I you we questions examples Total
Lévi-Strauss (700 words) (%)
Feyerabend (558 words) (%)
Semprini (2,141 words) (%)
1.10 1.30 2.10 0.00 0.00 4.50
0.00 0.20 0.50 0.40 0.70 1.80
0.00 0.00 0.20 ~0.00 ~0.00 0.20
320 T. Inglese The final written exam results The final exam consisted of six questions: two dedicated to the audio- visible authors and one to the invisible author. Non-native Italian-speaking students were given the option of answering in French, English, Spanish, or German. From a class of 108 students, 86 (46 native Italian speakers and 40 non-native ones) took the exam, 23 did not. Descriptive statistics The grade The average grade for the 46 Italian-speaking students for the audio-visible authors was 7.98 (SD = 1.71) and the average grade for the invisible author was 7.17 (SD = 3.05). The average grade for the 40 non-native speakers for the audio-visible authors was 7.40 (SD = 1.74) and the average grade for the invisible author was 5.80 (SD = 3.57). Both groups performed better with the audio-visible authors. The number of written words The average number of written words for the 46 Italian-speaking students for the audio-visible authors was 161 (SD = 61.53) and the average number of written words for the invisible author was 160 (SD = 86.13). The average number of written words for the 40 non-native Italian students for the audio-visible authors was 156 (SD = 66.20) and the average number of written words for the invisible author was 111 (SD = 90.18). Both groups wrote more words with the audio-visible authors.
9
Mean of course grade
8 7
7.98 7.17
Visible Non-visible
7.41 5.8
6 5 4 3 2 1 0 46 Italians
40 non-Italians
Figure 23.4 Effect of language and author visibility on course grade.
Can TV interviews enhance students’ writing? 321 180 160
162
161
Visible Non-visible
157
Mean of words
140 120
112
100 80 60 40 20 0 46 Italians
40 non-Italians
Figure 23.5 Effect of language and author visibility on number of words.
The number of quotes The average number of quotes for the 46 Italian-speaking students for the audio-visible authors was 2.9 (SD = 1.31) and the average number of quotes for the invisible author was 0.60 (SD = 0.64). The average number of quotes for the 40 non-Italian-speaking students for the audio visible authors was 2.87 (SD = 1.81) and the average number of quotes for the invisible author was 0.32 (SD = 0.52). Both groups quoted more times the audio-visible authors. ANOVA tests We collapsed the data of the audio-visible authors into one visible variable, conducting a multivariate ANOVA to test the effects of language and the 3.5
Number of quotes
3.0
2.9
Visible Non-visible
2.8
2.5 2.0 1.5 1.0 0.6 0.5
0.3
0 46 Italians
40 non-Italians
Figure 23.6 Effect of language and author visibility on number of quotes.
322 T. Inglese author visibility on the three dependent variables. We found a main effect of visibility on grade, F(1,257) = 14.84, p < 0.001, and of visibility on number of words F(1,257) = 5.71, p < 0.05. There was a main effect of vis ibility on quotes F(1,257) = 191.87, p < 0.001. We found a main effect of language on grade, F(1,257) = 9.68, p < 0.01, and a main effect of language on number of words F(1,257) = 7.83, p < 0.01. There was no main effect of language on quotes. There was an interaction between visibility and language on number of words F(1,257) = 5.24, p < 0.05.
Discussion In both cases there is statistical evidence that (a) students learned better with written words, audio, and videos rather than with written words alone; (b) the visibility of the author has an effect on the three variables, for both groups of students; and (c) learning with more modes is better than with only one. From the intersubjective theory of reading and writing, in students’ written compositions the social presence of the audio-visible authors counts, and is expressed through a tendency toward writing more words and relating more times to the audio-visible authors. The multimedia and multimodal characters of computer-mediated learning bring new research questions such as the meaning-making learning impact of a range of representational modes (which includes image, movement, gesture, music, sound effect, voice quality). The key issue is to understand what contributions these resources promote in teaching and learning, and what resources are best suited for what tasks and for which kind of learner. The multimodality of the screen raises the question of what literacy is or can be, what does it mean to be literate in a digital era, and how the multimodal character of computer-mediated learning reshapes the traditional and print-based concept of literacy (Kress, 2003). The combination of multimedia and multimodality theories offers a new perspective for understanding how learning happens and how it can be assessed. In our cases, both groups of students engaged more with the audio-visible authors. What was to be learnt and read was reshaped by the multimedia and the multimodality formats. The resources of the screen enabled students to have different forms of engagement with these authors that were perceived as “persons,” with authorial intent and as sociocultural entities. Choosing the Nolen, Paxton, Shanahan, and Brandt framework and their concepts of visible authors and intersubjectivity meant interpreting reading and writing as involvement acts, where readers tried to reach across texts to other human beings. As Brandt would highlight, the motive for reading is to find other minds. It is the consideration of the author (voice, style, intention) that makes reading an activity of social interaction. Literacy requires the awareness of how language works to sustain intersubjectivity, enabling the mutual work between text and context. The social
Can TV interviews enhance students’ writing? 323 involvement represents the basis of interpretation in reading and writing, understanding of how human beings create reality together. Learning to read, and in our case it is an academic way of reading, requires learning to maintain and intensify reliance on social context even under new and precarious circumstances. Brandt’s important reflection, which we consider a key interpretation for our work, is that literacy failures are not failures of separation, but rather failures of involvement, because of the lack of access to a feasible context for making sense of print. We think that multimedia and multimodal audio-visible authors, designed as learning materials to promote involvement and motivation, can enhance the quality of students’ written academic compositions. More studies need to be done in this new area. We collected 70 interviews with students to understand how they read the texts of the two types of authors. A separate publication is foreseen.
Acknowledgments The author thanks Prof. Francesca Rigotti and the two classes, Prof. Richard Mayer of UCSB Psychology Department, Prof. Chuck Bazerman, Prof. Karen Lunsford, Paul Rogers, Suzie Null of the UCSB Education Department, and her friends Cheryl Cohen and Sylvia Curtis.
Bibliography Baddeley, A. (1992). Working memory. Science, 255, 556–559. Brandt, D. (1990). Literacy as involvement. The acts of writers, readers, and texts. Carbondale, IL: Southern Illinois University Press. Crismore, A. (1984). The rhetoric of textbooks: Meta-discourse. Journal of Curriculum Studies, 16(3), 279–296. Digicult report. Technological landscapes for tomorrow’s cultural economy. (2002). European Commission Directorate. Feyerabend, P. (1981). La scienza in una società libera. Milan: Feltrinelli. Halliday, (1994). An introduction to functional grammar. London and Baltimore, MD: Edward Arnold. Inglese, T. (1998). La produzione documentaristica della TSI di Lugano dal 1968 al 1984 attraverso l’analisi di due rubriche: 360 e Reporter. Tesi di laurea in Cinematografia Documentaria (DAMS). University of Bologna (Italy). Inglese, T. (2006). Pedagogic roles of audio-visual TV interviews for educational and instructional purposes in a multimedia and multimodal learning environment for social science disciplines. Results from three case studies. Unpublished PhD in Communication and Education, University of Lugano, Switerland. Inglese, T., Mayer, R. E., & Rigotti, F. (2007). Pedagogic role of audiovisual TV archives in college classes: Can visible authors reduce the learning gap between native and non-native language speakers? Learning and Instruction, 17, 67–77. Jewitt, C., Kress, G., Ogborn, J., & Tsatsarelis, C. (2001). Exploring learning through visual, actional, and linguistic communication: the multimodal environment of a science classroom. Educational Review, 53(1), 5–18. Jewitt, C., & Kress, G. (Eds.). (2003). Multimodal literacy. New York: Peter Lang.
324 T. Inglese Kress, G. (2003). Literacy in the new media age. London and New York: Routledge. Lévi-Strauss, C. (2002). Razza e storia, Razza e cultura. Turin: Einaudi. Manovich, L. (2001). The language of new media. Boston, MA: MIT Press. Mayer, R. E. (2001). Multimedia learning. New York: Cambridge University Press. Mayer, R. E. (Ed.). (2005). Cambridge handbook of multimedia learning. New York: Cambridge University Press. Mayer, R. E., Fennell, S., Farmer, L., & Campbell, J. (2004a). A personalization effect in multimedia learning: students learn better when words are in conversation style rather than formal style. Journal of Educational psychology, 96(2), 389–395. Mayer, R. E., & Moreno, R. (2004b). Personalized messages that promote science learning in virtual environments. Journal of educational psychology, 96(1), 165– 173. Mayer, R. E., Sobko, K., & Mautone, P. (2003). Social cues in multimedia learning: role of speaker’s voice. Journal of Educational Psychology, 95, 419–425. Nolen, S. (1995). Effects of a visible author in statistical texts. Journal of Educational Psychology, 87(1), 47–65. Nolen, S., Johnson Crowley, N., & Wineburg, S. (1994). Who is this “I” person, anyway? The presence of a visible author in statistical text. In R. Garner & P. Alexander (Eds.), Beliefs about text and about instruction with text (pp. 41–56). Hillsdale, NJ: Erlbaum. Olson, D. R. (1980). On the language and authority of textbooks. Journal of Communication, 30(1), 186–196. Ong, W. (1998). Digitization, ancient and modern: Beginnings of writing and today’s computers. Communication Research Trends, 18(2), 4–21. Ong, W. (2002). Orality and literacy. The technologizing of the word. London and New York: Routledge. Paivio, A. (1986). Mental representations. A dual coding approach. Oxford: Oxford University Press. Paxton, R. (1997). “Someone with like a life wrote it”: The effects of a visible author on high school history students. Journal of Educational Psychology, 89(2), 235–250. Paxton, R. (1999). A deafening silence: History textbooks and the students who read them. Review of Educational Research, 69(3), 315–339. Paxton, R. (2002). The influence of author visibility on high school students solving a historical problem. Cognition and Instruction, 20(2), 197–248. Rigotti, F. (2003). Le parole della democrazia. Equilibri. Rivista per lo sviluppo sostenibile, annoVII(2). Scardamalia, M., & Bereiter, C. (1991). Literate expertise. In K. Ericsson & J. Smith (Eds.), Toward a general theory of expertise. Prospects and limits (pp. 172–194). New York: Cambridge University Press. Semprini, A. (2000). Multiculturalismo. La sfida della diversità nelle società contemporanee. Milan: Franco Angeli. Shanahan, T. (1992). Reading comprehension as a conversation with an author. In M. Pressley, K. Harris, & J. Guthrie (Eds.), Promoting academic competence and literacy in schools (pp. 129–148). San Diego, CA: Academic Press. Sweller, J. (2005). Implications of cognitive load theory for multimedia learning. In R. E. Mayer (Ed.), Cambridge handbook of multimedia learning (pp. 19–30). New York: Cambridge University Press.
24 Social academic writing Exploring academic literacies in text-based computer conferencing Warren M. Liew and Arnetha F. Ball Stanford University
Introduction The recent incursion of academic writing practices into the virtual domains of computer-mediated communication (CMC) has excited considerable interest among scholars of literacy, communication, and writing instruction. In particular, the increasing use of text-based asynchronous virtual learning networks (ALNs)—better known as computer conferences or electronic discussion forums—in higher education classrooms has inspired numerous studies of their potential affordances for effective collaborative learning (see review by Luppicini, 2007). Purportedly more promising are the advantages of “blended learning” environments that combine face-toface with computer-mediated interactions to extend participatory learning beyond the spatio-temporal constraints of either traditional classroom instruction or fully virtual distance learning (see, e.g., Bonk & Graham, 2006; Garrison & Vaughan, 2007). Drawing on computer conference data from a ten-week graduate-level blended learning seminar that the authors co-taught in the winter of 2007, this study examines the extent to which the socially productive effects of computer conferencing are reflected in the discursive properties of online academic writing (cf. Aviv, 2000; Coffin & Hewings, 2005; Heckman & Annabi, 2005; Potter, 2008). Employing discourse-analytic techniques, this exploratory study describes and analyzes students’ online discussions to reveal the rhetorical operations of social academic writing—a “double- voiced” genre of communal discourse incorporating both cognitive and socio-affective functions. The contributions and implications of these findings for further academic literacies research are discussed in the conclusion.
Background During the academic quarter of winter 2007, a professor and her teaching assistant (the second and first author respectively) led a ten-week graduate seminar that explored the uses of discourse analysis in educational research.1 Course readings, seminar discussions, and written assignments
326 W. M. Liew and A. F. Ball critically examined the linkages among language, power, and knowledge construction in educational settings. In addition to weekly three-hour classroom seminars, students were required to participate throughout the quarter in week-long asynchronous computer conferencing via CourseWorkTM, the university’s main e-learning management suite. These online conferences were intended to deepen and extend the week’s seminar discussions, allowing students to cross-reference, compare, and connect the course readings with their own graduate research projects. Each week a student-moderator was designated to open and facilitate the online discussion. Word-processing functions on the discussion board allowed authors to review, retract, and revise their electronic posts at any time, while weekly discussion threads were continuously archived in students’ individual CourseWorkTM accounts. Students’ online discussions constituted a substantial 30 percent of the final grade. We decided against providing formal assessment rubrics, offering instead simple written guidelines regarding the expected content, length, and frequency of weekly online posts. Computer conferencing began in the second week of the course, with the first author “modeling” the role of discussion moderator. In a principled move to allow students full ownership of the online forum, the course professor elected not to participate in weekly online exchanges, acting merely as a “silent observer.” The idealized goal was to create a participatory, learner- centered community of practice, in which students would learn how to coordinate with, support, and lead others, to become responsible and organized in their management of their own learning, and to be able to build on their inherent interests to learn in new areas and to sustain motivation to learn. (Rogoff, 1994, p. 225)
Data collection and analysis This study followed an ethnographic style of research that drew on the following data sources: (1) the authors’ experiences as participant-observers of the ten-week seminar that they co-taught; (2) a 72,000-word corpus of students’ archived posts in the CourseWorkTM discussion forum; (3) weekly lesson plans and post-lesson conferences written by the professor and her teaching assistant; (3) anonymous student feedback officially collated by the university’s online course evaluation system. The first author conducted both online and face-to-face semi-structured interviews with students nine months after the course had ended (following Institutional Review Board approval), by engaging them in retrospective analyses of their online conference posts.2 Discursive categories emerging from comparisons of the researchers’ independent coding of online transcripts, interview data, instructors’ memos, and course evaluations were used to develop the discourse analytic framework in Table 24.1. For the purposes of this analysis,
Retrieval
Sociality
• Identify common ideas and themes across readings and discussions. • Critically evaluate arguments. • Critique the validity and generalizability of ideas/concepts/theories using knowledge synthesized from other readings and/or personal experience. • Reformulate or offer solutions to problems and questions.
Synthesis and evaluation Intersubjective modality
• Recognition: Reply directly to others’ posts, rather than start a new thread; directly reference (e.g., by quoting) points raised by others. • Affirmation: Express agreement with and/or appreciation for others’ views and arguments. • Polite humor: Express affiliation through friendly sarcasm, teasing, witty puns, in-group jokes, etc.
• Personalized evidentials: Use I-statements (e.g., I think, I feel, I believe) to avoid generalizing or imposing one’s views on others. • Hedging: Qualify opinion or argument with uncertainty, either to express genuine doubt or to enact modesty. • Emotional expression: Articulate emotionally charged responses to ideas, views, and arguments. • Self-disclosure: Offer details of personal experiences; express vulnerability such as honest confusion or incomprehension.
• Phatics: Use ritual social expressions such as salutations and valedictions. • Vocatives: Address participants by first name. • Inclusives: Refer to others using inclusive pronouns (e.g., we, us, our). • Informality: Use of sentence fragments, unusual spelling, slang, in-group idioms, emoticons, conspicuous capitalization.
Socio-affective processes
Comprehension, • Demonstrate understanding of an idea/ argument/concept through explanation and application, and analysis elaboration. • Use knowledge from readings to address Subjective problems and questions raised by others. modality • Extend validity of argument/concept/claim to “real-world” applications and personal experiences. • Identify supports, warrants, and possible rebuttals to claims and arguments.
• Recall information and ideas from course readings, seminar discussions, and online posts. • Initiate a discussion based on points previously raised.
Cognitive processes
Table 24.1 An analytic rubric for social academic writing in a computer conference
328 W. M. Liew and A. F. Ball a single post from Week 5’s discussion was selected to illustrate the descriptive validity of this analytic rubric (see Figure 24.1).
Analytic framework The elision of nonverbal social cues in electronic exchanges has led researchers to characterize CMC as inherently lacking in media richness, in terms of its “capacity for immediate feedback, the number of cues and senses involved, personalization, and language variety” (Rice, 1993, pp. 452–453). While text-based CMC has been shown to limit impression formation (e.g., Daft & Lengel, 1986; Kiesler, Siegel, & McGuire, 1984), more recent studies suggest that users adopt compensatory communicative strategies for conveying social information (see review by Walther, 1996). Indeed, virtual communication can facilitate self-disclosure, trust-building, friendship, and community, as experienced users creatively adapt their relational behaviors and rhetorical strategies to CMC’s affordances and constraints (Fleckenstein, 2005; Ridings, Gefen, & Arinze, 2002; Tidwell & Walther, 2002; Woods & Ebersole, 2003). In this study, we examined how computer-conference participants negotiated the “cues filtered-out” limitations of CMC interactions by exploiting the social resources of online written communication. Central to the analysis were the socio-affective processes of trust-building, impression management, and group cohesion that in turn support and sustain the intellectual practices of virtual academic communities. In describing and analyzing the intellectual and socio-affective dimensions of students’ digital texts, we adapt and expand on some of the previous discourse-analytic tools developed by researchers to investigate the discursive processes of ALNs in higher education settings (e.g., Anagnostopolous, Basmadjian, & McCrory, 2005; Heckman & Annabi, 2005; Henri, 1991; Rourke, Anderson, Garrison, & Archer, 1999; Potter, 2008). The following section outlines the categories of our analytic framework in Table 24.1, which are then used as a coding scheme to analyze a selected post from the online discussions. Cognitive processes Much of the cognitive work of academic writing pertains to the burden of argumentation: “a verbal, social and rational activity aimed at convincing a reasonable critic of the acceptability of a standpoint by advancing a constellation of propositions justifying or refuting the proposition expressed in the standpoint” (van Eemeren, 2001, p. 11). Effective argument structures move through a recursive pattern of logical exposition, in which claims are made in tandem with supporting evidence and warrants (Toulmin, 1958). The ability to compose effective formal arguments valued by the academy represents what Scollon and Scollon (1981) have referred to as “essayist literacy.” Such scientific-academic writing typically includes the following discursive features: textual cohesion, informational density, Greco-Latin
Social academic writing 329 vocabulary, and frequency of nominalization, relational and mental process clauses, and scholarly citations (Corson, 1985; Halliday & Matthieseen, 2004). In terms of cognitive processes, essayist discourse may be aligned with the exercise of critical thinking skills modeled in our analytic rubric on Benjamin Bloom’s (1956) famous taxonomy: retrieval, comprehension, analysis, application, synthesis, and evaluation. Socio-affective processes Wegerif (1998) has noted that “many evaluations of asynchronous learning networks understandably focus upon the educational dimension, either learning outcomes or the educational quality of interactions, overlooking the social dimension which underlies this” (p. 34). Indeed, analysis of our data surfaced a range of rhetorical and lexico-grammatical indicators attesting to the salience of trust-building processes in online communities of practice. Within the scope of these sociable behaviors we distinguished between the categories of sociality and modality (see Table 24.1). Under sociality, four recurrent discourse features were noted. While phatic expressions (e.g., “You’re welcome”; “Have a good weekend, everyone!”) were widely used as ritual markers of friendly social exchanges, more strategic was the use of vocatives (e.g., pronouns and first names when addressing and acknowledging others) and inclusives (e.g., collective pronouns we/our/us) to signal shared knowledge, beliefs, and experiences. A variety of informalities (e.g., interjections, ungrammatical sentences, intentional misspellings, emoticons) served also to create a conversational tone of in-group familiarity and intimacy. While sociality denotes sociable linguistic behaviors, the category of modality signals the writer’s affective and cognitive relationship with the texts he/she writes and reads. Not only do writers position themselves epistemically by expressing their degree of commitment to, or confidence in, the statements they make, but in doing so they create solidarity or distance between themselves and others. Two common types of modality were identified. First, subjective modality (Fairclough, 1992) indexes the writer’s epistemic and affective stance toward his/her own knowledge claims. For example, I-statements—e.g., “I think”/“I feel”/“It seems to me”—not only qualify statements with varying degrees of conviction, certainty, and emotional investment, but imply personal responsibility over assertions that might otherwise be seen as having limited generalizability. The use of similar and other hedging strategies further serve a “negative politeness” (Brown & Levinson, 1987) function, allowing the writer to advance his/ her views without imposing them on others. Equally common were disclosures of personal information and experiences that served as illustrative warrants or examples for arguments. These were sometimes of an emotional and sensitive nature, in which writers appeared to trade personal vulnerability for intimacy with, and sympathy from, their readers.
330 W. M. Liew and A. F. Ball The second category, intersubjective modality, is “signaled by explicit references to another person’s statement” in order “to create solidarity and familiarity with that person” (Anagnostopolous et al., p. 1711). For example, direct responses to specific posts reflect one’s studied engagement in others’ contributions, whether by way of affirming or challenging them. Also incorporated into our rhetorical descriptors for intersubjective modality were communal strategies involving the use of shared humor as a “positive politeness” strategy (Brown & Levinson, 1987). Table 24.2 applies the analytic schema developed in Table 24.1 to an annotation of an online response posted by one of the students in Week 5 of the course (see Figure 24.1).
Discussion Jane’s3 online post skillfully goes about the intellectual task of identifying, analyzing, and synthesizing related ideas raised in the course readings, online posts, and class discussions (which included the contributions of Kris, a visiting professor who had been invited to participate in the week’s class seminar), while evaluating them in the light of her own knowledge and experiences. What is noteworthy, however, is the calculated skill with which these academic motions were coordinated with a range of socio- affective moves. Indeed, Jane’s writing exemplifies how students in their conference posts seek to enact a twin position of authority and humility while critiquing the views of others. Commenting on her post in an online interview, Jane explained: toward the end, i wasn’t only affirming others’ views/arguments, but also protecting myself from critique by acknowledging my inexperience and ignorance (“i’ve never taught”). which might have had something to do w/my “unassuming” yet “authoritative” academic talk. Others who were interviewed echoed similar views. Indeed, to qualify as a respected and respectful member of a collaborative learning community, one must appear intellectually confident but not over-confident, self- assured but not self-aggrandizing, constructive but not condescending. As with face-to-face class discussions, much of the “face-work” (Goffman, 1967) of sociable academic discourse consists in the ability to establish solidarity with group members while maintaining one’s intellectual competence and independence in the eyes of peers and assessors. What happens, then, when a writer fails to “disguise” his intellectual authority amid the social exigencies of an online discussion? Consider the following text posted by the teaching assistant in the first week’s conference:
Social academic writing 331 Sofia—others have indeed echoed and supported your intuitions throughout the ages! One of Literature’s unique affordances resides in the fact that novels, short stories, poetry, plays, film—whether “fictional” or “factual”— exist in and through narrative. And, as philosophers, literary scholars, and psycholinguists have shown, narratives are native to the organizing structures of human cognition. . . . Literary critic Barbara Hardy (1987) asserts that narrative is: a primary act of mind transferred to art from life. . . . For we dream in narrative, remember, anticipate, hope, despair, believe, doubt, plan, revise, criticize, construct, gossip, learn, hate, and love by narrative. In order really to live, we make up stories about ourselves and others, about the personal as well as the social past and future. (p. 1) Moreover, narrative as a meaning-making device can serve a deeper moral imperative. One argument suggests that sympathetic identification with a story’s characters and situations can engender greater compassion and moral awareness. Martha Nussbaum (1997) even contends that the “narrative imagination is an essential preparation for moral interaction” (p. 90). And there’s much more to this [. . .] by eminent thinkers including Roland Barthes (1982), John Dewey (1934), Maxine Greene (1995), and Jerome Bruner (2002) . . . References Barthes, R. (1982). Introduction to the structural analysis of narratives. In Susan Sontag (Ed.), A Barthes reader (pp. 251–295). New York: Hill & Wang. Dewey, J. (1934). Art as experience. New York: Minton, Balch & Co. Greene, M. (1995). Releasing the imagination: Essays on education, the arts, and social change. New York: Jossey Bass. Hardy, B. (1987). The collected essays of Barbara Hardy, Vol. 1: Narrators and novelists. Sussex: The Harvester Press. Nussbaum, M. (1997). Cultivating humanity: A classical defense of reform in liberal education. Cambridge, MA: Harvard University Press. Figure 24.1 Jane’s week 5 online post.
Scoring high on “cognitive processes,” this response invokes the hegemonic conventions of the formal academic essay, even as it gestures sociably toward affirming Sofia’s earlier post. Arguably, the studied formality of his register combined with the punctilious citation of copious scholarly references risked coming across as aggressively pedantic. Potentially belittling, too, was the assertion that “others have indeed echoed and supported your intuitions throughout the ages!”—as if suggesting the writer’s lack of originality in light of previous scholarship. Not surprisingly, perhaps, some students confessed to having interpreted this post as “pretentious” and “intimidating.” Others suggested that offline face-to-face interactions in a
a. Vocatives address multiple community members. b. Affirmation of others’ a contributions. Dina and Farah (and Nick and June), c. Hedging with “seems,” “like,” b “maybe,” and “could.” I also appreciated all of your thoughts on productivity, social processes/ d. Inclusive “we” indexes shared structures (the realities and pressures of the “actual” classroom), but I assumptions. want to think harder about striking “balances of productivity and student e. Typographic innovation: voice1 [. . .] It cseems like maybe creating balance, engaging in student asterisks replace quotation counterscript or bridging scripts is not as complicated as it looks. It seems marks. more like the work could be more ideological and pre-cursory to actual f. Capitalization for emphatic experiences than the huge task (which we’re also suggesting inhibits hedging. “productivity”) that wed perceive 3rd space creation to be. 2That’s why I g. Long, convoluted sentence found Kris’s strategic interjections about *her*e scheme of pre-service unedited, as if to evoke instruction to be so compelling. 3Applying and practicing (indeed informal “stream-ofconvincing oneself of) theory and giving student agency/voice (which is consciousness” style. not just simultaneous with being productive, that act CAN BEf itself productive) is something that should happen before people get into the classroom so that such ideas are not disruptive of our preconceived notions about what teacher-student relationships are like, what classroom discourse is like, what productivity and learning should be, etc. g The disruption of dominant ideologies (which i acknowledge are married to and maybe constitute the heart of structural processes) seems to be part of what might make this kind of exercise or theory seem so hard.
Subject: Discourse Analysis in Context, plus the counterscript Posted: Feb. 12, 2007 05:19
1 Recall and synthesize ideas from earlier posts to identify gaps for further inquiry 2 Affirmatively evaluate Kris’s “strategic interjections.” 3 Apply prior knowledge of cultural reproduction theories to evaluate theory-practice nexus and the intransigence of teachers’ beliefs.
Socio-affective processes
Online message text
Cognitive processes
Table 24.2 Annotated example of an online post from Week 5’s online computer conference
4 Identify gaps in Kris’ explanations while evaluating own knowledge gaps. 5 Venture possibilities for further syntheses of present and previous points raised in the online discussion. 6 Evaluate own arguments by inviting counterarguments.
4 I also didn’t really hear her give a direct answer about the relationship between 3rd space and student-centered learning except to say that they were not the same thing. i I have an idea about what some of the differences might be, but since i I don’t know much about the concept of student-centered learning (here we are with the disjunctures: i DIDN’T teach! h), i I could be very wrong. 5 My other thoughts on 3rd space possibilities are connected to the post I put up on the other thread. It’s only slightly related because I think I’m also trying to question the potential neglect of structures and ideologies when we talk about the possibility of a “3rd space” or even the possibility of a counterscript. Is it really?6 j k Hope you’ve all had a good weekend! l jane
h. Alternation between upper and lower case as a gesture of self-abnegation (e.g., “i DIDN’T TEACH). i. I-statements concede personal fallibility and inexperience. j. Invite comments from peers with a “semi-rhetorical” question. k. Phatic valediction. l. Lower-case sign-off to convey humility and informality.
334 W. M. Liew and A. F. Ball blended learning context had played a crucial role in mitigating interpersonal tensions arising from misinterpretations of online behaviors. As one student wrote in an online interview: I thought you [the teaching assistant] were one of the more personable, approachable people . . . But perhaps it’s because I also talk with you face-to-face and gotten to know you personally and that may have influenced me to think this way. Whatever “wordy, reference-clogged” stuff you had in your replies was more helpful than intimidating. Again, it’s because I’ve gotten to know you personally and like you as a human being:-) Although out-of-class and in-class face-to-face interactions could have ameliorated the negative effects of social-cues reduction on online sociality, our virtual discussions remained fraught with potential misunderstandings. Eventually, group norms emerged that called for the tactical coordination of formal essayist literacy with the ability to engage in relational dialogue among student-peers. Unaware of these socio-affective demands, the teaching assistant (then a second-year PhD student) had invoked in his online post an institutional power relationship by “dutifully” assuming the “discoursal identity” (Ivanicˇ, 1998) of a capable Teaching Assistant knowledgeably addressing his students. Indeed, a comment in the course evaluations suggests that the delicacy of this power balance had to some degree been neglected by the professor and her assistant: Also, sometimes I felt like I was being spoken down to by both Dr. Ball and Warren in this class. I think that cultivating safer, more respectful norms would have helped promote the risk taking I think both instructors were hoping to see. Meanwhile, an undercurrent of social tension could be discerned beneath the routine displays of online and offline sociality among the students. Particularly revealing, given its tacit character, was the consciousness of status differentials. During the seventh week of the course, one student wrote: Here I am, a master’s student in a classroom with PhD students. Sometimes I feel intimidated because I feel like I’m the freshman and you’re the seniors. However, it’s nice to hear that the “seniors” feel the same way as I. This class and the small number of students helped me open up and converse with you. I enjoy being around people that can challenge me intellectually and maybe that’s what graduate school is about; sitting in a class with people who care about the same subject as you (in our case, education).
Social academic writing 335 Others expressed insecurity over the pressure to perform in the presence of capable peers: “I’ve been putting off posting this week, mainly because I didn’t feel like I had anything fabulous to say—but hell—I can certainly handle adequate and mediocre.” If the formal evaluation of students’ online participation had been intended to motivate higher levels of intellectual engagement, they nonetheless militated against the concurrent desire for a safe learning space free from “surveillance pressures.” Some experienced mounting self- consciousness amid the unspoken competition for recognition and reward. According to one student: I don’t think it [the discussion forum] made much for the class as a community ’cos it was so competitive. I felt absolutely self-conscious. Sometimes more so than others. It depends on what the mood or tone of the thread was. Or sometimes if you were the moderator. Here, a more ambivalent picture of sociality emerged than that touted in popular endorsements of “student-centered” virtual learning. Indeed, to what degree does power function through the invisible inspecting gaze of teachers and peers? Do competitive pressures compete with cooperative ideals as graduate students assume the roles and identities conferred on them by an elite academic institution? To what extent did the hidden curriculum of grading practices result in unhealthy competition at the expense of learning? Was it reasonable to expect that the “absent presence” of an online professor would not have similarly invoked a “panopticon classroom interaction order” (Scollon & Scollon, 2004), in which “[t]he teacher speaks to everyone in the class, [and] everyone in the class who speaks speaks either to the teacher or possibly through him or her” (p. 39)? In a similar Foucauldian vein, one might ask: What internalized norms, dispositions, and desires underlie the disciplinary mechanisms of students’ academic socialization?
Conclusion The foregoing analysis has examined how learners in online academic settings construct “social presence” through their digital textual productions (cf. Gunawardena & Zittle, 1997; Rourke et al., 1999). Despite the absence of embodied social cues, computer conference discussions could be seen to generate a “hybrid” mode of formal/informal, public/private discourse, captured in the cognitive and socio-affective dynamics of social academic writing. Implicit in the dialogic nature of such writing is a critique of “essayist literacy,” whose emphasis on cognitive competences represents an anachronistic abstraction from the social realities of online collaborative inquiry. More importantly, the performance of academic writing was closely identified with identity work, as students grappled with the question: How can I construct an academically literate discoursal
336 W. M. Liew and A. F. Ball identity in relation to hegemonic institutional norms, without compromising my diplomatic identity as member of a collaborative learning community? According to Lea and Street (1998), two models of student writing are subsumed under the larger framework of Academic Literacies: a study skills model that identifies academic writing with certain technical and instrumental competences; and an academic socialization model in which students are assimilated into the cultural and institutional norms of the academy. In the third model, practices of reading and writing within academic disciplines are seen as constitutive of users’ identities within power/ knowledge arrangements. This view of literacies as social practices implicated in identity construction and power relations takes its cue from the New Literacy Studies (Baynham, 1995; Gee, 1996; Street, 1984) and its methodological ties with Critical Discourse Analysis (Fairclough, 1995; van Dijk, 1999). By framing the analysis of computer conferencing within this critical perspective, this study has attempted to show how social academic writing participates in the representation, legitimation, and negotiation of learners’ literate identities within a hidden curriculum of academic and social imperatives underpinned by institutionalized assessment practices (cf. Goodfellow & Lea, 2005; Lillis, 2001). Indeed, one might discern in students’ facility with social academic writing their complicity in the power structures that authorize their positions as academically literate subjects. Beyond trite valorizations of “difference” over “deficit” models, therefore, it is important to recognize how students skillfully enact their socialization into higher education discourse communities by performing certain literate identities within deficit-based assessment regimes. In conclusion, questions emerge that demand further investigation. What compulsory performances of academic identity are implied in students’ academic socialization? Are literacy practices fundamentally structured by the “pedagogical contract” of institutionalized assessment systems? How did the students, teacher, and teaching assistant enact their resistance toward the hegemonic norms of academic discourse through social academic writing? To what extent can “progressive” educators safely divest themselves of the authority vested in them by the institutional arrangements that unequally empower teachers and students?
Notes 1 The course drew a total of nine graduate students from the Stanford University School of Education (one Master’s student, seven PhD students, and one post- doctoral student). Aged between 24 and 36, these students included two men and seven women of African American, Asian American, and European American descent. All students were native English speakers who had had previous experiences with computer conferences in other graduate courses. 2 Students’ participation in this study was solicited only after the course had ended. At no point during the actual course did the researchers reveal their intentions to study the class’s online communications. Interviews were entirely
Social academic writing 337 c onducted by the first author, a fellow graduate student who interacted with the other students on a friendly basis. 3 All names, except for the authors of this chapter, are pseudonyms.
References Anagnostopolous, D., Basmadjian, K. G., & McCrory, R. S. (2005). The decentered teacher and the construction of social space in the virtual classroom. Teachers College Record, 107(8), 1699–1729. Aviv, R. (2000). Educational performance of ALN via content analysis. Journal of Asynchronous Learning Networks, 4(2). Retrieved December 27, 2004, from www.aln.org/alnweb/journal/Vol. 4_issue2/le/reuven/LE-reuven.htm. Baynham, M. (1995). Literacy practices. London: Longman. Bloom, B. S. (Ed.). (1956). Taxonomy of educational objectives: The classification of educational goals. New York: Longman. Bonk, C. J., & Graham, C. R. (Eds.). (2006). The handbook of blended learning: Global perspectives, local designs. San Francisco, CA: Pfeiffer. Brown, P., & Levinson, S. (1987). Politeness: Some universals in language usage. Cambridge: Cambridge University Press. Coffin, C., & Hewings, A. (2005). Engaging electronically: Using CMC to develop students’ argumentation skills in higher education. Language and Education, 19(1), 32–49. Corson, D. (1985). The lexical bar. Oxford: Pergamon Press. Daft, R. L., & Lengel, R. H. (1986). Organizational information requirements, media richness and structural design. Management Science, 32(5), 554–571. Fairclough, N. (1992). Discourse and social change. Cambridge: Polity Press. Fairclough, N. (1995). Critical discourse analysis: The critical study of language. Harlow: Longman. Fleckenstein, K. S. (2005). Faceless students, virtual places: Emergence and communal accountability in online classrooms. Computers and Composition, 22(2), 149–176. Garrison, R. D., & Vaughan, N. D. (2007). Blended learning in higher education: Framework, principles, and guidelines. San Francisco, CA: Jossey-Bass. Gee, J. P. (1996). Social linguistics and literacies: Ideology in discourses (2nd ed.). London: Falmer Press. Goffman, E. (1967). Interaction ritual: Essays on face-to-face behavior. New York: Pantheon. Goodfellow, R., & Lea, M. (2005). Supporting writing for assessment in online learning. Assessment & Evaluation in Higher Education, 30(3), 261–271. Gunawardena, C. N., & Zittle, F. (1997). Social presence as a predictor of satisfaction within a computer mediated conferencing environment. The American Journal of Distance Education, 11(3), 8–25. Halliday, M. A. K., & Matthiessen, C. (2004). An introduction to functional grammar (3rd ed.). London: Arnold. Heckman, R., & Annabi, H. (2005). A content analytic comparison of learning processes in online and face-to-face case study discussions. Journal of Computer- Mediated Communication, 10(2). Retrieved November 12, 2006, from http:// jcmc.indiana.edu/vol. 10/issue2/heckman.html. Henri, F. (1991). Computer conferencing and content analysis. In A. Kaye (Ed.), Collaborative learning through computer conferencing: The Najadeen papers (pp. 117–136). London: Springer-Verlag.
338 W. M. Liew and A. F. Ball Ivanicˇ, R. (1998). Writing and identity: The discoursal construction of identity in academic writing. Amsterdam: John Benjamins. Kiesler, S., Siegel, J., & McGuire, T. (1984). Social psychological aspects of computer mediated communication. American Psychologist, 39(10), 1123–1134. Lea, M., & Street, B. (1998). Student writing in higher education: An academic literacies approach. Studies in Higher Education, 23(2), 157–172. Lillis, T. (2001). Student writing: Access, regulation, desire. London: Routledge. Luppicini, R. (2007). Review of computer mediated communication research for education. Instructional Science, 35(2), 141–185. Potter, A. (2008). Interactional coherence in asynchronous learning networks: A rhetorical approach. The Internet and Higher Education, 11(2), 87–97. Rice, R. E. (1993). Media appropriateness: Using social presence theory to compare traditional and new organizational media. Human Communication Research, 19(4), 451–484. Ridings, C. M., Gefen, D., & Arinze, B. (2002). Some antecedents and effects of trust in virtual communities. Journal of Strategic Information Systems, 11(3–4), 271–295. Rogoff, B. (1994). Developing understanding of the idea of communities of learners. Mind, Culture, and Activity, 1(4), 209–229. Rourke, L., Anderson, T., Garrison, D. R., & Archer, W. (1999). Assessing social presence in asynchronous, text-based computer conferencing. Journal of Distance Education, 14(3), 51–70. Retrieved December 27, 2007, from http://cade. athabascau.ca/vol. 14.2/rourke_et_al.html. Scollon, R., & Scollon, S. B. (1981). Narrative, literacy, and face in interethnic communication. Norwood, NJ: Ablex. Scollon, R., & Scollon, S. B. (2004). Nexus analysis: Discourse and the emerging Internet. London: Routledge. Street, B. (1984). Literacy in theory and practice. Cambridge: Cambridge University Press. Tidwell, L. C., & Walther, J. B. (2002). Computer-mediated communication effects on disclosure, impressions, and interpersonal evaluations: Getting to know one another a bit at a time. Human Communication Research, 28(3), 317–348. Toulmin, S. (1958). The uses of argument. Cambridge: University Press. van Dijk, T. A. (1999). Critical discourse analysis and conversation analysis. Discourse and Society, 10(4), 459–450. van Eemeren, F. H. (Ed.). (2001). Crucial concepts in argumentation theory. Amsterdam: Amsterdam University Press. Walther, J. B. (1996). Computer-mediated communication: Impersonal, interpersonal, and hyperpersonal interaction. Communication Research, 23, 1–43. Wegerif, R. (1998). The social dimension of asynchronous learning networks. Journal of Asynchronous Learning Networks, 2(1), 34–49. Woods, R., & Ebersole, S. (2003). Becoming a “communal architect” in the online classroom: Integrating cognitive and affective learning for maximum effect in web-based learning. Online Journal of Distance Learning Administration, 6(1). Retrieved May 30, 2007, from www.westga.edu/ %7Edistance/ojdla/spring61/ woods61.htm.
25 Between peer review and peer production Genre, wikis, and the politics of digital code in academe Doreen Starke-Meyerring McGill University, Canada
Increasingly, digital technologies have become the subject of vigorous debate among academics about the future of academic writing and publishing (Bazerman, 2007; Willinsky, 2006; Solomon, 2007). One of the technologies that has attracted particular interest is that of wikis, a collaborative web-based writing and editing software famous for enabling the largest collaborative writing and knowledge-making project in human history, the Wikipedia. An openly accessible online encyclopedia with millions of entries collaboratively created and constantly updated in more than 200 languages by hundreds of thousands of contributors worldwide, the Wikipedia has been identified as “one of the most successful collaborative enterprises . . . in the first five years of the twenty-first century” (Benkler, 2006, p. 70). Not surprisingly, wikis have also garnered the attention of academics, giving rise to numerous questions: To what extent could wikis as spaces for collaborative knowledge making, sharing, and publishing take hold in academe? What roles might they play? What consequences might they have for academic knowledge production? More importantly for writing studies researchers, these questions also beg larger questions such as those raised by the IText working group (Geisler et al., 2001) about how theoretical traditions in writing studies can account for the workings of digital writing spaces or ITexts with their “blend of IT [information technologies] and texts” (p. 270). This chapter addresses these questions through the lens of genre theory as a theory that holds “enormous potential to contribute to our understanding of . . . ITexts” (Geisler et al., p. 293), illustrating how an integrated framework consisting of genre and technology theories (Starke-Meyerring, 2008) can help account for ITexts as blends of text and technology. For this purpose, I begin by examining the established practices that have governed academic writing. Although situated in diverse disciplinary, institutional, and national academic traditions, academic writing is largely governed by a set of practices called peer review, which therefore constitutes my focus. I then draw on critical theories of technology to examine the digital code of wikis as currently realized in the open- source software Mediawiki, the software that has enabled the Wikipedia,
340 D. Starke-Meyerring for the ways in which it might interact with established practices of peer review. To illustrate how academics negotiate this interaction, I then provide two brief examples of active academic wiki spaces, the Scholarpedia, and OpenWetWare, both of which use the same open-source wiki code, Mediawiki, but produce different solutions to the interaction between established practices and digital code. I conclude with implications of a critical genre-technology framework for the study and teaching of writing in digital environments.
Peer review in academe: a genre perspective As the dominant practice of verifying and shaping academic knowledge, peer review is commonly defined as the evaluation of proposed publications, presentations, or research projects by a few (typically 2–4) experts in the discipline, whereby the most highly valued form of this practice, double-blind peer review, involves concealing the identity of the authors and of peer reviewers (Bazerman, 1988; Benos et al., 2007; Biagoli, 2002; Shatz, 2004; Zuckerman & Merton, 1971). Through the lens of genre theory, discursive practices, such as those of peer review, are regularized through genres, the routine patterns of social action that emerge and evolve in human collectives over time because they meet recurring needs or exigencies and thus accomplish the work of these collectives (e.g., Artemeva & Freedman, 2006; Bazerman, 1988, 1994, 2002; Bazerman & Prior, 2005; Coe, Lingard, & Teslenko, 2002; Devitt, 2004; Giltrow, 1994; Miller, 1984; Paré, 2002, 2005; Paré, Starke- Meyerring, & McAlpine, forthcoming; Schryer, 1993). As a result of their evolution through recurrent situations, genres gradually become the way things are done—common sense (Paré, 2002)—and play a strong normalizing role. Indeed, over time, the exigencies that gave rise to them originally are often forgotten, although they continue to shape assumptions and practices of participants. Retrieving these forgotten exigencies is therefore vital for uncovering the cultural logic of genres from beneath the cloak of their normalcy. Having originated with the emergence of scientific societies, such as the Royal Society of London, and scientific publication enabled by the printing press (Bazerman, 1988; Biagioli, 2002; Eisenstein, 1979; Zuckerman & Merton, 1971), peer-review practices evolved from state and religious censorship bodies instituted to curtail the spread of potentially incendiary messages. Overwhelmed by the explosion of printed materials and considering the technical treatises of the Society to be of relatively low risk, state and religious censorship bodies granted the Royal Society the right to self- censor its publications following established protocols (Biagioli, 2002). Gradually, new needs emerged for protecting the reputation of the Society and its publications to ensure continued royal support, thus transforming the state-originated disciplinary practices into internalized disciplinary procedures (Biagioli, 2002). Nevertheless, as Biagioli maintains, despite the
Between peer review and peer production 341 eventual decline of absolutist forms of governments and censorship, the censorship and selection function of peer review remained and “eventually came to characterize the whole of academic and university science” (p. 32). To achieve their outcomes—be these the censorship and selection of publications or the protection of reputations—genres assemble a social order, regularizing who participates in what role, what is sayable and doable, and what strategies are habitually used. In Paré’s (2005) words, genres function as “systems of discourse control, more or less prescribed and explicit that determine who speaks (or writes), as well as when, why, where, and to whom” (p. 80). In peer review, participants fulfill clearly delineated roles—those of author, editor, or referee, with the assumption that referees produce impartial, disinterested reviews of the work at hand for publication, revision, or rejection—an assumption rooted in the enlightenment ideal of the rational individual, unconstrained by their situatedness in relation to an audience, community, or their own interests; and therefore capable of disinterested autonomous judgment through pure universal reason (Henry, 2007). As the practices of peer review unfold repeatedly, so do the Enlightenment assumptions on which they rest. Because of their repeated unfolding, genres have (re)productive force; they contribute to “the stabilization and (re)production of social institutions and communities” (Bazerman & Prior, 2005, p. 8). With its reliance on established disciplinary experts, the peer-review system, for example, has been instrumental in producing disciplinarity by reinforcing existing disciplinary orthodoxies, values, norms, and epistemological commitments, thus allowing for the stabilization of disciplinary knowledge-making practices, and ultimately for the production of highly specialized knowledge. As Biagioli (2002) notes, peer review functions to (re)produce a “distinctive kind of discipline” for academics, “something that is simultaneously repressive, productive, and constitutive of their knowledge” (p. 11). On the flipside, because of whom the genre system includes (disciplinary experts) and whom it excludes (experts from other disciplines, practitioners, and the public), the system has privileged disciplinary concerns over interdisciplinary, practitioner, or public concerns. In addition, peer review has (re)produced individual or small-group authorship, along with clearly delineated, portable, and tradable knowledge products, such as articles or books, that lend themselves well to assigning and protecting scholarly recognition and intellectual property rights. Today, this function of peer review reproduces critical economic practices of academic publishers and of universities by securing intellectual property and by serving as the principal practice for distributing research funding; for making merit, tenure, and promotion decisions; and for securing funding since peer-reviewed publications are central to winning grants and to building a university’s reputation and thus to attracting sponsorship. Their normalizing power and their function in (re)producing the material practices of institutions bestow genres with formidable conservative force and stability.
342 D. Starke-Meyerring This is not to stay that genres are static or defy change. Quite the contrary, genres are dynamic, existing in a constant productive tension between stability and change. As Schryer (1993) stressed, genres are “stabilized-for-now or stabilized-enough sites of social and ideological action” (p. 108). To some extent, their dynamic nature is produced by internal conflicts between the residual cultural logic of their origins and new situations faced by participants. For example, with its Enlightenment denial of the deeply rhetorical, interested, highly situated, and contingent nature of knowledge making, the peer-review system has traditionally provided little space for addressing conflicts of interest (e.g., for financial reasons or for priority) or incommensurable epistemological paradigms (Kuhn, 1972) between authors and reviewers. Accordingly, the system has been shown to reject publications that disrupted dominant research paradigms, but were later recognized with the Nobel Prize for their paradigm- shifting significance (Kilwein, 1999). Peer review has also been questioned for its failure to identify inaccurate or fraudulent claims (Solomon, 2007), for its limited effect on improving the quality of publications (Jefferson, Wagner, & Davidoff, 2002), for its bias against researchers from smaller, less well-known research institutions (Peters & Ceci, 1982), and for its marginalization of researchers from diverse backgrounds who may question established dominant discursive practices (Kumashiro, 2005), which peer review is designed to reinforce. In addition to internal tensions, the normalcy of established genres can be challenged or “cracked” (Paré, 2002) in a number of ways, including through technologies.
Wikis as peer production code: a critical technology perspective To account for the interactions between the genres with which participants approach a digital writing space, such as a wiki, and the technology, that is the wiki software code that inscribes the space, I draw on critical theories of technology (e.g., Benkler, 2006; Feenberg, 2002; Lessig, 2006). Most importantly, from this perspective, technologies are not mere tools or mediating means. Instead, technologies are highly political artifacts, whose design, use, and regulation are deeply implicated in reproducing, challenging, or reshaping existing material and social orders. Feenberg (2002) describes technologies as creating “a framework of activity, a field of play” (p. 82), which constitutes a “scene of struggle . . . in which civilizational alternatives contend” (p. 15). Referring specifically to software code, Lessig (2006) stresses that code constitutes the “architecture” or “built environment” (p. 121) of digital writing spaces, enabling or constraining certain social practices at the cost of others. Similar to genre theory, critical theories of technology emphasize the sociohistorical situatedness of technology. Most immediately, the development of wiki code is attributed to Ward Cunningham, a software programmer who developed the wiki software as a solution to a need for
Between peer review and peer production 343 collaborative problem solving in software design (W. Cunningham, personal email, September 10, 2007). However, wiki software also exists in a larger sociohistorical shift from proprietary production to large-scale peer production, which Benkler (2006) describes as a new mode of production that is “radically decentralized, collaborative, and nonproprietary; based on sharing resources and outputs among widely distributed, loosely connected individuals who cooperate with each other without relying on either market signals or managerial commands” (p. 60). For Benkler, wikis are the example par excellence of peer production not only because the software code facilitates peer production, but also because it reflects the spirit of the open-source movement in software programming, a movement designed to develop alternative software production modes—peer production—in the face of the increasing dominance of proprietary software. Rather than limiting software development and debugging to a few programmers to protect proprietary rights for sales, the open-source model opened up the code for collaboration and sharing through a license called GNU, which allowed anyone to develop and debug the software under the condition that the new version would likewise be freely available (Benkler, 2006). With its origins in the facilitation of peer production, wiki code questions the epistemological politics inscribed in peer review in a number of ways. For example, the code allows for gathering different participants and blends their roles. To facilitate peer production, the code privileges multiple collaborative authorship with editing tabs placed by default on each page and on each section; history tabs placed on each page for easy version tracking, comparing, or reversal; and watchlists for participants to follow peer contributions to the space. However, the code does not facilitate fine-grained distinctions among participants or hierarchical approval processes ascribed to different roles, thus blending the roles of authors, reviewers, and editors. Moreover, the code facilitates deliberation among participants through the “discussion” tab on each page, which allows participants to negotiate divergent perspectives and to deliberate the direction of a page. Rather than masking the epistemological, ontological, and ideological commitments of anonymous reviewers (Kumashiro, 2005) and possibly suppressing incommensurable theoretical paradigms, the open discussion spaces provided by wiki code facilitate the public deliberation of such commitments as the basis for knowledge making. It is through facilitating these practices of collaboration and deliberation for peer production that wiki code enables participants to achieve their ideals of quality. In Wikipedia, for example, “featured articles,” those considered of particularly high quality by Wikipedians, have been shown to have both a higher number of contributors and a higher number of edits (Wilkinson & Huberman, 2007). In this way, wiki code facilitates an epistemological politics that thrives on the deeply rhetorical nature of knowledge making through constant deliberation. Such alternative epistemological politics have consequences for disciplinary knowledge making, the economics of scholarly reputation, and
344 D. Starke-Meyerring intellectual property. With its inclusiveness and ability to coordinate and track the work of large numbers of members, wiki code has the potential to enable new levels of disciplinary development, integration, and negotiation of parallel or conflicting paradigms. At the same time, disciplinarity might be peer produced, that is by peers openly deliberating disciplinary norms, orders, practices, and values. For the economics of scholarly reputation, wiki code would still allow for contributions and reputations of individual contributors to be tracked, for example, through lists of contributor edits or the “popular pages” function. However, the code undermines the intellectual property practices so easily facilitated by peer review. Rather than clearly bounded, portable, definitive knowledge products by individual authors, wiki code privileges peer-produced, multi- perspectival, constantly negotiated, and continuous knowledge. Not surprisingly, the Wikipedia, with all work peer produced, has regulated its intellectual property practices in line with those of the open-source Mediawiki software code on which it runs, using a GNU Free Documentation License, which allows for unlimited copying as long as the site is acknowledged as source and derivative works are likewise free for copying (Wikipedia, 2007).
Negotiating the tensions between genre and digital code As this analysis shows, the genre system of peer review and the code of wikis exist in considerable tension, with each favoring a particular epistemological order for knowledge production with different outcomes. Researchers may resolve this tension in a number of ways: (1) they may reshape, circumvent, or regulate the wiki software code so as to reproduce established peer-review practices, perhaps by stipulating individual authorship for pages and anonymous reviewer functions for the discussion tabs or by linking the wiki to eventual publication in traditional, institutionally sanctioned venues; (2) they may choose to relegate the writing space to knowledge-making practices outside the purview of peer review, such as to the sharing of raw data or to teaching; (3) researchers may decide to use the code as an opportunity to “crack” (Paré, 2002) the established genred practices of academic writing, perhaps opening up the genre system of peer review for deliberation, reconsideration, or even new visions for peer- produced academic knowledge. Given the conservative force of genres as historical practice constitutive of institutional stability, however, the latter option seems less likely, although it may well hold the greatest potential for academic knowledge making in digital environments. As Bazerman (2007) observes, “the large economic stakes along with the complexity, stability and power of . . . [existing] social systems mean that the technology gets designed to facilitate the existing work and arrangements, making it cheaper and quicker, but not disrupting it.” Accordingly, Bazerman suggests that technologies are prone to be “bent to the needs of the robust social system.”
Between peer review and peer production 345 Two prominent examples of active research wikis, the Scholarpedia and OpenWetWare, both using the same Mediawiki code that also houses the Wikipedia, may suffice here to illustrate how these tensions between genred practices and digital code play out.
Scholarpedia: reproducing peer review in a peer-production space As one of the early and perhaps most active research wikis, the Scholarpedia describes itself as “the free peer reviewed encyclopedia written by scholars from all around the world.” Created and run by one scientist, Dr. Eugene M. Izhikevich, with the title “Editor-in-Chief,” the site currently coordinates roughly 3,000 users around three encyclopedias, the Encyclopedia of Computational Neuroscience, the Encyclopedia of Dynamical Systems, and the Encyclopedia of Computational Intelligence, with plans to expand the coverage of each of these encyclopedias. As the phrase “free peer reviewed encyclopedia” in its tagline suggests, the site is closely aligned with established practices of peer review and quite openly solves the tension between wiki code and peer-review practices by bending and regulating the code to make peer-review practices “more efficient.” On its home page, this emphasis is highlighted several times and rationalized with facilitating citation and publication in traditional journals as well as the eventual publication of peer-reviewed articles in a print version of the encyclopedia: “Each article is anonymously peer reviewed to ensure accurate and reliable information.” Aware that participants come with expectations grounded in traditional genres, the site justifies its hardcopy strategy with the need for “archiving” and the concern that “many academicians have a preconception that the prestige of an online article is not as high as that of a printed one.” Hence, the site goes to great lengths to reproduce the feel of print genres and to make the wiki “similar to a print journal so that Scholarpedia articles could be cited.” The site even provides page numbers not only for each article, but also for each revision, along with instructions for how to cite both “approved articles” and revisions. In addition to aligning the purpose of the wiki as closely as possible with peer-review practices, the site reproduces the epistemological order of peer review, distinguishing clearly between authors or curators of articles and reviewers, providing detailed task descriptions for each. Accordingly, the discussion tab of the wiki code that would potentially allow for collaborative shaping of an article and for the deliberation of incommensurable research approaches, conflicts of interests, etc., has been turned into a tab for anonymous peer review. Reviewers are also encouraged to add comments directly into the article, which the site claims makes these procedures of peer review more efficient than those in print journals: “instead of explaining why a statement is wrong and how to fix it, the reviewer just fixes it” (Instructions for authors page).
346 D. Starke-Meyerring The site also goes to great lengths to ensure credit and reputation for authorship as well as for reviewing. In fact, the institutional reputation system reproduced by peer review is made more efficient by invitation-only author assignments, author elections, and an intricate point-based scholar reputation system, according to which participants collect points for reviewing articles and gain or lose points for revisions to other articles, depending on how the curator of an article judges such revisions on a scale from −10 (“mostly wrong”) to +10 (“major contribution”). Different numbers of points translate into different levels of rights and privileges in the site and ultimately reputation for participants. Intellectual property is also protected by copyright; as the Instructions for authors page explains, materials are “protected by Scholarpedia copyright,” so nobody can reproduce these materials without permission of Scholarpedia or the author. The site does, however, allow for authors to choose from three copyright options: “(1) author owns the copyright and licenses the content to Scholarpedia, (2) Creative Commons, or (3) GNU FDL,” although there is no explanation for how a publisher of a print version of the encyclopedia might handle a GNU FDL license as such a license requires that the material also be available for further reproduction and use. The Scholarpedia, then, is largely an example of the way in which the alternative peer-production potential offered by the digital code is bent and circumvented to reproduce established practices of peer review. No open spaces for the deliberation of these genred practices, their internal tensions, the role of conflicts of interest, or incommensurable research paradigms are offered; the peer-production potential of the wiki remains untapped. The site is powerful testimony to the conservative force of genre as reproducing the traditional institutionally sanctioned practices of peer review.
OpenWetWare: encouraging peer production Having received NSF funding as well as media attention similar to that of Scholarpedia, OpenWetWare portrays itself as “an effort to promote the sharing of information, know-how, and wisdom among researchers and groups who are working in biology & biological engineering” (OWW Home Page). Coordinated by the Endy and Knight biology labs at MIT, the site offers research labs and groups opportunities to host their web presence in the wiki, create spaces for their classes, keep online lab notebooks, share protocols and techniques for lab procedures, and maintain blogs. Its stated purpose is to provide a place for labs, individuals, and groups to organize their own information and collaborate with others easily and efficiently. In the process, we hope that OWW will not only lead to greater collaboration between member groups, but also provide a useful information portal to our colleagues, and ultimately to the rest of the world. (OWW About page)
Between peer review and peer production 347 As this list of activities suggests, the site circumvents the established practices of peer review by limiting its activities to knowledge-sharing practices that do not fall under its purview but were in the past constrained by print technologies, such as the sharing of lab protocols. In addition, the site focuses heavily on new researchers and teaching and learning—again academic realms outside dominant practices of peer review. As the site states in its Frequently Asked Questions, Many of the researchers on OpenWetWare are relative newcomers to biological research and are struggling to cope with the steep learning curve in biology. We’re trying to accumulate enough information on OpenWetWare such that a person with little lab experience could, for example, go to the DNA Ligation page, read it and be able to carry out a successful DNA ligation reaction. As the original name of the site, “Endipedia,” suggests, original plans were more focused on reproducing dominant genred practices of peer- reviewed encyclopedic writing. Some of the comments on the community portal refer to such ambitions as well, but indicate resignation, or perhaps simply a temporary focus on non-peer-reviewed knowledge sharing: “Anyone interested in being a curator for a review in their field of expertise? I can’t promise you scientific rewards yet, but I suspect this will become an accepted form of merit down the line.” Embracing instead the peer-production potential of wiki code, the site also facilitates an alternative epistemological order to that of peer review. References to “our colleagues,” the “user community,” and numerous invitations to participants to join the steering committee or to otherwise participate in shaping the site, for example, by sharing ideas, requesting code changes, or joining mailing lists, emphasize the peer-production atmosphere of the space. Accordingly, the space is organized around a collaborative order—there are no distinctions between authors and reviewers, and there are no intricate reputation-ranking mechanisms, and instead of being run by one researcher, the wiki is governed by a collaborative structure open to all researchers. The Etiquette page also directly encourages authors to articulate some of their epistemological commitments when they edit pages: “If you change the existing protocol, please justify and explain your changes.” The site also reflects its support for peer-production practices in its approach to intellectual property by embracing a creative commons license as well as a GNU FDL license to facilitate easy sharing while also allowing contributors to retain their copyright if they so wish. In its Frequently Asked Questions, the site even anticipates concerns about a possible shift toward proprietary models in the future and addresses them outright, declaring that “OpenWetWare will always remain a free and open site dedicated to encouraging and celebrating open scientific information.”
348 D. Starke-Meyerring In contrast to Scholarpedia, OpenWetWare, then, heavily embraces the peer-production potential of the wiki code, albeit in knowledge-making practices outside the purview of peer review.
Conclusion As this exploration of the interaction between genre and technology in academic wikis illustrates, digital writing spaces assemble complex dynamic constellations of digital software code and participants steeped in diverse genred practices they bring to their interaction in the space. Neither these genred practices, nor the code that programs these spaces are neutral; rather they inscribe sociohistorically situated cultural logics, social orders, and epistemologies, whereby the digital code can be designed to reproduce, question, or challenge the established genred practices of participants. In achieving whatever “stabilized-for-now” (Schryer, 1993) equilibrium may be possible in a digital writing space with its constantly recodable architecture, the power of genred practices, however, may be substantial since they are often deeply implicated in the reproduction of institutional practices, including economic ones. Accordingly, what may at first sight appear to be a matter of remediation or the “representation of one medium in another” (Bolter & Grusin, 2000, p. 45) may actually be the result of an institutional battle won by the conservative force of genre as normalized and institutionally sanctioned practice and an opportunity lost for the consideration of alternative practices. A critical genre-technology theoretical framework suggests, then, that digital writing spaces are not a neutral medium into which genred practices simply travel; rather, digital writing spaces are sites of struggle between alternative social orders. Through the lens of a critical genre-technology framework, the defining characteristic of digital writing spaces, then, may not be representation, but rather contestation. A key question for digital writing spaces then becomes how the tension between established social practices and digital code are negotiated—whether the code will be designed or “bent” (Bazerman, 2007) to reproduce established practices and social orders or whether these established practices and social orders become “cracked” (Paré, 2002) and available for deliberation, reconsideration, and—perhaps even conscious change. Most importantly, perhaps, what opportunities, roles, and preparation do participants have for partaking in such decisions? These questions have implications for how we study, design, and teach writing in such spaces. A critical genre-technology framework allows us to see the contestation in these spaces and what is at stake; it allows us to examine the political nature of design, where even seemingly innocuous practices of user needs assessment and usability testing are highly political acts of either reinforcing a particular social order or offering alternatives. Most importantly, the framework helps us account for the highly political blend of genre and technology that characterizes digital writing spaces as
Between peer review and peer production 349 scenes of struggle over the future of our societies, institutions, and communities and thus as vital scenes of civic engagement for us as teachers and students of writing.
References Artemeva, N., & Freedman, A. (Eds.). (2006). Rhetorical genre studies and beyond. Winnipeg: Inkshed Publications. Bazerman, C. (1988). Shaping written knowledge: The genre and activity of the experimental article in science. Madison, WI: University of Wisconsin Press. Bazerman, C. (2007). Electrons are cheap, society is dear. Keynote address delivered at the Conference of the Canadian Association of Teachers of Technical Writing. Saskatoon, SK, May 27–29. Bazerman, C., & Prior, P. (2005). Genre, disciplinarity, interdisciplinarity. In J. Green, R. Beach, M. Kamil, & T. Shanahan (Eds.), Multidisciplinary perspectives on literacy research (2nd ed., pp. 133–178). Urbana, IL: National Conference on Research in English. Benkler, Y. (2006). The wealth of networks: How social production transforms markets and freedom. New Haven, CT: Yale University Press. Benos, D. J., Bashari, E., Chaves, J. M., Gagger, A., Kapoor, N., LaFrance, M., et al. (2007). The ups and downs of peer review. Advances in Physiology Education, 31, 145–152. Biagoli, M. (2002). From book censorship to academic peer review. Emergences, 12, 11–45. Bolter, J., & Grusin, D. (2000). Remediation: Understanding new media. Cambridge, MA: MIT Press. Coe, R., Lingard, L., & Teslenko, T. (Eds.). (2002). The rhetoric and ideology of genre. Cresskill, NJ: Hampton. Devitt, A. J. (2004). Writing genres. Carbondale, IL: Southern Illinois University Press. Eisenstein, E. L. (1979). The printing press as an agent of social change: Communications and cultural transformations in early-modern Europe. Cambridge: Cambridge University Press. Feenberg, A. (2002). Transforming technology: A critical theory revised. Oxford: Oxford University Press. Geisler, et al. (2001). IText: Future directions for research on the relationship between information technology and writing. Journal of Business and Technical Communication, 15(3), 269–308. Giltrow, J. (1994). Genre and the pragmatic concept of background knowledge. In A. Freedman & P. Medway (Eds.), Genre and the new rhetoric (pp. 155–178). Madison, WI: University of Wisconsin Press. Henry, J. (2007). Science and the coming of the Enlightenment. In M. Fitzpatrick et al. (Eds.), The Enlightenment world (pp. 10–26). London: Routledge. Jefferson, T., Wagner, E., & Davidoff, F. (2002). Measuring the quality of editorial peer review. Journal of the American Medical Association, 287(21), 2786–2790. Kilwein, J. H. (1999). Biases in medical literature. Journal of Clinical Pharmacy and Therapeutics, 24, 393–396. Kuhn, T. S. (1972). The structure of scientific revolutions (3rd ed.). Chicago, IL: University of Chicago Press.
350 D. Starke-Meyerring Kumashiro, K. K. (2005). Thinking collaboratively about the peer-review process for journal-article publication. Harvard Educational Review, 75, 257–285. Lessig, L. (2006). Code version 2.0. New York: Basic Books. Miller, C. (1984). Genre as social action. Quarterly Journal of Speech, 70, 151–167. OpenWetWare (2008). Retrieved April 15, 2008, from www.openwetware.org. Paré, A. (2002). Genre and identity: Individuals, institutions, and ideology. In R. Coe, L. Lingard, & T. Teslenko (Eds.), The rhetoric and ideology of genre (pp. 57–71). Cresskill, NJ: Hampton. Paré, A. (2005). Texts and power: Toward a critical theory of language. In L. Davies & P. Leonard (Eds.), Social work in a corporate era: Practices of power and resistance (pp. 76–90). Hants, UK: Ashgate. Paré, A., Starke-Meyerring, D., & McAlpine, L. (forthcoming). The dissertation as multi-genre: Many readers, many readings. In C. Bazerman, D. Figueiredo, & A. Bonini (Eds.), Genre in a changing world. West Lafayette, IN: Parlor Press and WAC Clearinghouse. Peters, D. P., & Ceci, Stephen J. (1982). Peer-review practices of psychological journals: The fate of published articles, submitted again. Behavioral and Brain Sciences, 5, 187–195. Scholarpedia. (2008). Retrieved April 15, 2008, from www.scholarpedia.org. Schryer, C. F. (1993). Records as genre. Written Communication, 10, 200–234. Shatz, D. (2004). Peer review: A critical inquiry. New York: Rowman & Littlefield. Solomon, D. J. (2007). The role of peer review for scholarly journals in the information age. Journal of Electronic Publishing, 10. Retrieved February 16, 2007, from http://hdl.handle.net/2027/spo.3336451.0010.107. Starke-Meyerring, D. (2008). Genre, knowledge, and digital code in web-based communities: An integrated theoretical framework for shaping digital discursive spaces. International Journal of Web-Based Communities, 4(4), 398–417. Wikipedia. (2007). GNU Free Documentation License. Retrieved August 20, 2007, from http://en.wikipedia.org/wiki/Wikipedia:Text_of_the_GNU_Free_Documentation_License. Wilkinson, D. M., & Huberman, B. A. (2007). Assessing the value of cooperation in Wikipedia. First Monday, 12(4). Retrieved August 20, 2007, from www.firstmonday.org/issues/issue12_4/index.html. Willinsky, J. (2006). The access principle: The case for open access to research and scholarship. Cambridge, MA: MIT Press. Zuckerman, H., & Merton, R. K. (1971). Patterns of evaluation in science: Institutionalization, structure and functions of the referee system. Minerva, 9, 66–100.
Part V
Theories and methodologies for understanding writing and writing processes
26 Writing in multiple contexts Vygotskian CHAT meets the phenomenology of genre David R. Russell Iowa State University
Texts largely structure the activity of the modern world and—a fortiori—the postmodern world, with its reliance on hypertextual networks. But they do so always in contexts—often in multiple contexts. Texts are given life through activity, through use in context(s). And to study them without studying their contexts (as has often been the case) is to separate writing from its very being. Yet the problem of theorizing context and contexts, plural—and of operationalizing the theory in empirical research—is one of the thorniest but most important in writing studies. Sociocultural theories of literacy emphasizing the role of context and contexts have been developed in the last 25 years in North American writing research and applied in a number of fields: primarily organizational (business, technical, and scientific) communication and education (Russell, 1997b; Bazerman & Russell, 2003). In this chapter I sketch out elements of a theory of multiple contexts based on a synthesis of Vygotskian cultural-historical activity theory (growing out of his notion of tool mediation) with a theory of genre as social action (Miller, 1984, 1994) (growing out of Alfred Schutz’s phenomenology). The relationship between cultural-historical activity theory (CHAT) and genre as social action has been developed in various ways by several North American writing researchers to provide a principled way of analyzing written texts in their multiple contexts, such as Bazerman’s theory of genre systems (1994, 2004), Prior’s theory of laminated activity (1998, 2007), and the Canadian genre research group (Dias, Freedman, Medway, & Paré, 1999). My particular contribution has been to analyze the ways writing is deployed and learned across contexts by seeing genre systems operating in both the social-psychological (subjective and intersubjective) plane and the sociological (objective and institutional) plane. I have turned to Vygotskian theories for the former and Schutzian theories for the latter. The key to synthesizing these two, for me, has been Miller’s idea of genre as social action, drawn from Schutz. I return to Schutz’s phenomenology and methodology to develop the theory of genre as social action to allow the analyst to make principled meso-level (institutional) and macro-level (ideological) generalizations based on observations of micro-level phenomena, and thus to trace the uses of writing across scales of time and level of generality.
354 D. R. Russell What the synthesis of AT and phenomenological genre theory helped me to do is analyze typifications of participants operating in multiple contexts, realized and analyzable in specific and concrete ways, in relation to reading and writing, the genre systems or enduring types, and how those typifications both enable and hinder participants from mastering the situation, from learning (Russell, 1997a). I will illustrate with examples from my group’s research on higher education and workplace pedagogy: (1) studies of the genre systems of history, and (2) studies of online multimedia simulations we are developing to represent engineers’ communicative activity within and between complex organizations.
Writing as tool mediation: Vygotskian CHAT and multiple contexts In this synthesis I am developing, context is not a problem of describing what is outside of the mind, as in some AI cognitive approaches. Social context is not what contains the interaction. Context (con-text) is actually from the Greek term for weaving, as in textile, or texture. In this sense, context is what is “woven together with” (Cole, 1996) a weaving together of people and their tools in complex networks. The network is the context. And network or system metaphors dominate. For Vygotsky and the tradition of cultural psychology he generated, that weaving together of people and tools is mediated activity. Subjects act upon objects not directly but through tools, often by marks on surfaces, writing, texts, as well as sounds in the air, both beyond and within any immediate situation (as Prior, 2007, points out). Marks on surfaces activate people’s thoughts, direct their attention, coordinate their actions, provide the means of relationship. It is in the contexts of their activities that people consider texts and give meaning to texts. Engeström’s (1987) version of activity theory (Figure 26.1) expands Vygotsky’s basic mediational triangle (subject-tools-object) to consider other essentials for making sense of activity, and Engeström calls this unit of analysis the activity system. This expanded model adds rules or norms, community, and division of labor, to provide an expanded unit of analysis for describing activity systems. Note that this neat diagram describes a very messy network. The direction or motive of an activity system and its object are contested, as subjects bring many motives to a collective interaction. Indeed, the division of labor in the system itself guarantees diversity. Dissensus, resistance, conflicts, and deep tensions are constantly produced in activity systems. In Engeström’s version of AT, these tensions within and among activity systems are viewed as symptoms of deeper dialectical contradictions, “historically accumulating structural tensions within and between activity systems” (Engeström, 2001, p. 137). All human activity is contradictory at a very basic level. Human actions are at once individual and social. In each
Writing in multiple contexts 355 Tools
Subject(s)
Rules/norms
Object/motive
Community
Outcome
Division of labor
Figure 26.1 Engeström’s (1987) expanded mediational triangle: an activity system.
culture and each activity system specific contradictions arise out of the division of labor. These contradictions are the source of discoordinations, tensions, and conflicts. In complex activities with fragmented division of labor, the participants themselves have great difficulties in constructing a connection between the goals of their individual actions and the object and motive of their collective activity. Within these contradictions, the identities of the participants are also formed and negotiated. But to theorize the ways texts mediate activity across different contexts, one must theorize the relations of all these elements in multiple activity systems, what Engeström, Engeström, and Kärkkäinen (1995) call polycontextuality. Participants within one activity system, one context, come from various contexts, and will enter various contexts. And they interact with subjects in other contexts or activity systems. To understand the various ways participants interpret and use the tools, object, motive, rules/ norms, etc. of an activity system, it is often necessary to analyze the relations among various contexts. We are now discussing context not in material terms alone, but also in terms of the structures of consciousness as experienced from the first- person point of view, whether first-person singular or first-person plural. And here concepts from phenomenological approaches to sociology, particularly Alfred Schutz, have been helpful understanding writing in multiple activity systems or contexts.
Genre as social action: Schutz and multiple contexts Vygotsky and Schutz, though from different fields (psychology and sociology) and traditions (Soviet Marxism and Western European phenomenology), share several crucial understandings of the relation between thought and action, communication and contexts, or situations (Table 26.1). For both theorists, humans act on the world using tools, including signs. “A tool is a thing-in-order-to,” Schutz (1967) says. “It serves a useful purpose and for the sake of this purpose it was produced” (p. 201). A key concept for both is intentionality: consciousness and action are
356 D. R. Russell Table 26.1 Shared understandings between thought and action, communication and contexts, and situations Vygotsky
Schutz
Tool mediation
Tool as thing-in-order-to-do
Behavior object-oriented
Actions chosen for relevance
Knowledge intersubjective
Knowledge intersubjective
Society historical
Society historical
Internationalization of social
Sedimentation of experience
Externalization
Objectivation
Generalization
Typification
Language central to thought
Language central to thought
always directed toward something, some object. Activity is oriented to an object, as Vygotsky insists, or as Schutz says, chosen for its relevance. It is motivated by some need. It is, then, always related to a context/s. For both theorists, knowledge is socially derived, intersubjective. And human thought and action are deeply historical. Moreover, Vygotsky describes behavior, language use, and thought arising out of concrete social interaction, as a process of internalization on the psychological plane and then of externalization in concrete social action, most importantly communication. Similarly, for Schutz knowledge developed socially is internalized through a process of what he calls sedimentation of experience in individual consciousness and then externalized through what he calls objectivation in material form, primarily communication (Schutz & Luckmann, 1973). But the most important connection is that for both, thought and language are both the result of an ongoing, dynamic process of categorizations arising out of immediate experience but enduring beyond it: generalization, as Vygotsky calls it (“Word meaning is nothing other than a generalization,” in his famous phrase (Vygotsky, 1987 p. 244)) or typification as Schutz terms it. Typifications are the habitual, routinized, socially shared, intersubjective categorizations that are at the heart of social- psychological stability—and the basis of our recognition of contexts and mastery of them, our learning (Schutz & Luckmann, 1973). Similarly, it is the construal of the atypical that gives rise to change. It is from Schutz’s phenomenological understanding of typification that much North American writing research has taken its concept of genre.
Genre as social action meets activity theory From the mid-1980s, North American writing research has developed the concept of genre as social action in order to analyze the role of documents
Writing in multiple contexts 357 (and artifacts in various media) in organizational change and learning. The concept of genre as social action originated not with Bakhtin’s (1987) notion of genre (though this has proved very influential) but with Schutz’s (1973) concept of typification. Carolyn Miller (1984, 1994) introduced the concept of genre as “typified rhetorical actions based in recurrent situations” (1994, p. 31). Genre is not seen as similar formal features or as packeted speech (Wertsch, 1994), but as typified actions that over time have been routinized, “stabilized-for-now” (in Schryer’s phrase, 1993) in ways that have proven useful in some recurring situation—that is, in some context recognized (interpreted) as similar, as typical, by participants. This phenomenology of genre is deeply compatible with Vygotsky’s view of mediated action. Put simply, a genre is the ongoing use of certain material tools (marks, in the case of written genres) in certain ways that people recognize as having worked once and might work again, a typified, tool-mediated response to conditions recognized by participants to be recurring. Discursive actions are not seen, in Bakhtin’s metaphor, as voices ventriloquted from and contributing to social languages, but rather as motivated actions in practical activity (see Bazerman, 2004, for the relation to speech act theory). Miller’s (1984) seminal article “Genre as Social Action” emphasizes the situatedness of communication by conceiving of genre as “more than a formal entity” for classifying textual features (p. 153). Miller says, following Schutz, that situations are social constructs that are the result not of “perception,” but of “definition.” Because human action is based on and guided by meaning, not by immediate material causes, at the center of action is a process of phenomenological interpretation. Before we can act, we must interpret the indeterminate material environment; we define or “determine” a situation (p. 156). According to Miller, this determination is accomplished by an attribution of “types” we assign to situations in which we find ourselves. Genre is “typified rhetorical actions based in recurrent situations” (1984, p. 159) but “what recurs is not a material situation (a real, objective, factual event) but our construal of a type” (p. 157). Miller goes on to argue, again following Schutz, that as we gain more experience in particular domains, our stock of knowledge is usually enough to master most of the situations we “define” during our day-to-day life in the world. Thus, genres are more than categories of tools classified according to formal features. They are traditions of using a tool or tools, “forms of life, ways of being, frames for social action” (Bazerman 1994, p. 79). A genre conveys a world view—not explicitly, but by “developing concrete examples” that allow participants “to experience the world in the genre’s way” (Spinuzzi, 2003, p. 42). Genres allow subjects to recognize (in Schutz’s term determine) the activity and the appropriate actions in the presence of certain constellations of tools (marks on surfaces and other phenomena). And genres make it possible to act with others over time in more or less
358 D. R. Russell but never entirely predictable ways, individually and collectively, institutionally and culturally. In this phenomenological sense, genres are also central to object formation, transformation, and maintenance of activity systems. As Engeström says, “The object is an enduring, constantly reproduced purpose of a collective activity system that motivates and defines the horizon of possible goals and actions” (1999, p. 170). But the object of activity can be seen to attain its stability, reproduction, and continuity through genres, the mutual recognition (determination) necessary for joint action to occur over time. And when the object is contested (offering potential for change), it is against the landscape of existing genres, existing typifications. Genres are also deeply involved in the construction of motives. Genres are, in a sense, classifications of artifacts-plus-intentions. They enact social intentions, offering ways of using tools to accomplish collective activity. As Miller (1984) argues, following Schutz, “What we learn when we learn a genre is not just a pattern of forms or a means of achieving our own ends. We learn, more importantly, what ends we may have” (p. 165). A genre offers not only a landscape of possible action but also a horizon of potential motives or direction (Bazerman, Little, & Chavin 2003). In this sense, genre provides a way of including motives in the analysis of activity. As such, genres can be seen as crucial links between subjects, tools, and objects.
Polycontextuality: multiple activity systems and complex genre systems In complex activity systems, there are typically many written genres, typified, intersubjective understandings, which participants use together to structure (and change) their interactions within and among various contexts or activity systems (polycontextuality, in Engeström’s phrase). North American writing research has developed the concept of genre systems, following Bazerman (1994), or in Spunuzzi’s (2003) formulation, genre ecologies, to understand how genres (particularly written ones) work in and between complex organizations. Bazerman defines a genre system as “interrelated genres that interact with each other in specific settings” (1994). In a genre system, “only a limited range of genres may appropriately follow upon another,” because the conditions for successful coordinated action are conditioned—but never finally determined—by their history of previous actions (Bazerman, 1994, p. 80). Analyzing the genre systems allows us to see routine or typified interactions of reading and writing not only within but also among contexts. For example, IRS tax form 1040 is intertextually (now often hypertexually) linked to other documents in other genres and in other contexts (activity systems): in a taxpayer’s files, employers’ files, bank records, government regulations, tax laws, accounting standards, addresses, calendars, and so on, and to material property (real estate, factories, farms, etc.) and con-
Writing in multiple contexts 359 crete actions (buying, selling, renting, theft, gambling losses, etc.) that those documents in various genres represent. Analysis of genre systems or ecologies charts horizons into which the object has expanded already through existing genres, and the territory it may expand into. For literate organizations, the expansive reach of the object and the identities of the subjects involved (actual or potential) can be traced by following the written genres. Genre systems provide the skeleton of the structure of modern activity systems, made visible through genre systems analysis.
Example 1: WAC and the genre systems of (university) history There has been a major effort in the United States in the last 30 years called the Writing Across the Curriculum (WAC) movement (Russell, 2002). WAC encourages university departments and teachers to improve students’ writing in their disciplines and to use writing to support learning in their disciplines, rather than use writing only as a tool of assessment. Although university administrators and teachers have seen student writing as a “problem” and had favorable attitudes toward improving students writing, WAC has encountered many obstacles and hesitations in implementation, beyond the obvious ones: lack of time and money. Yanez and I (2003) studied a third-year Irish education history course in a large Midwestern public university (MWU), that students in fields other than history took to satisfy a university general education (GE) requirement (common in US universities to broaden students’ education). We wanted to understand obstacles to WAC (and the deeper attitudes, practices, and structures involved) in multiple contexts: the classroom, the broader university, and professional and civic contexts beyond it. We first did an activity/genre systems analysis of the course, to construe the typifications (and thus genres) perceived by the teacher and students, drawing on classroom observations, student and teacher interviews, and documents. We found the assignment genres (book report, research paper) were defined very differently by the teacher and the students, which produced frustration in the students and tensions and disturbances in the classroom. By broadening our analysis to other activity systems (professional academic history, secondary school history teaching, and journalism) we found the tensions were symptoms of deeper contradictions between the students’ and teachers’ construction of the object and motive of the course. The teacher perceived the assignments as genres of professional academic history useful for deepening students critical thinking and making them more critically aware citizens. The students perceived them as linked to the activity system of secondary school history or popular history for leisure reading, and they did not perceive the genres as relevant to their diverse professional pathways or future citizenship. Students expressed their sense of “just doing it for a
360 D. R. Russell grade” rather than for their future involvements, and they seemed alienated by the writing tasks. However, our analysis of the Irish History course suggests this alienation was overcome when one student, with the help of her instructor, saw the textual pathways (genre systems) of academic history were linked to the genre system of the field she intended to enter, journalism. Comments by the instructor—a graduate student—about the tensions he felt in using writing led us to gather interview data and curricular/policy documents from the department and the university written over the last 50 years, in order to extend our analysis of the obstacles to WAC to the broader institutional and cultural levels, and across wider time scales. We traced the intertextual and intersubjective links between the classroom and the institution to identify deeper contradictions underlying those tensions teacher and students felt. We were thus able to connect the micro-level classroom and faculty interview data with larger patterns institutionally (the university and the professions of history and journalism) and ideologically (Yañez & Russell, in press). Our analysis suggests that writing at MWU is marked by strategic ambiguity. When convenient, writing is conceived in terms of unproblematic transmission: a container or conduit for thought. “Content” is placed into written “form” and sent. Writing is a generalizable set of discrete skills necessary for critical thinking and democratic life. Students do not have to understand the relationship between the practices of academic history and their own pursuits. Citizenship is not a social practice into which one is enculturated but an accumulation of knowledge and skills taught to the masses—an ideology of mass education. But when convenient, writing is alternatively conceived as a tool of enculturation in some specific social practice, such as the activity of doing professional academic history—an ideology of elite, meritocratic education. This strategic ambiguity allowed MWU to pursue contradictory motives in general education without confronting their consequences at the human level of teaching and learning. The strategic ambiguity made it possible for faculty and administrators to alternatively invoke one and ignore the other of these two official motives when necessary or convenient in working out the division of labor. Graduate students, for example, taught the general education courses, freeing tenure-line faculty for teaching majors and doing research. Not having to examine the relation between GE courses and students in terms of the writing (and share clear goals and expectations for the GE mission of the department) allowed administrators and faculty much more flexibility in apportioning faculty and teaching assistants time for teaching and research. And the contradiction in motives allowed writing to realize both motives, at least in terms of faculty discourse and official documents. However, this strategic ambiguity over conceptions of general education and writing left the instructor, and his students from many disciplines (none in history), to wrestle with the consequences. Despite the instructor’s best attempts, the institutional and disciplinary contradictions operated so
Writing in multiple contexts 361 powerfully that it was difficult at best to use writing as a tool of learning. In this analysis, then, classroom contradictions are linked intertextually to genres and activities of departmental, university, and professional academic history and the wider institutional and ideological contradictions beyond.
Example 2: multimodal simulations of professional contexts In the next example of our research, we used the synthesis of AT and genre to construct a fictional context to represent, for the participants in one context (an engineering classroom) the activity systems and genre systems of another context: the engineering organizations toward which the students are headed. We researched and created multimodal simulations of professional contexts, using the affordances of the World Wide Web (Fisher, 2006, 2007; Fisher, Russell, Williams, & Fisher, 2008; MyCase, 2006). We are developing and researching these multimedia simulations using a content management system (CMS) to model the circulation of documents within or among fictional organizations, represented by fictional internet and intranet sites. Students in professional curricula (business administration, bio-systems engineering, and genetics, thus far) role play as they collaboratively engage in workplace-like activities using the sorts of tools and genres typical in workplaces (databases, files of documents, meeting minutes, videoed meetings, synchronous and asynchronous communication, etc.). This is radically different than genre pedagogies that either teach students the genres “in” the classroom (Swales, 1990; Martin, 2000) or those that send students to do “ethnographies” of genre in situ (Devitt, 2004; Johns, 2002). Students play the role of consultants to a fictional organization, such as Omega Molecular, a start-up biotechnology company used in the engineering simulation. The teacher plays the role of the engineering students’ boss, the lead consultant. They must produce texts in a range of genres—oral, written, visual, and electronic—which are submitted to characters in the simulation, such as the CEO pictured here in a video. And the characters reply to the students-as-consultants through email (though it is actually— as the students are told—the teacher who is replying, using a special role- sensitive email system). The simulation also contains a universe of documents placed in a company document server, arranged in various departments, and linked intertextually, so that the students can (re)construct the history of the organization, its problems, its directions, its crimes, even (which we have “seeded” into the simulation). And in the actions of characters and students-as-consultants, that document universe is brought into circulation through the genre system, where students must act on deadlines, face ethical dilemmas “seeded” into the simulation, and deal with the dialectical contradictions among motives, tools, rules, and objects that we constructed.
362 D. R. Russell To create the simulations, we used something like Schutz’s sociological research method. We conducted interviews, videotaping, and participant observation at similar real organizations, then constructed what Schutz, following Weber (Hekman, 1983), calls an “ideal type” of such engineering organizations—a generalized summary of the typifications of participants, checked against their understandings. Schutz refers to these ideal types as “puppets” created by the social scientist (1962, p. 41). And it is these “puppets” that the students are manipulating, with the goal of constructing for themselves the typifications, the genres, of the target professional activity system as they write. But they do so not in the lifeworld (the classroom context) but in a play world, what Schutz (1962) calls an alternative reality. Our research into students’ learning in this environment (using observation, surveys, focus groups, hit counts, and textual analysis of student work) suggests that they are much more likely to attribute their learning in the online simulation environment to contexts of professional work than to contexts of schooling, as compared to their attributions of other parts of their courses that use more traditional learning environments (e.g., WebCT and face-to-face instruction) (Fisher, 2006). These attributions seem to be shaped by the changes in classroom rules, division of labor, and community that the simulation affords, and by the contradictions between the activity systems of schooling and workplace (mediated by the simulation as teaching tool). For example, in the engineering and business simulations, students draw freely from each other’s work as it is posted to a shared file space, and from the student work published in the simulation (students add to the simulation over time). This literacy practice is extremely atypical in classroom settings, but is extremely typical in the workplace, where people often draw from a common pool of documents and where documents cycle through multiple readers in the division of labor.
Conclusion I have outlined here a way of theorizing multiple contexts that synthesizes activity theory and a phenomenological approach to genre theory. It tries, like several other theories, to incorporate both the phenomenological first- person point of view, whether first-person singular or first-person plural, and generalizations that reach beyond that—though still rooted in the subjectivity (or rather intersubjectivity) of participants studied. Schutz’s sociological phenomenology was crucial to the development of two central methodologies for contemporary writing research: Garfinkle’s ethnomethodology, and conversation analysis—as well as to linguistic anthropology. Yet it is important to return to Schutz’s sociological phenomenology as more than an interesting antecedent. These successors of Schutz focus on examining micro-level interactions, and the research of Prior (1998) and his group (2006) suggests how useful this can be to writing research in their analysis of the lamination of contexts. But here I am returning to Schutz’s phenomenology and methodology to quite explicitly make meso-level (institutional) and macro-level (ideo-
Writing in multiple contexts 363 logical) generalizations, as he did and as many activity theorists do (e.g., Engeström, 1987, 2001). Despite the limitations of this kind of generalization (Garfinkle, 2002) of puppet-making, if you will, we have found it useful in understanding writing across contexts, and in creating environments for researching and teaching writing in use.
References Bakhtin, M. M. (1987). Speech genres and other late essays. Austin, TX: University of Texas Press (second paperback edition). Bazerman, C. (1994). Systems of genres and the enactment of social intentions. In A. Freedman & P. Medway (Eds.), Genre and the new rhetoric (pp. 79–101). London: Taylor & Francis. Bazerman, C. (2004). Speech acts, genres, and activity systems: How texts organize activity and people. In C. Bazerman & P. Prior (Eds.), What writing does and how it does it: An introduction to analyzing texts and textual practices (pp. 309– 339). Mahwah, NJ: Erlbaum. Bazerman, C., Little, J., & Chavin, T. (2003). The production of information for genred activity spaces. Written Communication, 20(4), 455–477. Bazerman, C., & Russell, D. R. (Eds.). (2003). Writing selves/writing societies: Research from activity perspectives. Fort Collins, CO: The WAC Clearinghouse. Cole, M. (1996). Cultural psychology: A once and future discipline. Cambridge, MA: Belknap. Devitt, A. (2004). Writing genre. Carbondale, IL: Southern Illinois University Press. Dias, P., Freedman, A., Medway, P., & Paré, A. (1999). Worlds apart: Acting and writing in academic and workplace contexts. Mahwah, NJ: Erlbaum. Engeström, Y. (1987). Learning by expanding: An activity-theoretical approach to developmental research. Helsinki: Orienta-Konsultit. Engeström, Y. (1999). Communication, discourse and activity. The Communication Review, 3, 165–185. Engeström, Y. (2001). Expansive learning at work: Toward an activity theoretical conceptualization. Journal of Education and Work, 14(1), 133–156. Engeström, Y., Engeström, R., & Kärkkäinen, M. (1995). Polycontextuality and boundary crossing in expert cognition: Learning and problem solving in complex work activities. Learning and Instruction, 5, 319–336. Fisher, D. (2006). Remediating the professional classroom: The new rhetoric of teaching and learning. Unpublished doctoral dissertation, Iowa State University, Ames. Fisher, D. (2007). CMS-based simulations in the writing classroom: Evoking genre through game play. Computers and Composition, 24(2), 179–197. Fisher, D., Russell, D., Williams, J., & Fisher, D. (2008). Space, time, and transfer in virtual case environments (VCEs). Kairos: Rhetoric, Technology, Pedagogy, 12(2). http://kairos.technorhetoric.net/12.2/binder.html?topoi/fisher-etal/article Intro.html. Garfinkle, H. (2002). Ethnomethodology’s program. New York: Rowman and Littlefield. Hekman, S. (1983). Weber, the ideal type, and contemporary social theory. Notre Dame, IN: University of Notre Dame Press.
364 D. R. Russell Johns, A., (Ed.). (2002). Genre in the classroom: Multiple perspectives. Mahwah, NJ: Erlbaum. Martin, J. R. (2000). Design and practice: Enacting functional linguistics. Annual Review of Applied Linguistics, 20, 116–126. Miller, C. R. (1984). Genre as social action. Quarterly Journal of Speech, 70, 151–167. Miller, C. R. (1994). Rhetorical community: The cultural basis of genre. In A. Freedman & P. Medway (Eds.), Genre and the new rhetoric (pp. 67–78). London: Taylor & Francis. MyCase: Across departments, across disciplines. (2006, September 24). Retrieved Jan. 30, 2008, from http://mycase.engl.iastate.edu. Prior, P. (1998). Writing/disciplinarity: A sociohistoric account of literate activity in the Academy. Mahwah, NJ: Erlbaum. Prior, P. (2007). From Bakhtin to mediated multi-modal genre systems. Tubarao, Brazil: SIGET, Retrieved from www3.unisul.br/paginas/ensino/pos/linguagem/cd/ English/26i.pdf. Prior, P., Hengst, J., Roozen, K., & Shipka, J. (2006). “I’ll be the sun:” From reported speech to semiotic remediation practices. Text and Talk, 26, 733–766. Retrieved from www.atypon-link.com/WDG/doi/pdf/10.1515/TEXT.2006.030?cookieSet=1. Russell, D. R. (1997a). Rethinking genre in school and society: An activity theory analysis. Written Communication, 14(4), 504–554. Russell, D. R. (1997b). Writing and genre in higher education and workplaces: A review of studies that use cultural-historical activity theory. Mind, Culture, and Activity, 4(4), 224–237. Russell, D. R. (2002). Writing in the academic disciplines: A curricular history. Carbondale, IL: Southern Illinois University Press. Russell, D. R., & Bazerman, C. (1997). The activity of writing; the writing of activity. Special issue of Mind, Culture, Activity, 4(4). Russell, D. R., & Yañez, A. (2003). “Big picture people rarely become historians”: Genre systems and the contradictions of general education. In C. Bazerman & D. R. Russell (Eds.), Writing selves/writing societies: Research from activity perspectives (pp. 331–362). Fort Collins, CO: The WAC Clearinghouse and Mind, Culture, and Activity. Schryer, C. F. (1993). Records as genre. Written Communication, 10(2), 200–234. Schutz, A. (1962). Collected papers, vol. I. The problem of social reality. The Hague: Martinus Nijhoff. Schutz, A. (1967). The phenomenology of the social world. C. Walsh and F. Lehnert, trans. Evanston, IL: Northwestern University Press. Schutz, A., & Luckmann, T. (1973). The structures of the life-world. Evanston, IL: Northwestern University Press. Swales, J. M. (1990). Genre analysis. Cambridge: Cambridge University Press. Spinuzzi, C. (2003). Tracing genres through organizations: A sociocultural approach to information design. Cambridge, MA: The MIT Press. Vygotsky, L. (1987). Thinking and speech. N. Minick, Trans. In R. W. Rieber & A. S. Carton (Eds.), The collected works of L. S. Vygotsky: Problems of general psychology: Vol. 1 (pp. 39–285). New York: Plenum. Wertsch, J. V. (1994). Voices of the mind: A sociocultural approach to mediated action. London: Harvester Wheatsheaf. Yañez, A., & Russell, D. R. (in press). The world is too messy: The challenge of historical literacy in a general-education course. In J. Castner (Ed.), Teaching writing in the liberal arts. Creskill, NJ: Hampton Press.
27 The contributions of North American longitudinal studies of writing in higher education to our understanding of writing development Paul Rogers George Mason University
For many decades researchers have investigated the development of writing abilities (Britton, 1975; Applebee, 2000). While this research has included attention to children’s acquisition of print literacy prior to schooling as well as writing in the workplace and the professions, a good deal of this research interest has been sustained by the need for an evidence-based foundation for the teaching and learning of writing in schools. Longitudinal studies in particular with their emphasis on change over time and across contexts have proven a particularly appropriate method in understanding writing development. As Emig (1971, p. 95) noted in her pioneering work Longitudinal case studies of a given sample of students, following them from the time they begin to write in the earliest elementary grades throughout their school careers, up to and including graduate school . . . would make better known the developmental dimensions of the writing process, both for the individual and for members of various chronological and ability age groups. To date, no studies of the scope for which Emig called have been conducted; yet, in North America a number of longitudinal studies of writing have been conducted, particularly in higher education settings. Methodologically longitudinal approaches have proven to have incorporated four out of the six levels of inquiry in research on the composing process proposed by Bereiter and Scardamalia (1987, p. 34): reflective inquiry, empirical variable testing, text analysis, and process description. Rather than comparing cohorts to control groups, as in theory-embedded experimentation, findings for these studies were derived by examining changes in the performances, attitudes, beliefs, and values of the participants. Moreover, they were based on viewing events and their impact on the cohorts or participants as they occurred over time, and did not include experimental interventions.
366 P. Rogers For a variety of reasons these studies do not, however, allow for direct causal arguments: the high number of variables bearing upon different units of analysis; the small incremental nature of change; the complex interrelationships between observed phenomena; and the near impossibility of determining the extent to which an individual variable contributed to an overall perceived change. However, the studies reviewed here have added much to our understanding of writing development; in particular, they have contributed to the generation of valuable hypotheses concerning how people learn to write and the impact of writing on learning, and they have provided a rich description of the multiple variables that contribute to the development of writing abilities. The earliest longitudinal studies of writing focused on children in their early years of schooling and the development of print literacy (Tierney & Sheehy, 2005). Loban (1963), for example, undertook a longitudinal study of 338 urban students during the entire course of their K–12 schooling showing that socioeconomic factors played a significant role in the development of their reading, speech, and writing. Rental and King (1981, 1983) followed 36 children stratified by sex, socioeconomic status, dialect, and school at intervals of four months over the children’s first four years of schooling. Primarily using an analysis of student narratives, this study looked at the transition to writing from earlier forms of literacy. The scarcity of such studies attests to the many challenges of longitudinal research, including high attrition rates, the need for significant human and financial resources to cope with logistics and large quantities of data, and the desire for quick results by funding sources. In spite of these difficulties and because of the urgent need for evidence regarding the need for and efficacy of writing instruction in higher education, a number of researchers have conducted longitudinal studies of writing in North American colleges and universities. In the rest of this chapter I summarize the developmental findings of these studies, looking first at single-subject longitudinal case studies, and then the multi-subject longitudinal studies. The earliest longitudinal case study of a college writer was Lucille McCarthy’s (1987) dissertation study. Conducted at a private liberal arts university between 1983 and 1985, McCarthy used classroom observations, informal discussion, textual analysis, think-aloud protocols, as well as retrospective and semi-structured interviews, with teachers and her subject to investigate how the student came to understand the requirements of different disciplines and teachers, and how the student produced the writing to meet those demands. McCarthy found a great similarity among assigned writing tasks; yet, the students attentional focus was different on each task, and he was unable to identify the connections among the tasks, i.e., the student interpreted each writing situation as being totally different from other writing tasks. McCarthy’s research highlighted the dominant role the immediate environment of the classroom played in the student’s learning, and described other factors which influenced the student’s writing success: the role the writing served for the student personally, the impor-
Longitudinal studies and writing development 367 tance of teacher–student relations including differences and similarities in ethnic and class backgrounds, and the learning that resulted from the student’s interactions with peers and their writing. Haas (1994) described the rhetorical development of one biology major whom she followed between 1986 and 1990 at a private research university. Through extended, minimally structured, and discourse-based interviews, classroom observations, and think-aloud protocols, Haas sought to understand “if students’ views of academic discourse change over the course of their college careers” (p. 46), and more specifically if they recognized “the rhetorical nature of scientific action and scientific texts” (p. 50). Haas found in her subject a growing awareness of the rhetorical frame supporting written discourse, as the student began to see “behind scientific texts are human authors with motives” (p. 32). Haas claimed it was not the explicit introduction of theories of discourse which led to the development of her subject’s growing awareness of the social dimensions of scientific writing; rather, the student developed new ideas concerning the nature of disciplinary discourse through interaction and activity. In particular, work with scientists in the lab helped her to see more clearly the human agency involved in text production and the social context in which texts operated. Changes to her perception of academic discourse influenced the way she approached her writing, i.e., she moved from merely reporting what her textbooks said to understanding that authors are having a conversation. Haas’s research suggests that in order to understand the development of student writing abilities researchers must investigate student’s pre-existing conceptions of texts at the beginning of their college career, observe ways in which these conceptions change, and note the variety of factors influencing their change, especially in relationship to a rhetorical or social understanding of the ways texts operate within specific communities of practice. Undertaken at a major private research university between 1990 and 1993, Beaufort’s (2004, 2007) study was not concerned with the general development of writing abilities; instead, it attempted to describe the development of one college student in learning to write history in order to further a model of disciplinary writing expertise. Using interviews, textual analysis, and consultation with disciplinary experts on history writing, Beaufort aimed to understand “how expertise in history writing might be characterized; what developmental changes occurred that were related to writing; and what factors contributed to the development of the subject’s history writing skills” (2004, p. 137). The results of her study show the complex, simultaneous, interweaving of multiple factors in the development of discipline-specific writing expertise, and how a writer may develop in one area but not another. Beaufort argues that mature members of the community of professional historians choose appropriate genres for appropriate purposes, but in the student’s case the genres and purposes for the writing in his history classes were unclear: the subject described assignments as either the regurgitation of content or general explorations of
368 P. Rogers broad subject matter. Hence, his texts approximated features of several genres, and often exhibited different purposes within the same document (2004, p. 162). However, genre awareness gains did include a movement toward the formal register of history writing, and some of his texts approximated aspects of authentic history writing. The student’s history writing did improve across his freshman, sophomore, and junior years: the student made gains in subject-matter knowledge including “some understanding of historical themes, but his key interpretive skills were still weak” (2004, p. 154). The subject’s critical-thinking skills (what Beaufort refers to as the procedural aspect of subject-matter knowledge) were not consistently applied. While the student’s purpose for writing changed slightly from writing for self and grades toward writing for the discipline of history, “The classroom discourse community took precedence and was, in fact, the only social context the student was aware of, rather than any sense of the larger disciplinary discourse community” (2004, p. 150). Beaufort reports that the subject was unable to make the kinds of writing-related rhetorical choices that would be considered an authentic part of the discourse of historians. There were, however, signs that the student was beginning to understand what historians do when they write, and that he was to some degree attempting to take on features of that identity: he was beginning to understand how historians analyze historical texts and through these texts take on the analysis of broader social events. However, what the student learned about history writing seemed to happen inferentially rather than through explicit instruction. The only longitudinal study dealing explicitly with second-language acquisition, Spack (1997) reported on a student she followed for two years using classroom observations, textual analysis, interviews, and informal conversations. She wanted to understand how her subject(s) drew on multiple resources to succeed as readers and writers in a university setting, and what features of the subject’s linguistic, educational, and cultural background constrained and enabled her to manage coursework. Spack found that the student matured as a reader and writer as she received meaningful input from numerous classroom experiences; she linked the student’s writing development to constant practice, subject-matter immersion, and “talking about her projects with those who could share their expertise” (p. 47). In the beginning of the student’s experience, linguistic and cultural factors impacted the student’s writing performance; however, in Spack’s view the student’s lack of background knowledge and difficulty with academic argument in English posed the greatest challenges later on. As the student progressed through her course of study her literacy skills developed through active engagement in course activities, especially reading. While the student’s sense of self-efficacy as a writer grew over time, she continued to perceive her most difficult writing problem as overcoming a “Japanese style of writing” (p. 46). Although the student benefited significantly from the ESL courses she took in her first year the learning did not always transfer to future courses.
Longitudinal studies and writing development 369 Chiseri-Strater (1991) reported on two students she followed at the University of New Hampshire between 1987 and 1988—although she began following six. Her research interests included understanding “how students interpreted the literacy demands made on them and what it meant from the student’s point of view to be literate in a university” (p. 184). Drawing upon interviews with students and teachers, classroom observations, informal conversations, and textual analysis, Chiseri-Strater reported that texts, writing tasks, discipline specific knowledge and language provided support for the students as they moved toward greater levels of participation in academic life. Moreover, through class discussion, reading and writing groups, and in class modeling, students begin to see that their ideas were never generated totally in isolation, and that the reading and writing of texts served as a foundation from which writers make new knowledge. Furthermore, students came to a tacit understanding of intertextuality, that is their attention moved away from focusing on the lone writer and shifted instead to the “sources and social contexts from which discourse arises” (p. 21). However, this learning was fraught with difficulties because of the differences found between courses that even in the same discipline use writing and classroom discourse in very different ways (see also Thaiss & Zawacki, 2006). Herrington and Curtis (2000) investigated the development of four students between 1989 and 1993 at the University of Massachusetts, Amherst (the original cohort was 18 students). In their case studies they integrate interviews with the study participants and their teachers, textual analysis, and classroom observations, and describe, using chronological narratives, the students’ writing development and personal growth through their undergraduate years. The cases show positive change in the students writing across their college experience: “students writing became more developed, more coherent, more surely articulated; further, all were writing with more authority; their later writings conveyed more of a sense of personal assurance and purpose in communicating with readers” (p. 357). They also provide detailed descriptions of the way that writing for self- reflection, self-fashioning, and participation contributed to the students’ personal development, stating that “students’ personal histories, goals, and interests are closely implicated in their learning” (p. 124). Herrington and Curtis point to many influences on development beginning with students pre-existing abilities, cultural backgrounds, and prior writing experiences, both positive and negative, which contribute to shaping student perceptions. They posit that writing development is interrelated to and works in parallel with other forms of development taking place in students’ lives, in particular development in reading. Opportunities to revise, hard work and effort, plus student choice and interest in writing topics also contribute to the development of writing abilities. Emphasizing the way classroom contexts dominate student concerns, they illustrate how supportive classmates and their texts contributed to student growth, and provide examples of the significant role teachers played in
370 P. Rogers student writing development. For example, the ways that the presence or absence of “scaffolding that is specific to the writer” (2000, p. 93), clearly defined course and assignment expectations, and a responsive teacher attitude in relation to feedback can constrain or promote higher levels of literacy. Sternglass (1997) followed 53 study participants between 1989 and 1995 at the City University of New York (CUNY), through their initial composition courses into their upper division courses—in her published accounts she reported on nine. Using interviews, classroom observations, surveys, and textual analysis of student writing and course materials such as syllabi and instructor comments, she sought to understand “the ways in which the development of complex reasoning strategies were fostered by writing, the role that writing plays in learning, and how the multifaceted social factors in students lives affected their academic progress” (p. xiv). She also looked beyond students’ academic development and sought for “a full and rich picture of [student’s] development as whole persons” (p. xiv). While all of the students in her study developed as writers their growth was “neither neat nor linear” (p. xiv). The role of writing in their overall growth was very important for “only through writing did they achieve the insights that moved them to complex reasoning about the topic under consideration” (p. 295). Specifically, all of her subjects “were able to move beyond automatic acceptance of facts to recognize that new insights could be provided that might alter some unexamined assumptions in their field” (p. 295). Students also grew in their ability to create new knowledge for themselves through the integration of their own experience and academic discourse: “The worldviews of all students impinge on their attempts to make sense of the academic perspective and to integrate their experience into it” (p. 296). She claims that the relationship of college material to student’s experience, especially the issues of race, class, and gender on multicultural urban populations is significant. Sternglass argued that early instruction in composition was critical to fostering critical reading and writing skills, and was especially important for students with poor academic preparation who “have the potential to develop the critical reasoning processes that they must bring to bear in academic writing if they are given the time” (1997, p. 296). Her longitudinal perspective allowed her to frame composition as “only a single step in the progression of student writers” (p. 141). Qualifying the ways instruction influences writing development, Sternglass argued, “Writing instruction needs to be thought of from a long term perspective and issues of content and form should be pursued at every level, both in composition courses and in content level courses” (p. 160). Moreover, the types of tasks students are asked to perform both for writing and examinations, and institutional practices such as grading contribute greatly to creating “teaching facilitated learning” (p. 195), and “the enculturation of students into academic cultures” (p. 24). The nature of assignments, whether they are challenging or not, the amount of previous practice and experience students
Longitudinal studies and writing development 371 bring to an assignment, and the presence of new analytical demands were also important variables in student learning. Sternglass’s cautions that “personal life situations might interfere with the amount of time or concentration available to fulfill a task” (p. 3), reminding teachers that assignment design is only one factor in student writing performance. Sternglass’s analysis of teacher comments on student writings show how different kinds of comments, those focused on form and content, also played a role in creating a supportive environment for student learning. She also described how the student the degree of empathy between instructor and student can be an important source of support for student development; as students progressed through their schooling, they began to develop professional relationships with their instructors who frequently served as strong role models and mentors for them. Conducted at Pepperdine University between 1994 and 1998, Carroll (2002) followed 46 students over four years—with 20 staying through the duration of the study—to answer the question: “What do we actually know about how the writing of students develops over the course of several years of college?” Carroll’s data (portfolios of student writing, videotaped focus groups, textual analysis of student writing including instructor comments and grades, written self-assessments, interviews, and analysis of course materials) showed that students did develop new and complex forms of literacy over their four years of college through multiple interactions with teachers, peers, and texts. Carroll argues that a one- or two-semester first-year course in writing cannot meet all the needs of even our more experienced writers, for literacy skills develop slowly “as writers take on new roles” (p. xii). Carroll suggests that as students progressed through their undergraduate years they internalized a language and strategies for approaching new reading and writing tasks. For the students to accomplish the advanced writing tasks required in college they needed to grow in multiple dimensions rather than develop a general writing ability. Even students who came to Pepperdine feeling prepared and who were competent in writing certain kinds of argumentative and research writing were challenged by the greater analytical demands and presentation concerns of college writing; however, it was in working through these challenges that students developed as writers. Sommers (2002) conducted a longitudinal study of 422 students at Harvard University between 1997 and 2002. Drawing upon interviews, surveys, and textual analysis—including writing assignments, student papers, and teacher comments, Sommers and her colleagues set out to gain a comprehensive view of the role writing plays in the college years, to document students’ first-year writing experiences outside of their required expository writing courses, and the way students talk about writing changed throughout their undergraduate years. Although a complete report of the study has never been published, one finding to emerge was that students appeared to make the greatest gains in writing when they saw in writing a larger purpose than fulfilling an assignment (Sommers & Saltz,
372 P. Rogers 2004, p. 139). While Harvard students clearly valued writing their sense of interest needed to be sustained throughout their college career, by first accepting a novice role, then shedding it as they progress through their college careers. Sommers (2007) argues that feedback on writing plays an important role in undergraduate writing development “when, and only when, students and teachers create a partnership through feedback—a transaction in which teachers engage with their students by treating them as apprentice scholars and evolving writers, offering honest critique paired with instruction.” Sommers concludes that when students have been taken seriously as apprentice writers, and when instructors model the role of an attentive reader, such comments function to anchor students in their academic lives and, ultimately, make a vast difference in their college writing. In addition, Sommers argues that growth in writing knowledge and skills is not always immediately evident in student texts. In fact, gaps between what students know about writing and what they can perform can be observed throughout the undergraduate years. This sleeper effect between what students report understanding and what they can consistently perform presents additional challenges for writing researchers, as several studies of writing at the college level suggest that students regress as well as progress as writers throughout their college experience (see, for example, Haswell, 2000, p. 310). Haswell (2000) conducted a quantitative longitudinal study that used a direct comparison of writing samples written by the same 64 students during their first and third year of college. The goal of Haswell’s study was to “document normal longitudinal changes in the writing of students during the first five semesters of college, changes associated with the performance of competent post graduate workplace writers” (p. 324). This empirical study of texts employed a theoretical frame that held the work of professional writers as the direction toward which students develop (p. 337). Writing samples were drawn from each participant when they entered college and in the first semester of their junior year. In order to combat the attrition associated with longitudinal studies, Haswell drew his random sample after the students had completed the writing in their junior year. Analysis of the texts was conducted using a group of eight measures condensed and refined from Haswell’s earlier studies. No significant differences in the random sample in terms of pre-existing abilities (p. 330) or gender emerged; also, neither the order of the prompt nor the rhetorical frames from which students could answer showed any difference. Eight of the nine measures showed statically significant change for the group; still few of the students in the sample show “wholesale advance across all eight measures and some record hardly any change from freshman to junior year” (p. 333). In the same way the focus on individual cases in the other longitudinal studies reviewed here prohibits the generalization of effects, Haswell’s shows how conversely focusing on “group norms hides individual differences” (p. 333). While this study provides clear empirical evid-
Longitudinal studies and writing development 373 ence for a general advance in college writing abilities from the freshman to the junior year, many questions remain. Lunsford and her colleagues conducted a longitudinal study of writing between 2001 and 2006 at Stanford University (ssw.stanford.edu), which examined the writing practices and development of a random sample of 12 percent of Stanford’s incoming class of 2001 (n = 189), through their undergraduate years and into their first year beyond college. Using semi- structured interviews, textual analysis, and surveys, the Stanford Study of Writing was the first longitudinal study of writing to collect student writing through an online database and to pay systematic attention to students’ extracurricular writing. Additionally, the collection of data in the students’ first year post-graduation provided insight into the development of college students’ writing abilities as they entered graduate school or began professional careers, and helped to address issues of learning transfer across multiple contexts. The study had several major goals: “to provide an overview of student writing at Stanford; to trace student development in writing across a five- year period; and to use findings to inform writing instruction at the institutional level.” In addition, the study consciously sought to make useful contributions to longitudinal studies of writing and to the teaching of writing across the disciplines. Stanford students reported being deeply engaged with writing in and out of class; for these students, extracurricular writing was very often more important than in-class writing. Across the five years of the study, participants’ understanding of the scope and function of writing also changed considerably, from fairly instrumentalist definitions of writing to writing as a way of managing and making sense of enormous amounts of information and as a way of creating new knowledge. Rogers (2008) has reported on 40 of the Stanford Study of Writing participants, addressing the processes and products of their writing development through an analysis of interviews and a criterion referenced scoring of students’ academic writing across the five years of the study. Analysis of this data indicates that students developed through social interactions that took place at two levels: first, through interactions with the college curriculum itself; second, and more importantly, students grew through dialogic interactions with teachers and peers, that is, writing focused conversations (written or spoken) that were ongoing, invited active response, and which were addressed to the individual writer’s thinking and ideas. The actual mechanism of development was fine tuning in which writers adjusted, synchronized, and coordinated their writing with readers and teachers, as well as disciplinary and professional communities of practice. As students moved through the undergraduate curriculum and pursued disciplinary specializations their writing showed statistically significant growth in a number of important knowledge domains: rhetorical knowledge, genre knowledge, and domain-specific content knowledge, including critical thinking and argumentation. Moreover, student writers developed across
374 P. Rogers these multiple dimensions in a nonlinear fashion in the direction of increased, specialized, social participation. At the time of this writing, data analysis continues on the Stanford Study and is focused on issues of audience, agency, and collaboration.
Discussion While these studies demonstrate that writers develop across their college years in multiple ways, defining writing development remains a difficult task, as writing is a complex-cognitive and situated-social activity. Writing development therefore must always be seen as highly contextual. Moreover, theories and standards of development are cultural inventions rooted in changing social practices which are also in flux; therefore, tracing what counts as competence and development means invoking theoretical frames, not merely reporting empirical data. The studies reviewed here provide evidence that writing develops in multidimensional and nonlinear ways in higher education, supporting Kellogg’s (2006) model of development, which suggests that important aspects of writing development take place in and beyond the college years. This evidence directly challenges the idea that writing is a single general skill, which, like riding a bike, once learned is never lost (Russell, 2002). These studies also show that college students developed as writers through growth in multiple knowledge domains and by becoming enculturated into disciplinary practices (Rogers, 2008; Beaufort, 2004, 2007). As Haas (1994) noted, “at the college level to become literate is in many ways to learn the patterns of knowing about and behaving toward, texts within a disciplinary field” (p. 43). Thus, while growth in writing abilities is intimately connected to social interactions and is related to other forms of psychological and emotional change related to identity and self-efficacy, the bulk of the detectable changes exhibited by developing writers are arguably best viewed as movement toward greater levels of participation in particular communities of practice. Additionally, the longitudinal studies reviewed here show unequivocally that students develop as writers and people throughout their college experience through interactions with a variety of sociocultural inputs. These studies also provide a comprehensive view of the curricular and extracurricular factors that contribute to the development of writers and writing abilities (see Table 27.1). While non-classroom factors influence consider ably student writing development, of particular interest are those factors that relate to students’ classroom experiences; it is in the classroom where the students’ background and pre-existing abilities, teacher behaviors and talk, assignment design, writing tasks, peers, feedback, texts, and assessment practices influence students’ writing performance; moreover; these factors may be directly influenced by teachers and curriculum designers. The studies reviewed here attend differently to these variables, as some of the studies deliberately focused upon out-of-class variables within their
Longitudinal studies and writing development 375 Table 27.1 Factors contributing to the development of writing abilities Non-classroom-related factors
Classroom discourse
Students’ lives outside of the classroom or school context
What teachers say about writing in the classroom, including direct instruction
Psychological factors such as selfesteem, confidence, or anxiety
Peer to peer talk, reading and writing groups
Time – Natural development (growth, maturity, and development)
Whole-class discussion
Pre-existing abilities and writing experiences; cultural backgrounds, and gender
Conversation with teachers
Student engagement Institutional context, including assessment regimes Mentoring (in sociocultural settings) Teacher behaviors
Classroom genres
Teacher expectations
Teacher written response to writing, i.e., ongoing, performance-specific feedback
Responsive teacher attitude in relation to feedback
Model texts
Immediate rhetorical context, e.g., classroom and grades
Access to other student texts
Time to draft, revise, and reflect
Reading
Mentoring (by teachers)
General instructional supports: handouts, graphic organizers, assignments, and rubrics
Repeat performance opportunities, i.e. practice
Increased domain knowledge
Nature of tasks Teacher supportiveness, accessibility outside class
research design. Sternglass, for example, wanted policy makers to understand the ways in which outside financial pressures affected student performance; Chiseri-Strater and Herrington and Curtis wanted to show the intermingling of personal literacies with academic literacies, and make visible the complex interplay between the development of writing abilities and development of the whole person. These studies give more attention to non-schooling factors. On the other hand, Beaufort and Haas, whose interest was in student growth within a single discipline, focused almost entirely
376 P. Rogers on school-based factors. In this regard, the Lunsford study is the first longitudinal study to pay systematic attention to both school and extra curricular writing. These findings confirm Emig’s notion that “longitudinal case studies . . . make better known the developmental dimensions of the writing process” (1971, p. 95); furthermore, they suggest the importance of constructing rich learning environments where these variables are present in ways that promote rather than constrain student learning. Finally, these factors provide us with clear variables for future inquiry, where they can be tested for correlations, their relative importance in contributing to development, as well as the varying degrees in which individual elements contribute to the varying dimensions of writing development.
References Applebee, A. (2000). Alternative models of writing development. In R. Indrisano & J. Squire (Eds.), Writing: Research/Theory/Practice. Newark, DE: IRA. Beaufort, A. (2004). Developmental gains of a history major: A case for building a theory of disciplinary writing expertise. Research in the Teaching of English, 39(2), 136–185. Beaufort, A. (2007). College writing and beyond: A new framework for university writing instruction. Logan, UT: Utah State Press. Bereiter, C., & Scardamalia, M. (1987). The psychology of written composition. Hillsdale, NJ: Erlbaum. Britton, J., Burgess, T., Martin, N., McLeod, A., & Rosen, H. (1975). The development of writing abilities (11–18). London: Macmillan Education. Caroll, L. (2002). Rehearsing new roles. Carbondale, IL: Southern Illinois University Press. Chiseri-Strater, E. (1991). Academic literacies: The public and private discourse of university students. Portsmouth, NH: Heinemann. Emig, J. (1971). The composing process of twelfth graders. Urbana, IL: NCTE. Haas, C. (1994). Learning to read biology: One student’s rhetorical development in college. Written Communication, 11, 43–84. Haswell, R. H. (1991). Gaining ground in college writing: Tales of development and interpretation. Dallas, TX: Southern Methodist University Press. Haswell, R. H. (2000). Documenting improvement in college writing: A longitudinal approach. Written Communication, 17, 307–352. Herrington, J., & Curtis, M. (2000). Persons in process: Four stories of writing and personal development in College. Urbana, IL: NCTE. Kellogg, R. T. (2006). Professional writing expertise. In K. A. Ericsson, N. Charness, P. J. Feltovich, & R. R. Hoffman (Eds.), Expertise and expert performance. New York: Cambridge University Press. King, M., & Rentel, V. (1982). Transition to writing. Columbus, OH: Research Foundation. Loban, W. (1967). Language development: Kindergarten through grade twelve. NCTE Research Report No. 18, Urbana, IL: NCTE. McCarthy, L. (1987). A stranger in strange lands: A college student writing across the curriculum. Research in the Teaching of English, 21, 233–265.
Longitudinal studies and writing development 377 Rentel, V., & King, M. (1983). A longitudinal study of coherence in children’s written narrative. Columbus, OH: Research Foundation. Rogers, P. M. (2008). The development of writers and writing abilities: A longitudinal study across and beyond the college-span. Doctoral dissertation, University of California, Santa Barbara. Russell, D. R. (2002). Writing in the academic disciplines: A curricular history (2nd ed.). Carbondale, IL: Southern Illinois University Press. Sommers, N. (2002). The Harvard study of undergraduate writing. Retrieved April 20, 2006, from www.fas.harvard.edu/~expos/index.cgi?section=study. Sommers, N. (2007). Across the drafts. Unpublished manuscript. Sommers, N., & Saltz, L. (2004). The novice as expert: Writing the freshman year. College Composition and Communication, 56(1), 124–149. Spack, R. (1997). The acquisition of academic literacy in a second language: A longitudinal case study. Written Communication, 10, 235–261. Sternglass, M. S. (1997). Time to know them: A longitudinal study of writing and learning at the college level. Mahwah, NJ: Erlbaum. Tierney, R. J., & Sheehy, M. (2005). What longitudinal studies say about literacy development/What literacy development says about longitudinal studies. In J. Flood, D. Lapp, J. R. Squire, & J. M. Jensen (Eds.), Methods of research on teaching the English language arts. Mahwah, NJ: Erlbaum.
28 Statistical modeling of writing processes Daniel Perrin and Marc Wildi Zurich University of Applied Linguistics
1 Writing phases Phase models of the text-production process basically assume that different writing activities predominate during different time periods in the process. These time periods of activities can be recognized by more or less homogeneous time-series dynamics in the data. Most of the newer models describe the writing process as incremental, increasing at every level. Far- reaching decisions such as those concerning topic planning and local actions such as correcting a typographical error are possible at any time but not equally functional and therefore not equally probable at all times. Furthermore, phases can overlap each other and recur cyclically. Three types of phases of text production are strongly supported in the academic discourse on writing research and education: planning, formulation, and revision phases. They can be traced to earlier models, such as that of Hayes and Flower (1980). At that time, writing processes were investigated with simple tasks in experimental settings. However, text production in situ, for example as an activity in professional life, also includes understanding and determining the task as well as the interfaces for implementing the text product. Relevant research has been carried out by Keseling (1992) on planning in the pauses while writing and speaking; Hayes and Nash (1996) on planning while writing; Van der Geest (1996) on text planning from the perspective of professional writers; Wrobel (1997) on modeling formulation processes; Wrobel (2002) on “mental pretext,” a cognitively produced but not yet transcribed idea of a formulation; Baurmann and Ludwig (1985) and Rau (1994) for an overview of revision in the writing process; Sommers (1980) and Flower, Hayes, Carey, Schriver, and Stratman (1986) on revision strategies used by students; Van den Bergh, Rijlaarsdam, and Breetvelt (1994) on the relationship between the revision process and text quality; Severinson-Eklundh and Kollberg (2001) on revision patterns used by various authors; and Allal, Chanquoy, and Largy (2004) on revision in writing education. Writing undoubtedly is the central activity in the production of written texts. However, studies on text-production structure show that reading
Statistical modeling of writing processes 379 phases more or less systematically accompany the entire process of text development. The type and purpose of reading vary, contingent on the stage of text development and on the related text. Depending on the related text, two types of reading processes can be differentiated: production- oriented reading, to monitor the developing text product, and source- oriented reading, to understand texts being referred to that are written by other authors. Important research on reading processes in text production has been done by Bracewell and Frederiksen (1982); Ludwig (1983); Nelson Spivey (1990); Flower et al. (1990); Rau (1994); and Jakobs (1999). Writing and reading activities occur at all levels of text production, including (a) graphomotoric functioning, (b) formulation, (c) intermediate products, and (d) text versions. But the mode, frequency, and interplay of these activities depend on the progress in solving the emerging task. Therefore, the activities can be taken as indicators for writing phases. a
On a graphomotoric level, writers follow the appearance of characters on the recording surface and, in a scarcely conscious monitoring process of mental and visual feedback loops, check the graphomotoric realization of linguistic expressions. People anticipate and see what they write, just as they anticipate and listen to what they say, and can immediately detect and rectify motor errors and slips (cf. Berg, 2002; Alamargot, Dansac, Chesnet, & Fayol, 2007). Graphomotoric activities occur when writers are materializing their mental representations of what to write. b On a formulation level, writers formulate chunks of texts and re-read them to check them for grammatical, semantic, and logical coherence. If formulating becomes difficult, re-reading can encourage new ideas on how to continue (reactivation re-reading; cf. Rau, 1994) and help assess before writing them down whether draft formulations would fit with what has already been written (evaluative re-reading; cf. Rau, 1994). c On a text-parts level, writers read what they have already written in order to heighten their awareness of it before continuing (reading to comprehend vs. reading to evaluate; Hayes, Flower, Schriver, Stratman, & Carey, 1987) and then to correct discrepancies between the current and intended form (Molitor-Lübbert, 1984) and/or to encourage ideas for the next stages of work (Keseling, 1988). On this level, the interplay of writing and reading phases is especially important for bottom-up oriented writers—those writers who start writing with only vague ideas of the text purpose, content, and structure and then use the text-production process to clarify them (Molitor, 1987, p. 404). d On a text-version level, writers re-read the current version of the text they have just written in order to gain a new overall idea of the sense, meaning, and structure of it before they revise the text as a whole. This function of re-reading is important, for example, when an author
380 D. Perrin and M. Wildi wants to understand written comments from a co-author and incorporate them into the text (cf. Couzijn & Rijlaarsdam, 1996; Fischer, 2006). Activities on the text-version level stand for phases of planning and evaluation. Activities on levels (a) (graphomotorics) and (b) (formulation) are supposed to dominate formulation phases, whereas activities on levels (c) (text parts) and (d) (text version) are supposed to dominate revision and planning phases. Thus the writing (and reading) activities and their impact on production and product quality are expected to be phase-specific, but due to the lack of empirically sound models, phase concepts in writing research have been rather speculative until now. Figure 28.1 shows a heuristic model of the text production process and its interplay of phases, of reading and writing processes and activity levels as presented in Jakobs and Perrin (2008). The complex interplay of writing and reading phases that constitutes text-production processes has not yet been tracked in natural settings with large-scale samples to gain the data corpora needed for adequate and empirically sound theories of writing. Therefore, the theories, models and approaches presented so far have tended to be empirically vague: they are speculative or based on experiments or on single case studies. As a consequence, good practice models of writing lack empirically testable explanations of writing processes in general and phases in particular. We see this as a deficiency for systematic teaching and evaluating of writing (Wrobel, 2000, p. 468).
Figure 28.1 Key concepts concerning phases in text-production models.
Statistical modeling of writing processes 381
2 Progression analysis To track text production in natural settings, progression analysis was developed (Perrin, 2003, 2006a).1 Progression analysis is an ethnographic, computer-based multimethod approach with which data can be obtained on three levels: (a) the work situation, (b) writing movements, and (c) writing strategies. a
Before writing begins, details about the work situation are elicited with interviews and participatory observations. b During the writing process, progression analysis records every writing movement with a logging program that runs in the background behind the text editor. In the larger investigations with progression analysis, the logging programs run behind the text editors that the writers usually use, for instance behind the user interface of the text-editing systems. The logging follows the writing process over several workstations and does not influence the performance of the editing system. It records all the keystrokes and mouse movements as timed actions related to text entities and writer identification. c After the writing process is completed, progression analysis focuses on writing strategies: on the writers’ reinforced, conscious, and therefore articulable ideas of how decisions are to be made during the act of writing so that the writing process or text product has a great prob ability of taking on the intended form and fulfilling the intended function. This level of progression analysis is accessed with cue-based retrospective verbal protocols and is activated to validate the quantitative methods presented in the next sections. As outlined above (see section 1), a typical author goes through various writing phases during text production: phases dominated by specific practices such as goal setting, planning, formulating, and checking text, all merged with source reading and text reading. This interplay of practices and time organization, the “writing process,” can be explored through the dynamics of cursor movements as measured by a progression graph, which is one of the tools of level (b) of progression analysis. The progression graph represents cursor movements during text production. These cursor movements are interpreted as the author’s shifts of focus. The temporal sequence of revisions in the writing process is represented on the ordinal scale of the x-axis; the spatial sequence of revisions in the text product is on the y-axis, also ordinal (see Figure 28.2). A progression graph does not display information about the text content or absolute time. Instead, the focus is placed on the temporal dynamics of cursor movements: the data may be interpreted as a time series, where “time” is measured on an ordinal scale. The examples below show writing processes recorded at the Swiss National Bank as economists wrote short project reports for internal
382 D. Perrin and M. Wildi reviews. The first case is presented in Figures 28.2 to 28.7, the second in Figure 28.8 and the third in Figure 28.9. In all three cases, the writing time was about one hour and the text length up to one page. By abstracting from “real” time (absolute time scale) and from the particular text content (words and phrases), we can eliminate “nuisance” effects2 that might otherwise mask the most interesting writing phases. In analyzing large corpora of data from the field, we are not interested in the content of a particular text (the meaning of the words and phrases used) or in the particular text-producer (the author) but rather in the progression patterns as traces of writing phases. A writing phase is dominated by a particular task carried out by an author to accomplish text production (see section 1) and is formally characterized by a connected period of time during which series dynamics in the progression graph are “homogeneous” (in the sense that they share common dynamics). This definition will be refined gradually in the course of the following sections. It is important to note that the same writing phase can occur once, never, or several times in a particular graph. To illustrate these concepts, we can assume that a hypothetical author writes “linearly,” so the cursor movements correspond to a straight line with the negative slope of −1 in the progression graph. Such an author
Figure 28.2 Progression graph of a writing process.
Statistical modeling of writing processes 383 would have a perfect mental representation of the text (mental pretext, Wrobel, 2002) in his or her mind and the whole writing process would consist of one elementary writing phase (typing down letters in a pre- specified “deterministic” order). Typically, writers do not have such precise representations of text in mind. Therefore, they have to merge at least two types of practices: writing (and modifying) new text and reading parts of text produced so far (see section 1). That is why cursor movements are generally more erratic.
3 Cusum statistics The graph in Figure 28.2 indicates that more than 1,000 cursor movements were necessary in order to complete a particular text. A straight line with a slope of −1, i.e., the graph of a perfectly linear writing process, would indicate that an author writes the text from the first to the last letter without any corrections. It can be interesting to compare real writers to this construct (see the superimposed line in Figure 28.3). As a matter of fact, the deviations of a progression graph from a straight line can be very revealing. We call this an elementary de-trending procedure, whereby the theoretical trend corresponds to a particular “linear” progression graph (see Figure 28.4).
Figure 28.3 Progression graph compared with a perfectly linear writing process.
384 D. Perrin and M. Wildi
Figure 28.4 Elementary de-trending of progression graph (reference trend: perfectly linear writer).
Interestingly, these deviations seem to exhibit some “structure” which we can use to infer “writing phases.” In order to visualize this structure we rely on the well-known cusum statistic (cumulative sums)
t
cusumt = x k k=1
where xk are the deviations from the (linear) trend (see Figure 28.5). In Figure 28.5 the cusum statistic decreases more or less steadily from t1 to t350, meaning that the cumulated distances between a linear author and the real writer increase during that period. After the minimum is achieved (maximum distance) the real writer gets closer and closer (back on track) to the linear trend. The minimum cusum is chosen as a global structural break-point in the original progression graph. This break-point can be interpreted as a shift from primarily forward-oriented writing and planning to primarily backward-oriented text reading and revising (Jakobs & Perrin, 2008). If we decompose the time axis according to the previous break-point definition, as has been done in Figures 28.6 and 28.7, then we can see that the progression graph of the real writer lies systematically below (Figure 28.6) or above (Figure 28.7) the linear trend line.
Statistical modeling of writing processes 385
Figure 28.5 Cusum statistics with break-point at t350.
From a more “mechanical” point of view, the proposed break-point can be used to identify two fundamental writing phases: accelerating and decelerating writing progressions. Interesting statistics based on the above cusums are the time location of the minimum, the level of the minimum, and the asymmetry of the “cone” defined by the cusum statistics. Clearly, these features and their relations to writing phases remain to be explored more systematically.
4 Signal-extraction statistics The above-mentioned decomposition of time spans according to the proposed cusum break-point definition is very rough, and we suspect that many other time points could be indicative of changes in the underlying dynamics of the original progression graph. This is based on the assumption that changes in dynamics in the progression graph are associated with writing phases. In order to generalize the above procedure, it is therefore important to recognize that the cusum statistic can be interpreted as a linear filter. More precisely, the summation operator is a particular type of low-pass filter: it
386 D. Perrin and M. Wildi
Figure 28.6 Separation according to cusum break-point (t1–t350).
smoothes the input series by magnifying low-frequency and by weakening high-frequency “components” of the original time series. However, the design of the cusum filter is very rigid and its characteristics—in terms of amplitude and phase functions—may in some sense be “too extreme” when more subtle writing phases are of interest. Recognizing writing phases is a difficult task because series dynamics are contaminated by strong artifacts, which should be eliminated. In order to generalize from the crude cusum smoothing operator, we will need a more flexible design—we intend to extract various kinds of “signals”—and that requires the proper treatment of the particular dynamics of progression graphs, which often exhibit large “bursts.” Based on our experience in statistical signal extraction, we have chosen a state space approach in order to decompose the original series into interesting “features.” For illustration purposes, the following model is used for our computations: Ξt = FΞt–1 + Et x = (1, 1) Ξt + εt
(0.1)
Statistical modeling of writing processes 387
Figure 28.7 Separation according to cusum break-point (t351–t1,000).
where
1 1 F = 0 1
υt1 Et = υt2
Q1 0 Var(Et) = , Var(εt) = R 0 Q2
According to this definition, xt (the original progression time series) can be decomposed into an unobserved trend (level and slope parameters in Ξt) and an unobserved noise process εt, whereby the level and the slope of the trend are modeled as stochastic (integrated) processes. The ratio Qi/R, i = 1,2, determine the adaptivity of the trend.3 The unobserved trend level and slope can be estimated by relying on the Kalman filter, using appropriate initial values to start the recursions. One drawback of this procedure is its sensitivity to “outliers” (bursts) which can be overcome by a simple device introduced in Wildi (2008).4
388 D. Perrin and M. Wildi
To summarize, the proposed method has the following advantages:
• • •
It is adaptive. The transfer function of the resulting low-pass filter is controlled. It can be robustified in order to account for the unusual time-series structure.
The following figures (Figures 28.8 and 28.9, based on two other writing processes) illustrate the use of trend slopes in progression graphs (original series and levels are down-scaled to unit variance). Let us consider Figure 28.85 first and recall that the mean slope should be approximately −1. The emboldened line corresponds to the estimated slope. In this writing process, we can detect a phase lasting from t1 to approximately t350 with larger trend slopes, a phase from ~t350 to ~t580 with smaller trend slopes followed by larger ones and a period from ~t580 till the end of the time span with slopes near −1. We call the first phase “oscillation,” the second-one “flip-flop” and the third one “top-to-bottom.” •
In the oscillation phase the author jumps backward and forward, even in the middle of linguistic units, to add fragments to a growing, mostly incoherent text.
Figure 28.8 Trend level (bold) and slope (extra bold), series 1.
Statistical modeling of writing processes 389 • •
In the flip-flop phase the author writes whole paragraphs or parts of paragraphs at one go but then rearranges paragraphs or jumps back to insert longer sections in previously written text. In the top-to-bottom phase, the author formulates the text in the reading order, mostly without jumping back and forth and with a clear scheme in mind and a feeling of writing flow.
Close examination of another writing process (Figure 28.9) reveals a flip- flop phase (from ~t70 to ~t230) and a top-to-bottom phase at the end (from ~t290 to the end) where the trend slope is close to −1. In between we observe a new dynamic, called a reversed-trend phase (~t230 to ~t290), which is characterized by large increasing trend slope values when the author moves step-by-step from the bottom to the top in working through the text. As can be seen from the trend-slope parameter, a weaker expression of such a phase occurs at the beginning (~t46 to ~t57): this phenomenon can frequently be observed throughout the writing-data set. Figure 28.9 also suggests that there is room for improvement, particularly in phases characterized by strong level shifts (such as a flip-flop phase) which should be captured by suitable model features (intervention vari ables) rather than simply being “smoothed out.” Also, the effect of the preliminary robustification can be appreciated in the top-to-bottom phase from t290 to t465, where large bursts occur.
Figure 28.9 Trend level (bold) and slope (extra bold), series 2.
390 D. Perrin and M. Wildi Both the identification and description of these phases are new in writing research. We assume that we can prove the concept of regularities in writing processes with this approach. Regularities have often been presupposed but the assumption has yet to be well grounded with empirical data. Moreover, initial findings suggest that previous phase models proposed in writing research (see Jacobs & Perrin, 2008) have to be reconsidered. In this research framework, signal extraction (trend level/slope) seems to be a promising means to improve and refine the detection and visualization of writing phases.
5 The future It is apparent that our preliminary analysis based on cusum as well as the proposed (very simple) state space statistics is very rough. For example, we have focused exclusively on low-pass filters. We conjecture that suitably designed band-pass filters will deliver more refined information about the existence and the properties of writing phases. The above-mentioned simple filters are asymmetric and as such subject to well-known aberrations such as non-vanishing phases and a lack of amplitude functions. It would be interesting to consider the Kalman smoother in order to mitigate some of these methodological limitations. In any case, the asymmetric Kalman filter is very useful at the end of the time series, where important writing phases corresponding to the completion of the writing process can be observed. As can be seen in the above figures, filter outputs are still subject to bursts. Therefore we need a more robust smoothing device (robust signal extraction6). Moreover, the time series are often subject to local-level shifts. Preliminary results suggest that additional structures can be discovered by accounting for these irregularities. More precisely, structures have been detected in series linearized by intervention variables. For that purpose we intend to augment the state space model with intervention variables in order to establish a methodologically consistent framework. We have not yet analyzed the short-term dependency structure of the time series. This may be achieved by augmenting the state space model by a component that represents the stationary (mid-term) deviations about the trend. By allowing for time dependency of the corresponding parameters, we conjecture that more subtle variations in the dependency structure of the “linearized” time series can be analyzed.7 To summarize, we have presented some simple methods to identify accelerating and decelerating phases in large corpora of progression data from the field. This information processing is a first step toward a systematic identification and interpretation of writing phases. However, the results mentioned above are primarily descriptive in nature and lack statistical significance (established by tests, confidence intervals, etc.). Therefore, statistical refinement of the method will be the next step, combined with deriving new distance measures (between progression
Statistical modeling of writing processes 391 graphs) in order to characterize and compare various authors and to measure “performances” (e.g., before and after a dedicated writing course). From an applied perspective, this knowledge can be used to improve or compare writing performances of individuals (adults or children) and to (re-)design writing courses. Scientific approaches to writing might soon base their phase concepts and phase descriptions not only on introspection, experiments, and single case studies but also on statistical modeling based on large corpora of data from various fields.
Notes 1 The primary data for three representative corpora of educational, academic, and professional writing have been and are being generated with progression analysis in a non-intrusive and therefore ecologically valid way in the writing fields: Corpus I includes all of the writing processes produced for the Swiss newspaper Tages-Anzeiger by all 190 journalists between the years 1999 and 2002 (Perrin, 2006b). Corpus II includes the writing processes produced in the myMoment research project by 206 first- to fifth-graders between 2006 and 2009 (Gnach, Wiesner, Bertschi-Kaufmann, & Perrin, 2007; Perrin, 2007). Corpus III will include the writing processes produced by approximately 120 undergraduate students and 50 professional translators in undergraduate and postgraduate professional development courses between 2007 and 2009 (Ehrensberger & Perrin, in preparation). 2 Typically, text production is subject to perturbations (phone calls, travel, idle periods, etc.) which contaminate absolute time. 3 We select these parameters according to the shape of the amplitude function of the resulting filter rather than by traditional forecasting performances. 4 The resulting algorithm won the NN3 and NN5 forecasting competitions; see www.neural-forecasting-competition.com. 5 The first few observations have been excluded because of (still unresolved) initialization issues concerning the iterations in the Kalman-filter equations. 6 Extensive expertise in this domain was acquired in the context of various forecasting projects (mainly in the health-care sector). In particular, generalizations of the least-squares paradigm seem feasible with suitably operationalized robust MM-designs. 7 It is possible to generalize the state space model to a system with time-dependent level and autoregressive parameters in order to track these changes.
References Alamargot, D., Dansac, C., Chesnet, D., & Fayol, M. (2007). Parallel processing before and after pauses. A combined analysis of graphomotor and eye movements during procedural text production. In M. Torrance, L. Van Waes, & D. Galbraith (Eds.), Writing and cognition. Research and applications (pp. 13–29). Amsterdam: Elsevier. Allal, L., Chanquoy, L., & Largy, P. (Eds.). (2004). Revision: Cognitive and instructional processes (Vol. 13). Dordrecht: Kluwer Academic Publishers. Baurmann, J., & Ludwig, O. (1985). Texte überarbeiten. Zur Theorie und Praxis von Revisionen. In D. Boueke & N. Hopster (Eds.), Schreiben. Schreiben lernen (pp. 254–276). Tübingen: Narr.
392 D. Perrin and M. Wildi Berg, T. (2002). Slips of the typewriter key. Applied Psycholinguistics, 23(2), 185– 207. Bracewell, R. J., & Frederiksen, J. D. (1982). Cognitive processes in composing and comprehending discourse. Educational Psychologist, 17(3), 146–164. Couzijn, M., & Rijlaarsdam, G. (1996). Learning to write by reader observation and written feedback. In G. Rijlaarsdam, H. Van den Bergh, & M. Couzijn (Eds.), Effective teaching and learning of writing. Current trends in research (pp. 224–252). Amsterdam: Amsterdam University Press. Ehrensberger, M., & Perrin, D. (in preparation). Capturing translation processes: A multi-method approach. Zeitschrift Schreiben. Fischer, M. (2006). Mit strukturiertem Feedback zu qualitativ besseren Zeitungsartikeln. In O. Kruse, K. Berger, & M. Ulmi (Eds.), Prozessorientierte Schreibdidaktik. Schreibtraining für Schule, Studium und Beruf (pp. 271–278). Bern: Haupt. Flower, L. S., Hayes, J. R., Carey, L., Schriver, K. A., & Stratman, J. F. (1986). Detection, diagnosis, and the strategies of revision. College Composition and Communication, 1, 16–55. Flower, L. S., Stein, V., Ackerman, J., Kantz, M. J., McCormick, K., & Peck, W. C. (1990). Reading to write. Exploring a cognitive and social process. New York: Oxford University Press. Gnach, A., Wiesner, E., Bertschi-Kaufmann, A., & Perrin, D. (2007). Children’s writing processes when using computers. Insights based on combining analyses of product and process. Research in Comparative and International Education, 2(1), 13–28. Hayes, J. R., & Flower, L. S. (1980). Identifying the organisation of writing processes. In L. W. Gregg & E. R. Steinberg (Eds.), Cognitive processes in writing (pp. 3–30). Hillsdale, NJ: Erlbaum. Hayes, J. R., Flower, L. S., Schriver, K. A., Stratman, J. F., & Carey, L. (1987). Cognitive processes in revision. In S. Rosenberg (Ed.), Advances in applied psycholinguistics. Reading, writing, and language learning (vol. 2, pp. 176–240). Cambridge: Cambridge University Press. Hayes, J. R., & Nash, G. J. (1996). On the nature of planning in writing. In C. M. Levy & S. Ransdell (Eds.), The science of writing. Theories, methods, individual differences and applications (pp. 29–56). Mahwah, NJ: Erlbaum. Jakobs, E.-M. (1999). Textvernetzung in den Wissenschaften. Zitat und Verweis als Ergebnis rezeptiven, reproduktiven und produktiven Handelns. Tübingen: Niemeyer. Jakobs, E.-M., & Perrin, D. (Eds.). (2008) The Mouton-de Gruyter handbooks of applied linguistics: Communicative competence (vol. 1). New York: Mouton de Gruyter. Keseling, G. (1988). Textmuster und Klangstrukturen als Grundlage von Bewertungen beim Schreiben. In W. Brandt (Ed.), Sprache in Vergangenheit und Gegenwart (pp. 219–236). Marburg: Hitzerroth. Keseling, G. (1992). Pause and intonation contours in written and oral discourse. In D. Stein (Ed.), Cooperating with written texts. The pragmatics and comprehension of written texts (pp. 31–66). Berlin: Mouton de Gruyter. Ludwig, O. (1983). Der Schreibprozess. Die Vorstellungen der Pädagogen. In K. B. Günther & H. Günther (Eds.), Schrift, Schreiben, Schriftlichkeit. Arbeiten zur Struktur, Funktion und Entwicklung schriftlicher Sprache (pp. 191–210). Tübingen: Niemeyer.
Statistical modeling of writing processes 393 Molitor, S. (1987). Weiterentwicklung eines Textproduktionsmodells durch Fallstudien. Unterrichtswissenschaft, 4, 396–409. Molitor-Lübbert, S. (1984). Kognitive Prozesse beim Schreiben. Tübingen: Deutsches Institut für Fernstudien an der Universität. Nelson Spivey, N. (1990). Transforming texts. Constructive processes in reading and writing. Written Communication, 7(2), 256–287. Perrin, D. (2003). Progression analysis (PA): Investigating writing strategies at the workplace. Journal of Pragmatics, 35(6), 907–921. Perrin, D. (2006a). Progression analysis: An ethnographic, computer-based multi- method approach to investigate natural writing processes. In L. Van Waes, M. Leijten, & C. Neuwirth (Eds.), Writing and digital media (pp. 175–181). Amsterdam: Elsevier. Perrin, D. (2006b). Verstanden werden. Vom doppelten Interesse an theoriebasierter, praxisgerichteter Textberatung. In H. Blühdorn, E. Breindl, & U. H. Waßner (Eds.), Text: Verstehen. Grammatik und darüber hinaus (pp. 332–350). Berlin: de Gruyter. Perrin, D. (2007). Children’s writing processes when using computers: Insights based on combining analyses of product and process in the Swiss myMoment project. Paper presented at the NEAL. Rau, C. (1994). Revisionen beim Schreiben. Zur Bedeutung von Veränderungen in Textproduktionsprozessen. Tübingen: Niemeyer. Severinson-Eklundh, K., & Kollberg, P. (2001). Studying writers’ revision patterns with s-notation analysis. In T. Olive & C. M. Levy (Eds.), Contemporary tools and techniques for studying writing (pp. 89–104). Dordrecht: Kluwer Academic Publishers. Sommers, N. (1980). Revision strategies of student writers. College composition and communication, 31, 378–388. Van den Bergh, H., Rijlaarsdam, G., & Breetvelt, I. (1994). Revision process and text quality. An empirical study. In G. Eigler & T. Jechle (Eds.), Writing. Current trends in European research (pp. 133–147). Freiburg: HochschulVerlag. Van der Geest, T. (1996). Professional writing studied. Authors’ accounts of planning in document production processes. In M. Sharples & T. Van der Geest (Eds.), The new writing environment. Writers at work in a world of technology (pp. 7–24). London: Springer. Wildi, M. (2008). Customized optimization criteria in forecasting. Paper presented at the International Symposium on Forecasting. Wrobel, A. (1997). Zur Modellierung von Formulierungsprozessen. In E.-M. Jakobs & D. Knorr (Eds.), Schreiben in den Wissenschaften (pp. 15–24). Frankfurt/Berlin: Lang. Wrobel, A. (2000). Phasen und Verfahren der Produktion schriftlicher Texte. In G. Antos, K. Brinker, W. Heinemann, & S. F. Sager (Eds.), Text- und Gesprächslinguistik. Ein internationales Handbuch zeitgenössischer Forschung (vol. 1, pp. 458–472). Berlin/New York: de Gruyter. Wrobel, A. (2002). Schreiben und Formulieren. Prätext als Problemindikator und Lösung. In D. Perrin, I. Boettcher, O. Kruse, & A. Wrobel (Eds.), Schreiben. Von intuitiven zu professionellen Schreibstrategien (pp. 83–96). Wiesbaden: Westdeutscher Verlag.
29 Writers’ eye movements Mark Torrance* and Åsa Wengelin† * Nottingham Trent University, UK; † Lund University, Sweden
Introduction The challenge of developing an adequate cognitive theory of writing is to describe how the various inputs into the writing system that constitute the information available to the writer are processed to produce coherent text. This information comes in, broadly, four varieties: topic and rhetorical knowledge represented in the writer’s long-term memory, a statement of the writing task, written or pictorial reference materials (optionally), and the text itself as it emerges on the page or screen (the “text produced so far”). This last—the use that the writer makes of their own emerging text during text production—is the focus of this chapter. Our purpose is to argue studying writers’ eye movements is likely to provide important insights into the complex mental processes that underlie writing. As things stand, there is little hard evidence to support this case: With the exception of two methods-focused papers (Alamargot, Chesnet, Dansac, & Ros, 2006; Wengelin et al., in press) we know of no journal-published literature exploring where writers look in the text that they are composing. This is in stark contrast to the extensive literature exploring readers’ eye movements (reviewed, for example, by Rayner, 1998). The bulk of this chapter is taken up with describing and discussing a range of different kinds of eye-movement behavior that we have observed in our research. This draws on pilot work conducted by Torrance and co- workers at Staffordshire University, United Kingdom and more developed studies by Wengelin and co-workers at Lund University in Sweden. Our intention is to whet readers’ appetite for this kind of research rather than to discuss findings in detail (which would make for a very long chapter) or to draw firm conclusions about underlying cognitive mechanisms (which would be premature). First, however, we briefly describe the methods that we use to collect these data.
1 Methods Before we can start to draw conclusions about underlying processes it is necessary to establish ways both for tracking writers’ eye movement and
Writers’ eye movements 395 for analyzing the subsequent eye-movement records in a way that generates useful summary statistics across multiple participants. These analyses must be possible without taking up prohibitively large amounts of researcher time. Determining what is being looked at is rather more complex in writing than in reading, and this is perhaps one of the reasons why writing has not been a focus of eye-movement research. The research that we describe in this chapter focuses on adult writers composing at a computer by keyboarding. Eye-tracking systems are now sufficiently advanced to allow accurate prediction of which word a reader (or writer) is looking at in a screen full of text. The system used by Torrance and co-workers, for example, gives at least word-level accuracy, for most participants, with text displayed on a 21-inch screen in 12-point font and 30-point line spacing. This is possible without having to restrain the participant’s head, an important consideration when collecting data from tasks that take more than a few minutes, and from writers who need to see their fingers in order to type. Eye-tracking systems output the time (in milliseconds) and screen location (in x, y coordinates) for each fixation made by the participant on the computer screen (and, in the case of the system used by Wengelin and co- workers, on planes around the computer screen, including the keyboard and, perhaps, paper-based reference materials). A fixation is a period of time where gaze remains stable on a particular object. Reading the previous sentence, for example, would involve fixations on, at least, fixation, period, time, gaze, and so forth. In our data we typically find around 5,000 fixations of greater than 100 ms duration over the course of a 30-minute writing session Between these fixations are rapid movements, called saccades, as gaze is shifted from word to word. Findings from reading research indicate that objects are perceived and processed during fixations but not during saccades. If the participant is simply reading static text provided by the researcher, as is the case in reading tasks, then it is straightforward to associate fixation at particular screen coordinates with fixation on a particular word. However, for a composition task the text that the writer’s eyes move across develops as the task proceeds. This means that word locations cannot be prescribed by the researcher. Also line wrapping and scrolling mean that even once a word has been produced its screen location is likely to change from time to time. Deriving useful information about where in the text the writer is looking is, therefore, not altogether straightforward. The approaches that we have been using combine a log of the time and screen coordinates of each fixation with a record of each key press. In brief summary, Wengelin and co-worker’s strategy involves combining eye- movement data with keystroke data from an existing logging program (Scriptlog; Strömqvist & Karlsson, 2001). They then pass eye-movement data through a “reading filter” that uses statistical pattern-matching methods to identify specific fixation patterns that are likely to be associated with, for example, reading existing text or scanning backward
396 M. Torrance and Å. Wengelin
Gaze on monitor
Gaze on keyboard
Keystrokes Reading filter Fixation coordinates
Time (s)
Figure 29.1 Sample output.
through text. This outputs information of the form “between times t and t1 the writer was probably reading their own text.” This information is then synchronized with a log of the writers’ keystrokes and displayed graphically. Sample output from this system is shown in Figure 29.1. Torrance and co-workers adopt a rather different approach that involves writers using a dedicated text editor which records both key-press and fixation events into a single log. A second program then uses information about line spacing and font and display size to identify which word was being looked at any particular point in time. The resulting data can be used to determine, for example, how far back into the text on average writers tend to read when they pause. It can also be used to calculate, across a number of writers, mean frequency and duration of particular formally defined patterns of eye-movement and keystroke activity. For example, activity might be categorized as reading when writers make three or more forward saccades of no more than three words and without typing any characters. This analysis performs the same function as Wengelin and co-workers’ reading filter but on the basis of rigid definitions. This is in contrast to, and complements, the fuzzier (and therefore more flexible) categorization provided by the reading filter. Both approaches are described in more detail in Wengelin et al. (in press).
2 Phenomena In starting to explore the large range of different eye-movement patterns associated with text production, which are necessarily more varied than those associated with reading, we have found it helpful to make two distinctions. First, fixations can occur either concurrently with keyboarding or during periods (or “pauses”) when no keys are struck. Second, fixations
Writers’ eye movements 397 can be local to the last inserted character or further back within the text. These distinctions suggest four kinds of eye activity (concurrent-local, concurrent-distant, pausing-local, pausing-distant), each serving a different set of functions within the writing system, and each associated with different cognitive processes. 2.1 Fixations concurrent with typing Our findings so far suggest that the vast majority of typing-concurrent fixations are on or very near the word currently being written. When handwriting, this will mean that the writer is looking at the word in which the nib of the pen is currently positioned. When keyboarding, however, things are a little more complex because some writers habitually fixate the screen when typing and some writers habitually fixate the keyboard (typically those who have not developed touch-typing skills). These appear to represent quite distinct groups. Wengelin and co-workers cluster-analyzed eye- movement data from 28 undergraduate writers composing short texts. Of these, 17 formed a “keyboard-gazer” cluster with a mean of 80 percent of total time spent looking at the keyboard, and seven formed a “monitor- gazer” cluster with a mean of 95 percent of time spent looking at the screen. Only four writers could not be easily classified into one of these two groups. Monitor gazers tended to be faster typists, typing around seven mid-word characters per second compared to four characters for keyboard gazers (Johansson, Wengelin, Johansson, & Holmqvist, in press). 2.1.1 Monitoring the word currently being typed The existence of these two quite different approaches to coordinating gaze and key presses raises questions about exactly what role visual information plays in the process of moving from the intention to type a word to the word appearing on the screen. Keyboard gazers presumably require visual input from the keyboard in order to construct the motor plans necessary to execute a particular letter: In order to fulfil the intention to type the letter “a,” keyboard gazers need to be able to see the A-key but monitor gazers do not. However, the role played by visual monitoring of the word being typed as it appears on the screen is less clear. One way of exploring this is to look at the eye movements associated with making a typographical error (henceforward “typo”). Monitor gazers often correct typos immediately after they have been created. For example, a monitor-gazing student writer whom we recently studied generated the following sequence of keystrokes in an attempt to type the word “through” (arrows represent deletion using the backspace key). tthr←←←roug←←←←j←hrough
398 M. Torrance and Å. Wengelin All 23 keystrokes were completed within 3,900 ms. Her pattern of fixations was as follows: She first fixated “tt” at the start of the word and then moved her gaze slightly forward as she typed more characters (“hr”). Her gaze then moved back to the start of the word (back to the error “tt”) and remained there while she moved the cursor to the error location. A similar pattern occurred with the second error (omission of the “h”) and the third error (accidental typing of “j”). In all three cases both the writer’s gaze and typing moved beyond the error before returning to correct it. One interpretation of this pattern is that the writer became aware of the error by observing it as it occurred but took some time to process it and plan a response. During this processing and planning period existing keystroke and eye-movement motor plans continued to be executed. Thus error correction resulted directly from the writer seeing the error on the screen. This is a plausible explanation. We have, however, also seen this very rapid error correction in writers who habitually fixate the keyboard when typing. Figure 29.2 shows a sequence of fixations and keystrokes which starts with a keyboard-gazing writer mistyping “however.” They then fixate a word somewhat to the right of the mistyped word, then move their gaze to look directly at it, deleted the word up to the offending character, returned gaze to the keyboard, and retyped the remainder of the
Figure 29.2 Sequence of fixations and keystrokes, mistyping “however.”
Writers’ eye movements 399 word correctly. Again, it is possible that the writer identified that an error had occurred by looking at the incorrect word. However, this approach to error detection would involve looking at the screen after completing each word (or miss most errors). This was not the case for this writer. She typically wrote whole clauses or even sentences before looking at the screen. She very rarely looked at the screen mid-word, and when she did this was invariably after she had made an error. This suggests, therefore, that screen fixations around errors often, or even typically, occur as a response to the writer having already recognized that an error has been made: They result from, rather than cause, error detection. These findings suggest two different hypotheses about the role of visual feedback from the word currently being typed. One possibility is that monitor- and keyboard-gazers use different mechanisms to detect errors. Monitor-gazers become aware of mistyped words when they see them appear on the screen. Keyboard-gazers become aware of mistyped words, at least in some instances, without looking at the screen, perhaps by observing their fingers pressing an unintended key. Alternatively, it may be that both monitor- and keyboard-gazers detect errors, at least in some cases, through kinaesthetic feedback from their fingers: when the fingers press unintended keys they “feel wrong.” Lackner and Tuller (1979) discuss a similar mechanism in speech production. If this is the case then monitor-gazers’ fixations on the word currently being typed simply serve to keep track of the current screen-location of the cursor rather than processing the word that is currently being typed. This would speed up locating typos (and other features of the text-just-typed) on the screen, doing away with the need for fixations that seem to serve a general cursor- search function which are common in keyboard-gazers’ eye movements (e.g., fixation 2 in Figure 29.2). It may even be that visual processing actually interferes with the production process. We have observed that monitor-gazing participants often fixate the first and last words of a line directly but fixations between these points, though tracking the words being produced, are then substantially above the line of text. This pattern does not occur during reading. It may be that this represents strategic diversion of attention away from the detail of the words being typed. Research is clearly needed to determine exactly what functions are served by fixations on the word currently being typed. This is an interesting question in its own right. It is also one that needs to be answered if researchers interested in higher-level processes are to make informed decisions about choice of research participants. Are the writing processes of monitor-gazers and keyboard-gazers (and, by extension, people composing by pen) essentially similar or importantly different? 2.1.2 Fixations away from the word currently being typed The events that we are thinking of here are situations in which the writer continues to type, or breaks typing for a very short period, and looks back
400 M. Torrance and Å. Wengelin into the text that they have already written. Our impression is that for most writers the vast majority of fixations concurrent with typing are either on the keyboard or, in monitor-gazers, on the word currently being typed. For example, for an experienced, monitor-gazing writer that we studied completing a short (30-minute) expository text, 77 percent of keystrokes were made within four characters of the point where the writer was fixating, and 95 percent within eight characters.1 We have obtained very similar figures (79 percent and 95 percent) for a less experienced, monitor- gazing writer performing a similar task. It is, however, interesting to conjecture possible functions for the remaining 5 percent of typing-concurrent fixations. It is possible that these more distant fixations do not, in fact, represent a shift in attention. Occasionally gaze will move from the point of attention simply because the eyes need a rest. It may, however, also be that they represent genuine parallel processes in which the writer’s attention shifts from the word being typed to a different part of the text. In speech production there is evidence that as output of the current clause draws to an end speakers are able to shift attention to composing the next utterance (e.g., Power, 1986), and Chanquoy, Foulin, and Fayol (1990) have found similar effects with writers. Eye movements during these phases might, therefore, be associated with either reading previous text to cue generation of new content. It is also possible that distant typing-concurrent fixations help in the development of reference between sentences. We have observed that distant fixations are often on a word or phrase with the same semantic referent as the word being typed, for example the same word, a synononym, a hyponym or a hyperonym. Similar behavior was identified by Holmqvist, Holsanova, Johansson, and Strömqvist (2005). 2.2 Fixations during pauses Pausing, for present purposes, is defined as any period of time when typing ceases for two or more seconds. Writing researchers have typically assumed that processes occurring during these periods are qualitatively different from those that occur while the writer is keyboarding or scribing: Pausing is likely to be associated with either reviewing what has already been written, or planning what to say next, or some combination of these activities. 2.2.1 General patterns Pause frequency and duration are related to the location within the text at which the pause occurs (Foulin, 1998; Spelman-Miller, 2000; Wengelin & Strömqvist, 2004): It is much more likely that writers will pause at a sentence or, particularly, a paragraph boundary than between or within words. For example, in data from nine undergraduate writers composing short argumentative texts we found that the probability of pauses occur-
Writers’ eye movements 401 ring at character, word, sentence, and paragraph boundaries were, respectively, 2 percent, 6 percent, 31 percent and 45 percent. There was a similar trend in mean pause durations, with the longest pauses occurring at paragraph boundaries. It is possible that writers pause at higher-level boundaries to take stock of what they have already written. Crudely, pauses at paragraph boundaries may be associated with reading or at least scanning the paragraph just written, pauses at sentence boundaries, the sentence just written, and so forth. We have found some evidence that seems to point in this direction. For our sample of nine writers, 80 percent of fixations occurring during pauses at character boundaries were within 22 words of the last-typed character. Fixations at word boundaries showed a very similar pattern. However, at sentence and paragraph boundaries writers looked back to much greater depth in their text, with the cut off for 80 percent of fixations occurring at, respectively, 60 and 70 words distant from the last inserted character. However, a relatively small percentage of pause time was taken up with the kind of sustained reading activity that would be typical of study of a passage of unfamiliar text. Using an operational definition of reading as sequences of three or more fixations linked by forward saccades of not more than three words or 25 characters, we found that around 20 percent of total pause time at character, sentence, and paragraph boundaries was taken up with reading, with little variation across boundary types, although for some reason, this figure dropped to 9 percent for pauses at word boundaries. More generally, reading consumed on average 8 percent of writers’ time—around 2.3 minutes of the 30 minutes total time on task. Research by Wengelin and co-workers with a much larger sample of Swedish-speaking writers (n = 80), and adopting their heuristic “reading filter” approach to identifying reading episodes, suggests very similar total reading times (9 percent of total time on task). This drops to 8 percent for 15-year-olds completing the same task. They also sampled students with reading and/or writing difficulties who read slightly less (7.5 percent for university students and 5 percent for 15-year-olds). The remainder of eye activity during pauses was taken up with fixations at different locations within the text, linked by both forward and backward saccades. This jumping around within the text is, we believe, often interpretable within the specific textual context at which a pause occurs, as we hope to demonstrate in the next section. 2.2.2 Two examples Figure 29.3 shows activity during a short (three-second) pause that occurred at the end of a mid-paragraph sentence. It is associated solely with correcting a local error, specifically a semantic mismatch between the modifier and the head noun in a prepositional noun phrase. On completing the phrase the writer looked back first at the head noun, then at the modifier, then at the head noun again, before taking action to correct the error.
402 M. Torrance and Å. Wengelin
Figure 29.3 Activity during three-second pause.
This eye activity cannot be interpreted as routine monitoring. There is nothing in this sequence that suggests that the writer deliberately and strategically read back over what they had written. Also, study of the rest of this writer’s eye and keystroke protocol suggests that they did not routinely look back at the modifier and head noun in each prepositional phrase that they generated. It appears therefore, that the eye movements shown in Figure 29.3 result from, rather than cause, the writing realizing that they have made a mistake. This pattern, if replicated across most situations in which a similar error was made, would point to a three-component model of error detection in which becoming aware that an error has occurred, locating the error, and identifying the nature of the error are three distinct components which potentially occur independently and sequentially. In the Figure 29.3 sequence, error awareness appears to have occurred prior to, and to prompt the initial recursion (backward saccade) to the head noun (ice caps). There then follows movement back to locate the source of (or other component in) the error, and movement forward again to either identify or confirm the nature of the mistake. This account, if correct, leads to questions about the processing that occurred prior to and during the typing of “ice cap” that allowed the error to occur in the first place. It suggests that at the semantic level the writer held two related but competing representations of the message of the sentence. Figure 29.4 is a more extended example of eye activity during pauses. It shows two consecutive pauses, one lasting 1.4 seconds (which actually falls below the duration that we set above for what constitutes a pause) and a second of 3.8 seconds. This activity was associated with completing one paragraph and starting a new one. One possible account of the creation of a paragraph break is that a writer starts a new paragraph on the basis of what they plan to say next. If for discursive reasons the new content is best set out in a new paragraph then one is created. The eye and key activity shown in Figure 29.4 suggests that, in this instance at least, this was not the underlying process. Figure 29.4 points instead to two distinct stages, with termination of the existing paragraph being independent, to some degree, of knowing what to say in the next. On completing the sentence the writer pauses and reads back through the final verb phrase. This then prompts typing of a hard carriage return. The writer’s eyes then move with
Writers’ eye movements 403
Figure 29.4 Extented example of eye activity during pauses.
the cursor to the start of a new line. However, instead of immediately starting to type there is then another, relatively long pause while the writer’s gazes returns to a single word in the previous sentence. Exactly what is being planned during this second pause is not altogether clear. However, the word that is fixated in the previous sentence (“minimizing” with reference to petrol consumption) does appear to then be elaborated in the first sentence of the new paragraph. This two-stage process is also apparent in pause patterns at sentence boundaries: Wengelin (2006) found that where pausing occurred at a sentence boundary, in around 50 percent of cases the writer pauses both before and after typing the full-stop (period). Analysis of where the writer looks during these pauses at boundaries has potential of uncovering mechanisms underlying both text-span (clause, sentence, paragraph) termination, and how existing text is used to support planning of the next part of the text.
3 Discussion In this chapter we have briefly introduced a wide range of findings, many preliminary, that we hope have roused the curiosity of readers interested in the cognitive processes that underlie text production. We would like to finish with four observations. First, from a methodological perspective we think that our findings indicate that appropriate analysis of combined eye-tracking and keystroke data provides a useful approach to exploring writers’ eye movements. Second, and again concerning methods, we think that our understanding of the cognitive processes associated with writing are best advanced by a combination of large-sample studies, which are common in the psychology of writing literature, and detailed analyses of single writing sessions, which are less so. Third, study of writers’ eye movements suggests a research paradigm in which focus moves away from broad questions about which writing activities lead to the best quality text, to answering more fine- grained questions about how writers plan the next sentence, word, or
404 M. Torrance and Å. Wengelin clause. Finally, and less contentiously, we hope that we have convinced the readers that the study of eye movements can improve our understanding of the cognitive processes underlying text production. However, there is probably much to learn about basic attentional and eye-control issues in writing, before we can be confident about inferring writers’ psycholinguistic processes from patterns in their eye movements. For example, throughout this chapter we have assumed that fixated words are processed. The extensive literature on eye movement during reading suggests that in most cases this is a reasonable assumption for fixations over about 100 ms. It would seem reasonable to assume that the same holds true in writing. However, this needs to be established empirically rather than just assumed. For example, as we discussed above, in monitor- gazers a large majority of fixations tends to follow the cursor. Some of these are probably used for monitoring the word just being typed, but it is an open question as to the level to which these words are actually processed. We currently know little about where writers look when they write and, more generally, little about the detail of moment-by-moment text- production processes. Increasing availability of very accurate eye-tracking systems, and the development of methods by which these can be used in the context of writing, provide a basis for considerable future study. We hesitate to predict a trajectory, but we think that there is at least potential for considerably developing our knowledge of how writers move from intention to inscription.
Note 1 These values exclude keystrokes where the concurrent fixation was not within the existing text.
References Alamargot, D., Chesnet, D., Dansac, C., & Ros, C. (2006). Eye and pen: A new device for studying reading during writing. Behavior Research Methods Instruments & Computers, 38(2), 287–299. Chanquoy, L., Foulin, J.-N., & Fayol, M. (1990). Temporal management of short text writing by children and adults. Cahiers de psychologie cognitive, 10(5), 513–540. Foulin, J.-N. (1998). To what extent does pause location predict pause duration in adults’ and children’s writing? Cahiers de psychologie cognitive, 17(3), 601–620. Holmqvist, K., Holsanova, J., Johansson, V., & Strömqvist, S. (2005). Perceiving and producing the frog story. In D. Ravid & H. Shyldkrot (Eds.), Perspectives on language and language development: essays in honor of Ruth A. Berman (pp. 289–302). Amsterdam: Kluwer Academic Publishers. Johansson, R., Wengelin, Å., Johansson, V. & Holmqvist, K. (in press). Gazing at the keyboard or gazing at the monitor: Influences on text production. Reading and Writing.
Writers’ eye movements 405 Lackner, J. R., & Tuller, B. (1979). Role of efference monitoring in the detection of self-produced speech errors. In W. Cooper & E. Walker (Eds.), Sentence processing: Psycholinguistic studies presented to Merril Garret. Hillsdale, NJ: Lawrence Erlbaum. Power, M. J. (1986). A technique for measuring processing load during speech production. Journal of Psycholinguistic Research, 15(5), 371–382. Spelman-Miller, K. (2000). Academic writers on-line: Investigating pausing in the production of text. Language teaching research, 4(2), 123–148. Strömqvist, S., & Karlsson, H. (2001). Scriptlog for windows: User’s manual. Lund and Stavanger: Department of Linguistics, Lund University and Centre for Reading Research, University College of Stavanger. Wengelin, Å. (2006). Examining pauses in writing: Theory, methods and empirical data. In K. Sullivan & E. Lindgren (Eds.), Computer key-stroke logging and writing (pp. 107–130). Amsterdam: Elsevier. Wengelin, Å., & Strömqvist, S. (2004). Text-writing development viewed through on-line pausing in Swedish. In R. Berman (Ed.), Language development across childhood and adolescence: Psycholinguistic and crosslinguistic perspectives (pp. 170–190). Amsterdam: John Benjamins. Wengelin, Å., Torrance, M., Holmqvist, K., Simpson, S., Galbraith, D., Johansson, V., et al. (in press). Combined eye-tracking and keystroke-logging methods for studying cognitive processes in text production. Behavior Research Methods.
30 Text analysis as theory-laden methodology Nancy Nelson* and Stephanie Grote-Garcia† *University of North Texas (College of Education) † Texas A&M University – Corpus Christi
Researchers throughout the disciplines analyze written documents for features of interest to them and to their particular groups, and, in doing so, gain insights into human thought and social activity. Text analysis is especially important in the field of composition studies, which has its central research focus on the production of written texts, particularly as performed by academic writers. This chapter is about texts and about methodology, and it is also about theory. In some ways the adjective theory-laden in its title is redundant, since methodology is, by its very nature, infused with theory. Nevertheless, this term provides a needed emphasis because some uses of methodology refer only to method or technique, and methodology is much more than method or technique. It is description, explanation, and justification of method. The term theory-laden, which has been employed by many, can be attributed to Hanson (1958), who spoke of perception as “a theory-laden undertaking.”
Methodology and theory-ladenness It seems quite clear that methodology encompasses epistemology as well as method. Kaplan (1964) made this point, years ago, in his classic volume, The Conduct of Inquiry: “[T]he subject-matter of methodology consists— very roughly speaking—of the most basic questions that can be raised concerning the pursuit of truth” (p. 20). More recently, Tuchman (1994) restated this point when she said that methodology includes the “epistemological assumptions implicit in specific methods” as well as the assumptions regarding how “a specific method ‘captures’ the ‘object’ of study” (p. 306). To this, one might add that a researcher’s assumptions are reflected not only in the choice of objects that are studied but also in those that are not studied. In a choice of method, the researcher reveals a particular view of “reality.” Research is human activity imbued with questions not just about what counts as knowledge but also about who makes the decisions about what counts. It is an activity that is rhetorical and political as well as intellectual. Researchers themselves are culturally and historically situated, and that situated-ness affects what is looked for and what is seen. As members
Text analysis as theory-laden methodology 407 of the writing research community know so well, the positioning of a particular orientation toward research is often accomplished by means of “othering” another orientation. At times, the community has experienced rather heated debates over competing approaches and, more significantly, the assumptions that undergird them, since much more than the specific approach has been at stake. Thus, although the emphasis here is on analytic procedures, this chapter is more than a review of technique; it is—and must be—also an account of the differing and shifting theories and ideologies that have shaped the field. It thus addresses the issue of what counts as knowledge.
The text problematic Before getting to questions about what is in a text, it is necessary first to tackle the problem of what a text is—or, put another way, what is meant by the word text, a commonplace term that has become a contested one in recent years. Historically, the two features that have figured most prominently in defining textuality are unity and connectedness. In addition to these two features, there is the matter of interpretation: Whether something is or is not to be considered a text depends on the way in which it is interpreted in a particular situation. This latter point is critically important in a consideration of text analysis. Theory-ladenness in writing research is manifested, among other ways to be discussed further, in how text is conceptualized. As is well known, the notion of text has become increasingly complex, given poststructural and postmodern developments with respect to intertextuality. In the “post” discussions, texts are described as absorbing and transforming other texts and viewed as mosaics or tissues of quotations. Also the notion of text has been challenged by the development of hypertext and its network or rhizome metaphor. It must be acknowledged that such conceptions certainly complicate the concept of text by challenging its discreteness and its boundaries but they do not successfully negate it. This is because the notion of intertextuality does not succeed unless there are texts to be connected through the inter and also because hypertextuality must also retain the notion of a text, since it represents the linkages of texts to other texts as well as the embedding of texts within another text. Thus, in a sense, there are texts within a text (the mosaic or tissue idea) and there are also ties between texts and other texts (the network or rhizome idea)—but, importantly, there is also something that can be treated and talked about as a text. For this piece, one should think of a text as language that functions as a unit in communication and that is perceived as a functional unit. With respect to matters of function and perception, this means that it is treated, spoken of, and analyzed as a functional unit, even as a unit for analysis in composition research. In this discussion, we refer to text in its written form, even though that too is debatable, since the term can be used more broadly to refer to oral discourse as well.
408 N. Nelson and S. Grote-Garcia Relevant to this discussion is the notion of entextualization as employed by Silverstein and Urban (1996) in their consideration of methodology. These scholars used this term for the process through which a stretch of language comes to be seen as a text, and they explained: “To turn something into a text is to give it . . . a form and meaning that are imaginable apart from the spatiotemporal and other frames in which they can be said to occur” (p. 1). Writing researchers do this in their research: they entexualize the linguistic excerpts to be treated as texts. This is done not only with the written products being studied but also with field notes from observations, transcriptions of interviews, and typed protocols from think- alouds. A major point we hope to make in this chapter is that these products too become texts to be analyzed.
History: turns or lines? In seeing how theory relates to methodology, our focus is on three questions that have guided and continue to guide writing research: What is the nature of the product? What is the nature of the process (cognitive and social)? What is the nature of practices? In reviewing this research, it would have been possible to present a chronological account—adopting the familiar tropes about shifts and turns: cognitive revolution, social turn, post-process, sociocultural turn, and now what some are calling “post- social.” But the history of writing research is not so simple. Although one can discern a movement within the field of composition studies from focus on one question to another, the history of writing research is not a simple chronology of movement toward some shared objective, nor is it characterized by dramatic and absolute shifts in the field from one question to another. This history of traditions in text analysis, like other histories, is marked by discontinuities; research agendas overlap and are transformed. This is the case particularly when one takes a broad view of the field that includes international work as well as what is conducted in North America. For this reason, the present chapter is organized around questions—all of which have a history but continue to inspire research today.
What is the nature of the product? It has become conventional to mark the era of empirical research into writing, in the United States anyway, as beginning in the 1960s, even though much research had been conducted before that. For this empirical work at the mid twentieth century there was a need for numbers, and the numbers in writing research were to come most often from text analyses. Although the focus of the research was on the textual product, it was not the textual features themselves that were the major interest but was instead what those features would show about mental products—the kinds of knowledge used in writing and, in one line of the research, something that came to be called “mental representation.” Attention continues today to be
Text analysis as theory-laden methodology 409 directed to textual features and what they can reveal about those who produce them. What was measured early on was quite often some feature related to syntax. Instrumental here was transformational generative grammar, but more important than that to researchers was a largely unarticulated theory—and assumptions tied to it—that was associated with Hunt’s (1965) identification of T-unit, a main clause along with its modifiers and subordinate clauses. For research employing T-unit analyses, the numbers were easy to derive by counting the number of words in a T-unit, and there were high degrees of interrater reliability. The shared, and largely unexamined, assumption was that syntactic complexity, as measured by such counts, was related in a rather direct and linear way with maturity in writing and complexity of thought. Research focused mainly on developmental patterns and on effects of sentence-combining interventions. But the assumptions guiding this research were critiqued by Kinneavy (1979), who pointed to the lack of a theoretical rationale, and by Faigley (1980), whose article title said it all: “Names in Search of a Concept: Maturity, Fluency, Complexity, and Growth in Syntax.” Faigley argued that simply increasing clause length or T-unit length did not necessarily indicate increases in the constructs labeled variously as maturity, fluency, or growth in writing. And he questioned the assumptions on which such links were made, pointing out that “rhetorical considerations of audience, subject, and purpose [can] cause wide fluctuation in the syntactic profiles of essays” of writers (p. 293) and that syntactic complexity is not necessarily tied to cognitive complexity. His major point was that many researchers had “bought into” assumptions that do not necessarily hold when one considers actual writing in context. Thus he was not critiquing rhetorically oriented research, such as studies focused on variations in syntax when the audience or task varied. Although it receives much less attention in the research today, syntactic complexity is still being studied in text-analysis studies to examine possible differences resulting from some characteristic of the rhetorical situation. It is being employed, for instance, to study possible differences associated with various tasks for computer-based writing. And the T-unit persists as a relatively easy means of parsing a text into units in a fashion that is “reliable” in terms of agreement among coders. That work has addressed—and still addresses—the matter of syntax, but there is also the issue of semantics. In the late 1970s and in the 1980s other researchers became part of an international discourse-comprehensionand-production movement associated with, and informed by, cognitive science, and they began giving their attention to semantic features of texts (e.g., see discussions in Nelson, 2001, 2008, and Spivey, 19971). This became a time for analyzing a text into propositions and top-level structures in order to study its semantic content and organization and for analyzing its cohesive ties, referential overlap, or given new patterns in order to study its interconnectedness. Discourse researchers were developing a
410 N. Nelson and S. Grote-Garcia new, specialized vocabulary that is still useful today for talking about text and that included such terms as rhetorical structures, idea units, and textual hierarchy. Associated with much of this work today as in the past is a template approach to study transformations that writers make when performing various reading-writing tasks, such as summarizing a text or using a source text or multiple texts as the basis for one’s own essay. In this approach, templates are created and used in the analyses. When the summaries or essays are examined (and compared with the text template), the deviations from the original indicate the influence of some factor of interest to the researcher, such as the writer’s purpose. This approach has been undergirded with the theoretical assumption that the text base manifested in the writing is at least somewhat isomorphic to a mental text base—and thus conclusions can be drawn about cognitive products. An important verb in this line of work is constructing, or building, and what is built is a mental representation. Frederiksen (1986) explained this assumption in this way: The goal of cognitive studies of written discourse is to describe how meaning and language structures are constructed by a writer and interpreted by a reader. Such a description includes a specification of the meaning and language structures, called “representations,” that writers and readers construct and the rules upon which these representations are based. (p. 227) This is a major theoretical assumption: that the features of the text provide cues to a cognitive configuration. Ordering patterns are thought to give indications of organizational operations on the part of the writer, inclusion of particular material to signal the writer’s choices, and textual links to suggest relations that the writer made between units of content. Figure 30.1 is an illustration from a 1980s study (Spivey, 1984) examining discourse synthesis, an act of composing in which writers integrate material from multiple textual sources. The clusters of letters indicate particular topics the writer combined together in a content unit, and the large sections indicate thematic “chunks,” or organizational units, some of which are interlinked because of inferential connections the writer made. Text analysis in the cognitive tradition has proved particularly useful in the writing-to-learn research. The textual material to be learned can be analyzed as can students’ writing relative to that material, and then comparisons can be made to see how the students have appropriated or transformed what they read. Today interest continues in rhetorical structures and their functions in written discourse, and information technology is providing other research tools, such as text-mining, which can be used to study interrelated texts in a corpus.
Text analysis as theory-laden methodology 411
Figure 30.1 Representation of a writer’s text (source: Spivey, 1984).
What is the nature of the process of writing? For the question about writing process, it is necessary to respond in two ways—to consider the cognitive process and to consider the social process. Both lines of work focused on process developed concurrently, and both continue today. But in terms of methodology, research pursuing the two questions are associated with different traditions, assumptions, and approaches. Cognitive process. Among writing educators there was much talk in the 1970s of “teach process, not product,” and a similar theme came from many writing researchers: that one should “study process, not product.” Some, like Murray (1980), made strong statements: “Process can not be inferred from product any more than a pig can be inferred from a sausage” (p. 3). Two major assumptions guiding research labeled as “process- tracing” are that there are rather direct means of learning about processes and that process is manifested in temporal patterns. Process for many came to mean a set of subprocesses, such as planning, translating, and reviewing, as defined in the very influential Hayes–Flower (1980; Flower & Hayes, 1981) cognitive model of writing. It portrayed writing as hierarchical with recursive movement among these subprocesses, all of which have sub-subprocesses. A large number of studies have addressed one, several, or all of these subprocesses as writers compose their pieces with pen and paper or on the computer. Since its inception, much of this work has fit theoretically into the problem-solving paradigm of cognitive science. The writing task is the problem to be solved, and the process becomes the solving of it. Problem-solving has been—and still is—a powerful theoretical metaphor, a way of seeing one process through the lens of another.
412 N. Nelson and S. Grote-Garcia A favored and much-used research approach for many researchers in this process paradigm has been the think-aloud procedure. Its proponents view it as concurrent and direct—but admittedly incomplete. As Hayes and Flower (1980) pointed out when this research was just beginning, a protocol is an incomplete account like “the tracks of a porpoise, which occasionally reveals itself by breaking the surface of the sea. Its brief surfacings are like the glimpses that the protocol affords us of the underlying mental process” (p. 9). A protocol, transcribed from a tape, becomes the focal text—the text to be carefully analyzed. It becomes the text that will provide insights into plans and goals, and the writer’s text is absorbed into it, italicized, and marked as “translation” if the writer reads it aloud. The following is an example from Flower and Hayes (1981) showing how particular elements might be coded: (Plan) Ok, first day of class . . . just jot down a possibility. (Translate) Can you imagine what your first day of a college English class will be like? (Review) I don’t like that sentence, it’s lousy—sounds like theme talk. (Review) Oh Lord—I get closer to it and I get closer – (Plan) Could play up the sex thing a little bit. (Translate) When you walk into an English class the first day you’ll be interested, you’ll be thinking about boys, tasks, and professor – (Review) That’s banal—that’s awful. (p. 376) Protocol analysis was the subject of an ongoing argument in the literature that had to do mainly with the issue of reactivity: whether or not the means of data collection changes the process that is being studied. Criticisms of reactivity were answered by citations of research indicating that the think-aloud procedure slows a process but does not change it. Other issues have had to do with interpretive aspects of analysis, such as division of the protocol into episodes and derivation of categories. Despite such controversies, which are now largely past history, when it comes to concurrent collection of data, the think-aloud approach has been, and continues to be, very central to cognitive research in writing, and many researchers continue to use protocol analysis. Work in this area continues to provide detail about the cognitive process and the subprocesses of composing. For tracing of cognitive processes, there are other common approaches to data collection, such as interviews and directed retrospection, and some researchers are now also recording eye movements. It is important to note that, with all of these, as with protocol analysis, the data become entextualized into a textual product to be analyzed. It may be a protocol, a transcript, or some type of record. In the past, process-tracing sometimes also meant videotaping the writer and relating pauses and actions to the portion of the text being written or
Text analysis as theory-laden methodology 413 analyzing multiple versions of a piece—tracing initial plans and relating them to the text or simply tracing changes in the written language itself. But now there is keystroke logging, discussed in Chapter 29, which is a relatively new approach for tracing the cognitive process of writing. This is a computer-based approach, providing a detailed, annotated record of the actions of the writer, in which the logged record becomes the text to be analyzed. It is interesting to note that, with this procedure, the writer’s text absorbs the process-tracing data—in contrast to the think-aloud research, in which a writer’s text is absorbed in a text. By reading a keystroke log, a researcher can “replay” a composing episode, seeing where breaks were, where items were added, where they were deleted, and the sequence in which it all occurred. Much of the process-tracing research, which is now being conducted to a great extent in Europe, tends to be aligned with the cognitive-process model that was advanced by Hayes and Flower and has been further developed by others. Think-aloud protocols continue to provide glimpses into the process, but other approaches to process-tracing provide different kinds of glimpses. Social process. Back in the 1980s when the cognitive work in propositional analysis and also problem-solving research were developing, writing research was also experiencing the impact of qualitative approaches to research associated with sociolinguistics, sociology, and anthropology. Ethnography, ethnomethodology, conversation analysis—all provided insights into social interactions, and all brought their own theoretical assumptions, some shared and some not shared. The field was also experiencing newly discovered social theories articulated by Vygotsky and Bakhtin, among others. Researchers interested in the social nature of writing have, for decades now, been studying such activities as co-authorship, writing conferences, and discussion in peer-response groups. When writing is considered as a social process, attention must go to interactions, negotiation, and collaboration. To study these social processes, a text besides the writer’s text is needed because what is of central interest to the researcher is the process that surrounds the text—the context. A transcript of the (recorded) conversation or a set of detailed notes is entextualized; it is broken into units (episodes, exchanges, and so on), and the units are classified according to the type or nature of the exchange. Through this entextualization, the transcript becomes the central text, and it has its own textual conventions, which are coding devices for tying the written language to the oral. The following illustrates the transcription conventions employed by Ulichny and Watson-Gegeo (1989) in studying a writing conference between a teacher (T) and a student writer named Karla (K). The [ shows overlapping utterances and the double italics represent latched speech. Most relevant to this chapter is what appears in capital letters, since it is the writer’s text, which is being read aloud. It becomes incorporated into the new text, the transcription.
414 N. Nelson and S. Grote-Garcia T: (. . .)s:/ IT WAS EXCITING/ so it happened in the past// So you need to have a past tense// Not THEY SIT UP AND ROAR [but they: K: [they sat up/ = T: = a.nd/ ((slight rise, questioning)) K: Roared// T: Yeah//
(p. 316) It should be noted that transcription, which seems to be relatively straightforward, is theory-laden, like the other means of entextualization, and requires interpretation. Approaches to transcription vary considerably in the conventions for indicating elements and for interpreting their features and functions. As Mischler (1991) argued, “Different transcripts are constructs of different worlds, each designed to fit our particular theoretical assumptions. . . . They have a rhetorical function that locates them within a larger political and ideological context” (p. 271). This socially oriented research has provided insights into the influence of other individuals on writers and the joint efforts in composing that they make through social interaction. It has also introduced a different set of important matters relative to writing, such things as turn-taking and participation structures.
What is the nature of discourse practices? There is no clear line between research that focuses on social processes and research that focuses on social practices. Nevertheless, when one speaks of social factors that are sociocultural or sociohistorical, the term tends to be practice instead of process, and for research examining practices there is an intaglio of theories. For years now, some writing scholars have claimed that the field of composition studies is now “post-process,” meaning that a generalized theory of the process cannot be achieved because writing is always situated in events, in history, in communities, in social institutions and power relationships, and in history. Situated is a major concept today, and practice is a term that is increasingly in favor because it suggests acknowledgment of this aspect of writing. When one speaks of practices, writing is often interlinked with other forms of communication, including oral language, and is associated with communities of people who share ways of knowing, doing, and communicating. Genres become practices rather than structural forms. Research taking this kind of situated perspective addresses many issues, since its proponents maintain that there are multiple literacies. One major line of this research began several decades ago in studying the writing and publishing of specialized disciplinary communities, focusing on their rhetorical practices, such as citing or acknow ledging other scholars (Bazerman & Paradis, 1991). Genres, especially
Text analysis as theory-laden methodology 415 those associated with critiquing, proposing, reporting, and academic arguing, have been, and are being, studied as discourse practices. Increasingly important today is the sociocultural, historical theory of Soviet psychologists—Vygotsky, Luria, and Leont’ev—and of other scholars whose theories have been related to theirs. These include the practice theories of Bourdieu and of Lave and Wenger, and they also include activity theory, as articulated by Engeström, which is discussed in more detail in Chapter 26 of this volume. In this line of work, the unit of analysis is the activity system, whose boundaries can be difficult to determine, since systems interact with other systems. Activity theory, which focuses on action toward objects, is an orientation that breaks numerous boundaries. It is both cognitive and social and both individual and collective, and it blurs divisions between texts and other kinds of artifacts. Engeström and Miettinen (1999) have argued for moving beyond texts as a single focus of research: Exclusive focus on text may lead to a belief that knowledge, artifacts, and institutions are modifiable at will by means of rhetoric used by an author. Activity theory sees construction more broadly. People construct their institutions and activities above all by means of material and discursive, object-oriented action. (p. 10) This is a theoretical orientation—and approach—intended to capture the dynamism and historicity of human activity, which is said to include the following components: community, rules, subject, instruments (tools, artifacts), object, division of labor, and outcome. Since there is much movement among these components, texts, which may begin as artifacts or tools, can fill other roles, becoming, for instance, the object or even rules. A researcher would likely begin with Engeström’s (1999) triangular model, shown in Figure 30.2, but add to its complexity and show in this textual representation the nature of interaction among the nodes as the study progresses. In addition to cultural-historical activity theory, there are related theories, including actor-network theory and distributed cognition, in which computers and other inanimate objects are actors, or agents, along with people. This work too supports the “post-social” notion because agents are not necessarily human. Practice is the unifying theme tying together the various theories that undergird this orientation. As Turner (1994) argued, practice is an all- embracing notion, since, across various fields in the humanities and social sciences, various contributions are “[re]interpreted as assertions about practices, even though they were not originally couched in this language” (p. 1). As Wenger (1997) explained, the term practice connotes doing—but it is “doing in a historical and social context that give structure and meaning to what we do” (p. 47):
416 N. Nelson and S. Grote-Garcia Tools
Object
Subject
Rules
Community
Outcome
Division of labor
Figure 30.2 Engeström’s (1999) model of an activity system.
Such a concept of practice includes what is said and what is left unsaid, what is represented and what is assumed. It includes the language, tools, documents, images, symbols, well-defined roles, specified criteria, codified procedures, regulations, and contracts that various practices make explicit for a variety of purposes. But it also includes all the implicit relations, tacit conventions, subtle cues, untold rules of thumb, recognizable intuition, specific perceptions, well-tuned sensitivities, embodied understandings, underlying assumptions, and shared world views. (p. 47)
Conclusion: the status of text analysis in writing research Two conclusions may be reached regarding text analysis as used to study writing. The first is that, in research employing text analysis, the texts created by the writers being studied are not always the major texts to be analyzed and are not always the major support for claims that a researcher makes. In many cases, texts are entextualized from other sources, and what is the central text for analyses varies across the research paradigms, or traditions. In research focused on the product, the writer’s text is valuable, mainly for what it can reveal about something else, often the writer’s knowledge, ability, or mental “product.” In research called “cognitive process research” the central text is likely derived from the tracing procedures because it is important to have data providing evidence of the temporal aspects. It may be, for instance, a think-aloud protocol or keystroke record. In research that attends to the social process there is also entextualization resulting in such texts as interview transcripts, records of discus-
Text analysis as theory-laden methodology 417 sions or conversations, and fieldnotes. In research on social practices, much depends on the theoretical orientation of the researcher. For example, in studies of disciplinary practices, writers’ texts might be the major source, but in research on activity systems written texts become just one of many kinds of data that are collected. The second conclusion has to do with the figurative and figural nature of the theories that guide research. The theories provide verbs that are used in talking about writing, and those verbs guide research as well as understandings. Writing is, variously likened to building something, to solving a problem, to interacting (negotiating, collaborating), and to doing. The nature of the verb influences the source of data and the means of analysis as well as conclusions that can be drawn. If writing is constructing, attention must go to what is built. If writing is solving a problem, a problem must be presented or perceived; and then the writing process is seen through the lens of solving. If writing is interacting (or negotiating or collaborating), attention will go to engagement between two or more entities. If writing is doing, activity will be seen, and that activity will be physical and mental and probably social as well. This chapter has reviewed the development of text analysis in writing research and, in doing so, has offered a brief overview of research traditions in the field. That overlap has been unavoidable, since inquiry procedures are tied to the critical concerns in a field. It is true that our research approaches are adapted to our questions (we develop the means to get the answers we seek). However, it works the other way too. In accordance with what Kaplan (1964) called “the law of the instrument,” our questions are adapted to our approaches (we come up with questions that can be answered through the means we have available).
Note 1 The first author of this piece, N. Nelson, published previously as N. Nelson Spivey.
References Bazerman, C., & Paradis, J. (Eds.). (1991). Textual dynamics: Historical and contemporary studies of the rhetoric of the human sciences. Madison, WI: University of Wisconsin Press. Engeström, Y. (1999). Activity theory and individual and social transformation. In Y. Engeström, R. Miettinen, & R.-L. Punamäki (Eds.), Perspectives on activity theory (pp. 19–38). Cambridge: Cambridge University Press. Engeström, Y., & Miettinen, R. (1999). Introduction. In Y. Engeström, R. Miettinen, & R.-L. Punamäki (Eds.), Perspectives on activity theory (pp. 1–18). Cambridge: Cambridge University Press. Faigley, L. (1980). Names in search of a concept: Maturity, fluency, complexity, and growth in written syntax. College Composition and Communication, 31, 291–300.
418 N. Nelson and S. Grote-Garcia Flower, L., & Hayes, J. R. (1991). A cognitive process theory of writing. College Composition and Communication, 32, 365–387. Frederiksen, C. H. (1986). Cognitive models and discourse analysis. In C. R. Cooper & S. Greenbaum (Eds.), Studying writing: Linguistic approaches (pp. 227–267). Beverly Hills, CA: Sage. Hanson, N. R. (1958). Patterns of discovery: An inquiry into the conceptual foundations of science. New York: Cambridge University Press. Hayes, J. R., & Flower, L. S. (1980). Identifying the organization of writing processes. In L. W. Gregg & E. R. Steinberg (Eds.), Cognitive processes in writing (pp. 3–28). Hillsdale, NJ: Erlbaum. Hunt, K. (1965). Grammatical structures written at three grade levels (Research Report No. 3). Champaign, IL: National Council of Teachers of English. Kaplan, A. (1964). The conduct of inquiry: Methodology for behavioral science. San Francisco, CA: Chandler. Kinneavy, J. (1979). On the misinterpretation of syntactic complexity data. English Education, 11, 91–97. Lave, J., & Wenger, E. (1991). Situated learning: Legitimate peripheral participation. New York: Cambridge University Press. Mischler, E. G. (1991). Representing discourse: The rhetoric of transcription. Journal of Narrative and Life History, 1, 255–280. Murray, D. M. (1980). Writing as process. In C. R. Cooper & L. Odell (Eds.), Research on composing: Points of departure. Urbana, IL: NCTE. Nelson, N. (2001). Discourse synthesis: Process and product. In R. G. McInnis (Ed.), Discourse synthesis: Studies in historical and contemporary social epistemology (pp. 379–396). Westport, CT: Praeger. Nelson, N. (2008). The reading–writing nexus in discourse research. In C. Bazerman (Ed.), Handbook of research on writing: History, society, school, individual, text (pp. 435–450). New York: Erlbaum. Silverstein, M., & Urban, G. (1996). The natural history of discourse. In M. Silverstein & G. Urban (Eds.), Natural histories of discourse (pp. 1–20). Chicago, IL: University of Chicago Press. Spivey, N. N. (1984). Discourse synthesis: Constructing texts in reading and writing. Newark, DE: International Reading Association. Spivey, N. N. (1997). The constructivist metaphor: Reading, writing, and the making of meaning. San Diego, CA: Academic. Tuchman, G. (1994). Historical social science. In N. K. Denzin & Y. S. Lincoln (Eds.), Handbook of qualitative research (pp. 306–323). Beverly Hills, CA: Sage. Turner, S. (1994). The social theory of practices: Tradition, tacit knowledge, presuppositions. Cambridge: Polity. Ulichny, P., & Watson-Gegeo, K. A. (1989). Interactions and authority: The dominant interpretive framework in writing conferences. Discourse Processes, 12, 309–327. Wenger, E. (1998). Communities of practice. New York: Cambridge University Press.
31 On textual silences, large and small Thomas Huckin University of Utah
It is no secret to anyone who teaches writing, researches writing, or simply writes that language conveys meaning not only through the words and images on the page (or screen) but also through their very absence. In the words of Stuart Hall (1985), “Meaning is relational within an ideological system of presences and absences.” If we see writing in dialogic, Bakhtinian terms, it can be said (as do Nystrand, Greene, & Wiemelt, 1993) that “writers and readers work constantly to weigh what is said or what needs to be said against what can be (and perhaps must be) left unsaid in a given textual exchange,” or, as Rommetveit (1974) puts it: “perhaps ellipsis is the prototypical structure of linguistic communication after all” (Nystrand et al., 1993). Rhetoricians, writing scholars, discourse analysts, and others are of course well aware of the power of silence; for example, Perelman and Olbrechts-Tyteca, in their magisterial The New Rhetoric (1969), note that “the importance of presence [i.e., explicit elements] in argumentation has a negative as well as a positive aspect: deliberate suppression of presence is an equally noteworthy phenomenon, deserving of detailed study” (p. 118). Nonetheless, despite a number of book-length treatments of the subject (Picard, 1948; Dauenhauer, 1980; Tannen & Saville-Troike, 1985; Jaworski, 1993; Kalamaras, 1994; Montiglio, 2000; Farmer, 2001; Glenn, 2004), it remains what Glenn has called an “underexamined rhetorical art.” In particular, little has been done in the way of close textual analysis—the kind of analysis that can be readily put to use in writing pedagogy. If silence is so rhetorically powerful, and if its power is so widely recognized, why has so little attention been paid to it? Why, as Glenn puts it, is it so “underexamined”? I think the answer to that intriguing question resides in the simple fact of material availability: it’s difficult to analyze something that’s “not there,” especially for writing researchers and teachers whose stock-in-trade is looking closely at actual words on the page. For this reason, I have devoted much of my work on textual silences over the years to developing a theoretically informed methodological foundation for further research in this area. The pedagogical goal of the work is to sensitize students to the rhetorical power of the unsaid, thus making them more critical readers and more conscientious writers, especially in their ability to construct arguments about important public issues.
420 T. Huckin In order to identify and analyze textual silences, I find it useful to proceed in four steps (cf. Huckin, 2002): First, since background knowledge of the topic and genre under discussion is essential, one should select a text about some familiar topic in some familiar genre. Second, to establish what is typically relevant information about the topic and genre (what might be called the rhetorical universe of discourse), I find it helpful to compile a corpus of texts, whether formal or informal, addressing that topic. (For practical reasons, I usually work with small corpora, as I will show later.) Third, one then returns to the target text and does a close reading of it, making note of those subtopics or genre features that are missing; these are textual silences. Fourth, one does a critical/rhetorical analysis of these textual silences, speculating on how they might influence the intended reader. This step in the process is analogous to what Barton calls identifying “rich features” (Barton, 2002), except that the features are not physically present. Although I have presented them here as discrete, in practice these four steps interact recursively: it is best to go back and forth among them, informing the analysis at all levels. Textual silences are commonly used in benign ways, where there is no intent to deceive the reader. But they can also be used for deception, to hide important information from the reader without good cause, to the advantage of special interests. It is this latter, manipulative type of textual silence that we will focus on here, as it is especially important to the sort of critical reading and rhetorical writing that dominate current composition pedagogy in the United States with its emphasis on civic issues and public discourse. If we define textual silence as “the omission of some piece of information that is pertinent to the topic at hand” (Huckin, 2002), textual silences can be classified into at least six different types according to the textual mechanism that gives rise to them. These include topical silence, conventional (genre-based) silence, discreet silence, lexical silence, implicational silence, and presuppositional silence (see Table 31.1). Textual silences also vary in their transparency, that is, the ease with which they can be detected. The first four listed are what I call covert silences, in that when used manipulatively, their success depends on the reader not knowing what information has been elided. The last two are what I call collaborative silences, because they induce the reader to provide the appropriate information desired by the writer. The following discussion goes through each of these in order, starting with the least transparent type of silence (the topical silence) and working down to the most transparent type (the presuppositional silence). In each case, I will give both benign and manipulative examples, many of the latter from corpus-based case studies. (Note: With apologies to this volume’s international readers, all of the examples given are drawn from current-day American culture. While I am confident that the typology itself has broad, cross-cultural validity, I lack sufficient knowledge of other cultures to confidently offer specific examples in any of them. To help readers uncertain about the American examples, explanatory glosses are provided.)
On textual silences, large and small 421 Table 31.1 Textual silences by type, from least to most transparent Type of silence
Mechanism
Rhetorically benign examples
Rhetorically manipulative examples
Topical silence
Exploits textual coherence
Virtually any text
Congo genocide; international healthcare comparisons
Conventional silence
Exploits genre features
Obituaries, resumes
Advertisements; US history textbooks
Discreet silence
Exploits social sensitivities or security needs
Classified info, proprietary info, tactfulness
Photos of GI coffins; selfcensorship by White House reporters
Lexical silence
Exploits word choice
Hyponymy (“walk” vs. “stroll”), euphemisms
“Pro-choice” vs. “pro-life”; “death tax” vs. “estate tax”
Implicational silence
Exploits the Cooperative Principle (Grice)
Polite suggestions
Actors playing doctors in TV ads; Bush admin public statements in the run-up to war in Iraq
Presuppositional silence
Exploits syllogistic logic, linguistic rules, or cultural norms
“Blessed are the meek”; “Three blood samples were taken”; “I was here before you.”
Opinion poll questions; Powell at the UN
Topical silence Perhaps the most rhetorically potent yet least detectable type of textual silence is the topical silence, where some topic relevant to a larger issue is omitted from discussion. Since few texts include everything that could be said, topical silences are commonplace and usually benign. For example, the length restrictions on the chapters in this volume have no doubt forced authors to omit some material they would otherwise want to have included; such omissions are benign in that they are not designed to promote the writer’s interests over those of his or her readers. Of more interest are the manipulative uses of topical silence—cases where the speaker or writer elides relevant information in a way that surreptitiously disadvantages the listener or reader. Manipulative topical
422 T. Huckin silences are everywhere to be found in public discourse. On the broadest level one could point to a topic such as the Congo genocide, which in the past ten years has killed more than five million people and sexually traumatized millions of women and children yet is seldom even mentioned in the US news media (Phillips & Project Censored, 2006). This glaring textual silence is a manipulative one because it keeps the American public in the dark on a very important world issue, to the benefit of certain economic and political interests (Phillips & Project Censored, 2006). Another example of topical silence concerns international comparisons of healthcare programs. Although the US healthcare system is by far the world’s most expensive, it leaves some 50 million of our citizens uncared for and is outranked by many other countries on such measures as infant mortality, preventable deaths, longevity, etc. Given these unpleasant facts—and the topic’s high profile—Americans would surely be interested in knowing how other countries’ healthcare systems work, especially those countries that outrank us on a variety of measures. Yet the mainstream news media in the United States rarely mention such competing systems. Using the Lexis-Nexus Academic Universe database, I recently put together a corpus of 128 news articles (107,225 words) on “healthcare systems,” representing the totality of such articles published in major US news sources (print, TV, radio) during the two months leading up to the 2008 presidential primaries. None of these 128 articles provided any sort of comprehensive description of any foreign healthcare system; only ten even mentioned any foreign systems, and this only in passing. I consider this a manipulative textual silence because it deprives American citizens of important information, to the advantage of those powerful interests that want to make only minimal changes to the current healthcare system.
Conventional (genre-based) silence A second type of textual silence is the conventional or genre-based silence. These are cases where the genre (or the parent discipline; see Dressen, 2002) dictates that certain types of information be left unsaid. Benign examples include the obituary, which prototypically omits information unfavorable to the deceased, and the resume, which normally excludes marital and family information (at least in the United States). Swales describes cases of benign textual silence in the field report, citing such diverse fields as mycology, petrology, linguistic anthropology, and discourse analysis (Swales, 2004). In all of these cases there is no effort to deceive or mislead the reader. Conventional, genre-based silences can also be used, however, to manipulate an audience; and this is where they lack transparency almost to the same degree as topical silences. Advertisements, for example, routinely leave out information that is not favorable to the product or service being advertised. Political speeches, such as annual state addresses by national leaders, do likewise. Even secondary school history textbooks fall into this
On textual silences, large and small 423 category. In a detailed examination of 12 of the most popular American history textbooks used in high schools across the country, historian James Loewen found that they systematically omitted information that challenges mainstream myths about American glory (Loewen, 2007). For example, American icon Helen Keller is routinely depicted as a remarkable young woman who managed to graduate from college despite being deaf and blind. Nothing is said in these books, however, of her 68 years after college. Keller devoted her long adult life to progressive causes, becoming a powerful advocate for women’s suffrage, workers’ rights, socialism, and opposition to war, yet all of this is elided from the US history textbooks that millions of American teenagers read in school. Loewen argues that this is because Keller’s radical leftism runs counter to the orthodox glorification of American capitalism and militarism that characterizes the genre. American secondary school history textbooks also routinely sanitize the Vietnam War. For most Americans who lived through that time, there are five graphic images that are seared in our memory: the Buddhist monk protesting the war by immolating himself in a Saigon public square; the summary execution at close range of a suspected Viet Cong by a South Vietnamese high official; the bodies at the site of the My Lai massacre, where US troops slaughtered more than 400 Vietnamese civilians; the napalmed Vietnamese girl running naked toward the camera; and the final rooftop evacuation of US embassy personnel in 1975. These photos are all omitted from American secondary school textbooks, replaced by less inflammatory ones such as US soldiers on patrol in rice paddies, B-52 bombers flying high over the North Vietnamese countryside, or naval vessels maneuvering offshore. To the extent that the secondary school US history textbook is a conventionalized genre, as Loewen asserts, these omissions are all genre-based silences.
Discreet silence Another category of textual silence is the discreet silence, where the writer refrains from mentioning sensitive information either to avoid offending the reader, avoid revealing organizational secrets to rival forces, or avoid infringing on another person’s privacy. Benign examples of discreet silence include the withholding of classified information by government officials, of proprietary information by business enterprises, and of personal client information by lawyers, bankers, priests, and physicians. Other benign examples include tactfulness in letters of recommendation, and cultural taboos such as public acknowledgment of racial traits in athletic performance (Entine, 2000). But discreet silences can also be used in manipulative ways. A well-known example is the Bush administration’s executive order prohibiting photos of the military coffins coming back from Iraq and Afghanistan. The administration claimed that such silencing was being done out of respect for these soldiers’ families, but many commentators have expressed skepticism about
424 T. Huckin such claims. They note that during the Vietnam War the photos of returning coffins were a powerful stimulus for antiwar sentiment, and that the recent censoring of such scenes conveniently served the administration’s political purposes by keeping such unpleasantness out of the public eye.
Lexical silence A fourth type of textual silence is the lexical silence, which occurs at the most micro-level of discourse. Lexical silences exploit paradigmatic word choice in that the choice of a particular word excludes other candidates. Consider, for example, cases of hyponymy, where two words are similar in meaning but one is more general than the other, such as walk and stroll. If you described someone as “walking” down the street when they were actually strolling, you would be silencing those features of stroll that do not inhere in walk. Euphemisms are a special case of hyponymy. Consider, for example, the distinction between cheap and inexpensive. Both terms convey the meaning of “low in cost.” But while inexpensive carries only that meaning, cheap has an additional connotation of “poor quality.” Thus, if you euphemized a cheap wine by calling it an inexpensive wine, you would be eliding one of its features. In most cases such lexical exclusions are rhetorically benign. But there are many cases in public discourse where the choice of a term over a competing term can have manipulative effects. For example, in the longstanding debate in the United States about terminating an unwanted pregnancy, one side uses the term pro-choice while the other uses the term pro-life. In the former case, reproductive freedom for the woman is foregrounded, silencing the interests of the fetus. In the latter case, it is the other way around: the term pro-life prioritizes the fetus and effectively silences the woman’s interests. As another example, consider the competing terms, estate tax and death tax. In the United States, there has long been a federal tax levied on large estates upon the death of the estate’s owner—a tax that conservatives have long wanted to do away with in the name of a general tax-cutting ideology. Because of exemptions, however, the estate tax affects only the largest 2 percent or so of estates; the other 98 percent pay no estate tax at all. For decades conservatives could not rally enough support from the general public to eliminate this tax. Then in the late 1990s a well-known conservative wordsmith, Frank Luntz, coined the term death tax as a substitute for estate tax. He knew that since everyone faces death at some time, use of the term death would have more emotional power than the term estate with its elitist overtones (Luntz, 2007). The new term was widely disseminated through conservative outlets in the mass media, and met with rapid success: 70 percent of the American public said they favored eliminating the “death tax,” and in 2001 the US Congress voted to do just that, albeit only for one year. As Luntz says, “Change the name and you change the fortunes” (2007, p. 166). In this case, the original name foregrounded
On textual silences, large and small 425 the fact that the tax affected only the richest estates; the new name (death tax) elides that fact and is thus a manipulative lexical silence. Many lexical items encode metaphors, and in such cases, as Lakoff and Johnson note, the metaphor can hide particular aspects of the world: The very systematicity that allows us to comprehend one aspect of a concept in terms of another (e.g., comprehending an aspect of arguing in terms of battle) will necessarily hide other aspects of the concept. In allowing us to focus on one aspect of a concept . . . a metaphorical concept can keep us from focusing on other aspects of the concept that are inconsistent with that metaphor. (Lakoff & Johnson, 1980, p. 10) Pro-life and death tax are cases in point, as they operate metaphorically and, in so doing, efface key aspects of the underlying issue.
Implicational silence A fifth type of textual silence is implication, defined as “the suggestion of a thought or idea by letting it be inferred from something else, such as a statement, that is more explicit” (American Heritage College Dictionary, 3rd ed.). Implicational silences, or insinuations, being an efficient and often polite way of conveying ideas, are commonplace and benign in everyday discourse. For example, if my wife and I were preparing to go to a dinner party and she said to me, “Are you sure you want to wear that jacket with those pants?,” I would infer that she disapproved of my attire even though she did not say so explicitly. It would be a textbook indirect speech act (Searle, 1979) that exploits the Cooperative Principle (Grice, 1975; Sperber & Wilson, 1986) and its very indirectness would be a polite way of getting her point across (Brown & Levinson, 1978; Leech, 1983). If there is any manipulation involved, it is gentle and benign, since there is no deception in the illocutionary force (Austin, 1962) of her utterance. Such innocent uses of implication are routine in daily communication. But implicational silences can be deployed for manipulative purposes as well, especially in promotional discourses (Bhatia, 2004) such as advertising and politics. In such cases, the advertiser or politician benefits from a distinct feature of implication, its deniability. That is, by insinuating something rather than flatly stating it, an advertiser or politician can convey some idea while reserving the ability to later deny it if necessary. For example, TV ads selling medications commonly feature a spokesman wearing a white jacket and a stethoscope draped around his neck. Many viewers infer from this garb that the spokesman is an actual physician, not an actor. A voice-over saying “Doctors recommend . . .” reinforces the deception. At no time, however, does the spokesman actually claim to be a physician, and if the FTC were to file suit for false advertising, the company running the ad could hide behind this deniability.
426 T. Huckin The most extensive and most consequential use of implicational silence in recent times can be found in public statements by the Bush administration leading up to their 2003 invasion of Iraq. During the summer and fall of 2002, top administration officials including the president himself conducted a virtual propaganda campaign that led the American public to believe (a) that Saddam Hussein was behind the 9/11 attacks; (b) that Saddam had a working relationship with Al Qaeda and its leader, Osama bin Laden; and (c) that Saddam had an arsenal of weapons of mass destruction (WMDs), including possibly nuclear bombs, which made him an imminent threat to the United States. Indeed, it was largely on the basis of these three false beliefs that a fearful American public was persuaded to throw its support behind the invasion. Many opponents of the war accuse the Bush administration of “lying us into a war.” But the three false beliefs that misled the American public resulted not so much from outright lies as from numerous insinuations by top administration officials. For example, in a speech to the United Nations on September 12, 2002, President Bush said: In the attacks on America a year ago, we saw the destructive intentions of our enemies. This threat hides within many nations, including my own. In cells and camps, terrorists are plotting further destruction, and building new bases for their war against civilization. And our greatest fear is that terrorists will find a shortcut to their mad ambitions when an outlaw regime supplies them with the technologies to kill on a massive scale. In one place—in one regime—we find all these dangers, in their most lethal and aggressive forms, exactly the kind of aggressive threat the United Nations was born to confront. Twelve years ago, Iraq invaded Kuwait without provocation. And the regime’s forces were poised to continue their march to seize other countries and their resources. Had Saddam Hussein been appeased instead of stopped, he would have endangered the peace and stability of the world. (Bush, 2002) Though never mentioning nuclear weapons by name, Bush alludes to them by first referring to WMDs (“the technologies to kill on a massive scale”) and then referring to “their most lethal and aggressive forms.” He does so in a context that invites the listener to infer that Iraq is harboring Al Qaeda cells and camps (he refers to them only as “terrorists,” but the implication is clear) and that Saddam’s regime could well supply Al Qaeda with nuclear weapons. This implication is of course contrary to the opinion of many experts, who point out that Saddam had always considered Al Qaeda an enemy of Iraq in general and of his regime in particular. But Bush only insinuates an Al Qaeda/Saddam connection, rather than claim-
On textual silences, large and small 427 ing one. By hedging his insinuation with “our greatest fear is,” Bush avoids making a flat-out claim that could be immediately disputed. The drumbeat of insinuations continued throughout the next few months, climaxing on February 5, 2003 with US Secretary of State Colin Powell’s infamous speech at the United Nations. In this speech, which he later called “the low point in my career,” Powell made numerous false statements about Iraq, including insinuations that Saddam was behind 9/11. Consider this one example, the concluding sentence of his hour-long presentation: “Leaving Saddam Hussein in possession of weapons of mass destruction for a few more months or years is not an option, not in a post- September 11th world” (Powell, 2003). In this one remarkable sentence, Powell manages to combine four themes that the Bush administration had long been pushing. He (1) presupposes that Saddam possesses WMDs, (2) insinuates that Saddam was involved in the 9/11 attacks, (3) implies that Saddam poses a threat to the rest of the world, and (4) implies that the rest of the world cannot wait but must act immediately to stop this threat. All four of these claims were unsupported by the available evidence, which would explain why Powell chose to present them indirectly. Despite overwhelming counterevidence, the administration continued to make insinuations about Saddam, 9/11, and WMDs. And it did so with very little resistance from an obsequious press. One interesting exception occurred on March 26, 2006, when US Secretary of State Condoleezza Rice made this comment on Meet the Press: We’re in Iraq because the United States of America faces a different kind of enemy in a different kind of war. And we have to have a different kind of Middle East if we’re ever going to resolve the, the, the problems of an ideology of hatred that was so great that people flew airplanes into buildings. Iraq was—Saddam Hussein’s Iraq was a threat. (Rice, 2006) By juxtaposing Iraq and people flying airplanes into buildings, Rice insinuates that Iraq was involved in 9/11. Normally, such insinuations would go unchallenged, but on this particular occasion host Tim Russert interjected, “But Saddam was not related to flying airplanes into buildings.” Caught off guard, Rice was forced to preserve her credibility by denying her insinuation: “No, and we have never said that Saddam—Saddam was not related to the events of 9/11.” The administration’s continuing use of implicational silences to insinuate a Saddam–9/11 link had a pronounced effect on the general public. In a Harris poll conducted in June 2004, for example, 69 percent of the randomly selected of 991 American voters polled answered yes to the question, “Do you believe that Saddam Hussein was supporting the terrorist organization al-Qaeda, which attacked the United States on September 11, 2001?” In a February 2006 Zogby poll of 944 US soldiers serving in Iraq,
428 T. Huckin 85 percent said the US mission is mainly “to retaliate for Saddam’s role in the 9–11 attacks.”1 This despite the fact that the 9/11 Commission Report, issued in the spring of 2004, stated that it had found “no credible evidence that Iraq and al Qaeda cooperated on attacks against the United States.”2 Bush himself reportedly admitted to White House journalists, quietly, that there was “no evidence that Saddam was involved in 9–11.”
Presuppositional silence A final type of textual silence is the presuppositional silence. These are cases where the writer omits relevant information on the assumption that it is already known to the reader. Presuppositional silence exploits syllogistic logic, sentence syntax, or cultural norms. For example, the third verse of the Beatitudes is cast in the form of an enthymeme: “Blessed are the meek, for they shall inherit the earth.” The middle term of this enthymeme (“to inherit the earth makes one blessed”) can be omitted because it logically connects the major premise and the conclusion and is thus presupposed. Certain syntactic constructions in English, such as passive voice and nominalization, enable the grammatical agent to be deleted. A sentence such as “Three blood samples were taken” in a lab report or police report elide the identity of the agent who took the samples. Most people would presuppose that it was some technician who did it. Presupposition can be based also on cultural norms. For instance, if a person standing in line said to someone else, “I was here before you,” he or she is probably presupposing a cultural norm of first come/first served (Jalbert, 1994, p. 133). In such cases, the use of presuppositional silence is quite benign. But there are other cases where it is manipulative. Consider this question from a Fox News poll in early 2003: Some people say the Bush administration is too focused on Iraq and not paying enough attention to other threats; other people say the Bush administration is using Iraq as an example to other countries and groups that support terrorists. Which is closer to your opinion? (emphasis added) The term other threats presupposes that Iraq is a “threat.” The term other countries and groups that support terrorists presupposes that Iraq is one such country. Both of these presuppositions were false. Since these unstated, false presuppositions were integral to the Bush administration’s campaign to garner support from the American people for its Iraq war, they qualify as manipulative presuppositional silences. Secretary of State Powell’s speech to the United Nations around the same time was replete with presuppositions. Here is one example: Nothing points more clearly to Saddam Hussein’s dangerous intentions and the threat he poses to all of us than his calculated cruelty to
On textual silences, large and small 429 his own citizens and to his neighbors. Clearly, Saddam Hussein and his regime will stop at nothing until something stops him. (Powell, 2003) Here, Powell presupposes that Saddam has dangerous intentions, that he poses a threat to all of us, and that he is out of control. These were all questionable if not unfounded assumptions that were manipulative because they unethically served the interests of those who wanted to wage war against Saddam at the expense of those who did not. The first two are existential presuppositions with overt referents and therefore not what I consider textual silences. The third, however, is more indirect, requiring some interpretation on the part of the listener to arrive at the underlying presupposition that Powell apparently intends, namely, that Saddam and his regime are wildly out of control in their all-out effort to destroy “all of us.” This latter, then, is a manipulative presuppositional silence.
Conclusion One of the tenets of critical rhetoric (McKerrow, 1989) holds that “Absence is as important as presence in understanding and evaluating symbolic action.” As with the general goals of critical rhetoric, my studies are all motivated by a desire to interrogate power and promote political consciousness and constructive civic action—goals no different than those of Athenian education in the fourth century bc. College-level composition instruction in the United States increasingly shares these concerns, as is clear from such measures as the number of textbooks and anthologies devoted to civic or social issues, the increasing number of service-learning courses, and the large number of scholarly papers on the subject. But the critical reading skills that are prerequisite to good civic writing are often limited by students’ inability to see what’s been left out of a text. In this chapter I have argued that textual silences are rhetorically powerful elements of written discourse that are too often neglected in critical analyses. I have delineated six types of textual silence that differ in their degree of transparency, their mechanisms, and their scope of application. To illustrate how these different types of silence are used manipulatively, I have taken examples from important recent public discourse, especially the Bush administration’s discourse leading up to the 2003 invasion of Iraq. Among other things, attention paid to textual silences should alert composition students to the importance of learning as much as possible about a topic and of incorporating counterarguments into one’s writing. In general, I believe that by cultivating an awareness of textual silences, writing instructors can enhance their teaching of critical reading skills, argumentation, genre knowledge, and subject-matter knowledge, especially of topics important to civic literacy and civic activism, and that such instruction is in keeping with current and future trends in composition pedagogy in the United States and, I believe, in other democratic nations as well.
430 T. Huckin
Notes 1 www.zogby.com/news/ReadNews.dbm?ID=1075. 2 FactCheck.org, June 22, 2004.
References Austin, J. L. (1962). How to do things with words. Oxford: Clarendon. Barton, E. (2002). Inductive discourse analysis. In E. Barton, & G. Stygall (Eds.), Discourse studies in composition. Cresskill, NJ: Hampton. Bhatia, V. K. (2004). Worlds of written discourse: A genre-based view. London: Continuum. Brown, P., & Levinson, S. (1978). Politeness: Some universals in language usage. Cambridge: Cambridge University Press. Bush, G. W. (2002). President’s remarks at the United Nations General Assembly, September 12, 2002. Retrieved from www.whitehouse.gov/news/releases/ 2002/09/20020912-1.html. Center for Public Integrity. (2008). Retrieved from www.publicintegrity.org/ warcard/?source=c0107e3a. Cheney, D. (2002). Interview on “Meet the Press.” September 14, 2002. Dauenhauer, B. (1980). Silence: The phenomenon and its ontological significance. Bloomington, IN: Indiana University Press. Dressen, D. F. (2002). Identifying textual silence in scientific research articles: Recontextualizations of the field account in Geology. Hermes, Journal of Linguistics, 28, 81–107. Entine, J. (2000). Taboo: Why black athletes dominate sports and why we’re afraid to talk about it. New York: Public Affairs. Farmer, F. (2001). Saying and silence: Listening to composition with Bakhtin. Logan, UT: Utah State University Press. Glenn, C. (2004). Unspoken: A rhetoric of silence. Carbondale, IL: Southern Illinois U. Grice, H. P. (1975). Logic and conversation. In P. Cole & J. L. Morgan (Eds.), Syntax and semantics, vol. 3: Speech acts (pp. 41–58). New York: Academic Press. Hall, S. (1985). Signification, representation, ideology: Althusser and the post- structuralist debates. Critical Studies in Mass Communication, 2, 91–114. Huckin, T. (2002). Textual silence and the discourse of homelessness. Discourse & Society, 13(3), 347–372. Jalbert, P. L. (1994). Structures of the “unsaid.” Theory, Culture & Society, 11, 127–160. Jaworski, A. (1993). The power of silence. Newbury Park, CA: Sage. Kalamaras, G. (1994). Reclaiming the tacit dimension. Albany, NY: State University of New York Press. Lakoff, G., & Johnson, M. (1980). Metaphors we live by. Chicago, IL: University of Chicago Press. Lazere, D. (2005). Reading and writing for civic literacy. Boulder, CO: Paradigm. Leech, G. (1983). Principles of pragmatics. London: Longman. Lewis, C., & Reading-Smith, M. (2008). False pretenses. Washington, DC: Center for Public Integrity. Retrieved from www.publicintegrity.org/warcard/?source= c0107e3a.
On textual silences, large and small 431 Loewen, J. (2007). Lies my teacher told me. New York: Touchstone. Luntz, F. (2007. Words that work: It’s not what you say, it’s what people hear. New York: Hyperion. McKerrow, R. (1989). Critical rhetoric: Theory and praxis. Communication Monographs, 56, 91–111. Montiglio, S. (2000). Silence in the land of logos. Princeton, NJ: Princeton University Press. Nystrand, M., Greene, S., & Wiemelt, J. (1993). Where did composition studies come from? An intellectual history. Written Communication, 10(3), 267–333. Perelman, C., & Olbrechts-Tyteca, L. (1969). The new rhetoric: A treatise on argumentation. Notre Dame, IN: Notre Dame University Press. Phillips, P., & Project Censored (2006). Censored 2007: The Top 25 Censored Stories. New York: Seven Stories. Picard, M. (1948). The world of silence. Washington, DC: Gateway 1948/1988. Powell, C. (2003). Transcript of Powell’s UN presentation. Retrieved from http:// edition.cnn.com/2003/US/02/05/sprj.irq.powell.transcript. Rice, C. (2006). Interview on “Meet the Press.” March 26, 2006. Rommetveit, R. (1974). On message structure: A framework for the study of language and communication. London: Wiley. Searle, J. (1979). Speech acts: An essay in the philosophy of language. Cambridge: Cambridge University Press. Sperber, D., & Wilson, D. (1986). Relevance: Communication and cognition. Oxford: Blackwell. Swales, J. (2004). Research genres: Explorations and applications. Cambridge: Cambridge University Press. Tannen, D., & Saville-Troike, M. (1985). Perspectives on silence. Norwood, NJ: Ablex.
Index
5.4 Movement 3–4 abstract 277; levels 299; theory 304 abstraction-decontextualization 19 academic 229, 258; discipline traditional 38; discourse 367, 370; integrity 299; knowledge 340; literacies 336, 375; program quality 125; reading 323; socialization model 336; success 297, 308; support programs 299; virtual communities 328 academic writing 32, 40–1, 44, 181, 218, 261, 284, 328, 339, 344; article 239–41; commitment 261; conventions 241; formal essay 331; performance 335; practices 325, 339; social 325, 335–6; student 373; teaching of 195; thematic content 46; thesis 260 accelerating phase 385, 390 action research 112–13, 284 activity 356; during pause 402; individual 66 activity system 354–5, 359, 361–2, 416; complex 358; maintenance 358; research 417 activity theory 46–7, 224, 353–4, 362, 415 adults illiterate or semi-literate 113, 115 age 138, 147, 154; effect 161, 174–5 alienation 360 alterity 119–20 analysis 128, 241, 407; of captions 136; of forms 20; frameworks 23; of genre system 359; of linguistic features 188; of spoken data 298; of student narratives 366; of teacher comments 371 analytic categories for questions 298, 301
analytic rubric 328; for academic writing 327 apprize option 301–3, 304, 307 archive 309–10; materials 310; posts 326 Argentina 262, 283; universities 285, 292 argument 98, 105, 120, 128, 266, 269, 277, 279–80; academic 368; complete set 277; for and against 315; formulation 267; hierarchical arrangement 274; practices 266; scientific 265; structures 328; written 266 argumentation 266, 328, 373, 400, 419, 429 Aristotle 85 Armenian 99, 102, 107; culture 104; classroom 103; diaspora 101; foreign students 104; identity 105; peers 106; students 97–9 article 255; conventions 245 assessment 62, 199, 284, 287–8, 308; changes 32; diagnostic 73; diversified practices 34; institutionalized practices 336; of planning processes 75; of students 181; targeted 206, 208; of writing 200, 209 Assessment Proforma 201, 204, 205, 206–9 assignment 36, 221, 251, 268, 280, 359, 367, 371; appropriate pattern 279; discussed 300; written 34 asymptote 171–2, 174 attitude to writing 76; changes 365 audience 241–3, 245; awareness 203 Auman, M. E. 215, 217; schema 216 Australia 198, 202–8, 261–2, 292; Writing Project 199–201, 209; universities 297–8, 200, 307 author 341, 345–7; audio-visible 313–23; awareness 311; construct
Index 433 25; foreign 53; invisibility 316; visibility 316–17, 320–2 authoritative discourse unnegotiable 299 background knowledge 298–9, 420; lacking 368; repair 300, 302; understanding 267 Baddeley, A. 310 Bakhtin, M. M. 46, 111, 114, 299 balance 26; teaching and research 37 band-pass filters 390 Bazerman, Charles 47, 93, 212, 344, 353, 357 Beaufort, A. 367, 375 Benkler, Y. 343 Bereiter, C. 316, 365 Berninger, V. 166, 174 Bhatia, V. K. 51 Biagoli, M. 340–1 biographical information 90, 93, 141 blended learning 325, 334 Bloom, Benjamin 329 Bologna process 31, 41n1, 262 Boscolo, P. 168, 170, 174, 177 Bourdin, B. and Fayol, M. 166 branch theories 10; of practical writing 8 Brandt, D. 310–13 Bratcher, S. and Stroble, E. J. 213 Brazilian society socioeconomic structure 116; education laws 45; Portuguese language 120 break-point 384–5 bridging 103–5, 107 bursts 386–7, 389 Caccamise, D. J. 170 Cameron, D. 112 captions 136, 147; context 138; frames 140, 143; restriction 137 Carroll, L. 371 Castillian 72, 122 Catalan 151–3, 159; vocabulary 155 censorship 103, 340–1 Central Committee of the Communist Party 5 central government 72 Central and South American countries 122 challenges 240, 242–3, 246 changes of grammatical person 149n4 chart 222; example 216; R/D/F & e 215–17 Chenoweth, N. and Hayes, J. R. 166
children 136, 177–8; requirements 149 China Writing Society 3–6 Chinese characters 13; tones 13n2 Chinese children 152–3, 156–7, 161, 163 Chinese language and literature 11, 152; archaic 4; Mandarin 153; terms 3; writing studies 5, 11 Chiseri-Strater, E. 369, 375 Christensen, C. A. 167 citations 329, 345, 412, 414 classification typology 188, 191 classroom 234; characteristics 186; contexts 58, 369; contradictions 361; discourse community 368; discussion 298, 314; environment 65, 366; good will 102; instruction 212; intervention 182; meaningful input 368; norms 244; observation 214, 366–7; practical norms 244; practices 94n1, 187, 239; questionnaires 184; radio project 224; rules changed 362; teachers 184, 200; teaching 243; texts 212; tools 122; traditional organization 186; writing practice 181 closures 141–2, 144 coding 22, 123, 300, 326, 425; cipher 116; color-coding 214; cross-case 240; dual coding theory 310; open 239; papers 269; scheme 170, 328; systems 310; visual images 300 cognitive-load theory 310 cognitive process model 413 cognitive processes 1, 127, 181–2, 328–9, 331, 394, 397; in adult writing 166; lack of clarity 21; research 416; temporal patterns 410–11; in text production 403–4; of writing 199, 413 cognitive science 409 Cognitive Self-Regulation Instruction 77–8 coherence 74–7, 90, 140, 202, 203, 208, 369; markers 73, 78; textual 194 cohesion 178, 194, 202–4, 208; analyses 272; building 201, 209; degree of 279; textual 328 cohesive 277; profiles 278; structure 280; terms comparison 274; terms per sentence 272–3, 278 collaboration 343, 346
434 Index collaborative activity 224, 226–8; authorship 343; knowledge making 339 collaborative learning 297; community 330, 336; literacies 232; online inquiry 335; semi-collaborative 222; silences 420; structure 347; webbased writing 339 collaborators 245, 246; cross-level 241 college 238; instruction to promote learning 283; literacy practices 286; texts 287; writing abilities 373; writing from sources 286 colonization 86; colonialism 94 communication 46, 88, 118, 138, 225, 289, 414; business program 299; computer-mediated 325, 328; cultural 114; daily 425; difficulty 119; face-to-face 115; intercultural 103, 106; multiple modes 225; sciences 314; situated perspective 357, 414; sociocultural 45; spontaneous verbal 114; unity 407; about writing 6 communicative 47; metacommunicative ability 312–13; purpose 195; semantic focus 123; skills 266; task 191 community 114, 246, 414; academic 33; building 126; colleges 257, 258; of experts 129, 228; online 329; problems 112; professional 241, 373; radio station 226; standards 266 compare and contrast 188; essays 219–21 competence 20, 61, 64, 181, 233, 374; academic 40; development 71, 75; macro-competence 61; obscured 23; of students 183; teachable 59; textual 184; unwanted transfer 23 composition 6, 10, 45, 123, 167, 174, 212, 370; college level instruction 429; curriculum 212; early leading line 89; length and quality 166; model 200, 203–6, 208–9; pedagogy 420; scholars 86; structure 46; studies 406, 408, 414; Summary of Methods 7; theory 18; stages 182 comprehension 127–8, 154–6; speed tests 123 computer conferences 325–7, 336; online 332–3 computer literacy 9, 13; skills 267 concurrent self-report method 74 conferences 6, 128–30, 252–3, 256,
326, 330; genre 44; literacy skills 8, 24, 125 conflicts 98, 115, 342, 354, 355; of interest 346; resolution 105; social 119 constraint 241; on ideas generation 174 construction 410; of motives 358; of texts 141 content 63, 167–8, 187; analysis 17; areas 60; coherent 176; familiar 137; generation 400; knowledge 202; management systems 361 context 27, 48, 98, 138, 238, 240, 312, 352–3, 357, 361, 413; lamination 362; of learning 311; multiple 354; production 63; real world 75, 267; sociopolitical 85; of use 164; for writing 184 contextualization 19–21; of discourse 51 conventional silence 420, 422 conversation 312; analysis 227; with the author 311; style 310 Cook-Gumperz, J. 115 copying 299, 308 copyright 346–7 corruption 103; educational 98; postSoviet 107 counternormative setting 240, 246 courses 101, 103, 106; English language 261; evaluations 326; materials 99; in narrative 100 covert silences 420 creative commons license 347 creative thinking 91, 370, 373 critical reading 420; skills 429 critical rhetoric 429; analysis 92 critical thinking 360, 373; skills 329, 368 Crosswhite, James 98 Cuban, Larry 40 cues 381; filtered out 328 cultural background 298, 309, 369 Cultural Revolution 5–6 culture 87, 118, 122, 177, 313; academic 117, 126; development 98; difference 284; influence 12, 46; logic of origins 342, 348; sessions 114–15; taboos 423; written 117 Cunningham, Ward 342–3 curriculum 222; elements 254; texts 212 cursor 398–9, 403–4; movement 383 cusum 390; break-point 385, 386–7; statistic 384
Index 435 data collection 113 data terms 272, 275–6, 278 decelerating phase 385, 390 deconstruction 201, 208 De La Paz, S. and Graham, S. 166 democratization or exclusion 61 Deng Yingchao 6 decentralization effort 137 derivation task 155–6; morphology 154 description 45, 185; definition 187–8, 191, 193; functions 27 descriptive analysis 52 development 12; factor 149; patterns 409; problems 168; in reading 369; of writing abilities 365, 367, 369, 375 deviant responses 156 deviations from original 410 dialectical contradictions 354, 361 dialogic 46; learning 297, 298, 299, 307; relationship 314; teacher–peers interactions 373 dictating 168, 172; orally 178 Didactext research group 181, 196n1 didactic 17–10, 21, 23–4, 60–1; intervention 184, 187–8, 191, 193–4; of language practices 292; motivations 65; perspective 22; sequence 62–3, 67, 181–2, 184–5, 194; of writing 59 didacticians 23; of different disciplines 18 difficulties 140, 157, 164, 208, 366, 368–9; in language learning 152; in learning 23, 116; reading and writing 283; of revision 135 digital 50, 309; code 340, 346, 348; images 225; literacy 9; technologies 339; texts 328; wiki codes 339; writing space 342, 348 diglossia 153, 163 directional terms 272, 276, 278 disciplinary 21–2, 344; architecture 3; contradictions 360; communities 414; development 344; frames 23; genre 24; knowledge making 343 discipline 27, 341; discursive space 126; specific knowledge 369; writing expertise 367 discourse 44, 47, 62, 336, 409; academic goals 305; analysis 51, 123, 325, 328; communities 233, 246; control 304, 341; critical 45, 48, 51; genres 111, 114, 117, 119, 228;
meanings 302; metadiscourse 312, 314; practices 415; promotional 425; synthesis 410; written 227, 299 discreet silence 420, 423 discursive 357; community 47; domains 50; features 328; heterogeneity 25; practices 48, 50, 53, 340; response 256 discussion 268, 297, 302, 304, 307, 315, 413, 416; across disciplines 256; class 369; cross-national 253; electronic forums 325; on European listservs 256; face-to-face 138, 330–1; online 326, 334; section 277; spaces 343; tabs 343, 345; threads 326 distracters 178 diversity 20, 119, 120 division of labor 354–5, 362 documents 359–61; written analysis 406 Donahue, Christiane 254–5 drafts 251, 379; final 181; group discussion 39; multiple 38; scripts for interviews 227–9 dropout rate 125, 283; solutions 24 economic 299; assignment 300; benefits 9; and political collapse 91 editing 181, 225; images 231; oral texts 234; process 230; software 226; system 381; tabs 343 editor 341, 345 education 6, 44, 112, 118, 313; authorities 88; bourgeois 91; macrolevel policies 31; meritocratic 360; practical 91; project failure 91 educational 93, 122, 314; context 67, 71–2, 79; foundations 94; intervention 22; knowledge 299; reforms 85, 92; researchers 309; research 53 educators 129; preparation of 124 effects of student language 316–18 eight-block framework 7 e-learning management suite 326 elementary de-trending 384; procedure 383 elementary school 46, 191; curriculum 87; education 92, 181; teachers 237; textbooks 89 ellipsis 419 Emig, J. 365, 375 empowerment 233; through critical literacy 233
436 Index enculturation 360; into disciplinary practices 374; of students 370 engagement 297, 330; audience 208; with authors 322; in course activities 368; reader 208; of student 34, 59, 91, 218, 220, 284; teacher 201, 372; with writing 373 Engeström, Y. 354–5, 358, 415–16 English 44, 46, 50, 86, 97, 177, 230, 255–6; fluency 104; learners 213, 224, 226–9; non-native speakers 253; punctuation 105; writing 252 English as a second language 368; classroom 233 English Journal 237–8 entextualization 408, 412–14, 416 epistemic 279; dimension 268; generality 269; profiles 278; status of claims 274; structure 280 epistemic level 274, 277, 279; of claim 269–70; of statements 278 epistemology 406; analysis 24; frames 23; order 344–5, 347; politics 343 equilibrium theory 300 error 398–9; of accent marking 194; correction 401; detection 402; realization 402 essays 176, 177, 215, 260; form 222; graded 219; longer 177; opening sentences 215; personal 217; writing 169, 175, 218 Essential Competences for Basic Education 59 ethics 118, 238 ethnic 111, 118–19, 168; background 367; conflicts 98; ethnocentric 104 ethnographic 52; ethnomethodology 362; field notes 227; microethnographic perspective 227; research 112, 326; studies 111 ethos 118–19 Europe 257; cross-national writing research 256; organizations of writing teachers 254; Western European culture 117 European Union 72, 98; Commission research project 309; Credit Transfer System 32, 41n2; Qualification Framework 41 European Writing Centers Association 253, 256 evaluation study 33–5 evidence 266; formation 265 exam 39, 124, 187, 284, 288, 313, 315–16, 320, 370; college entrance
45; final 34; licensing 262; oral 123; performance 3; related work 36; university admission 124; writing 123, 260 examples 188, 208; for arguments 329 exclusions 424 expanded mental model 105 expectations 104, 287, 291, 370; college-level 285; keeping pace with ability 76; macro-level 32; of research 37; writing 220 experience with computers 169 experimental reports 260 explanation 196, 298, 301; techniques 190 explanatory resources 188, 194 eye movement 394, 396–7, 400–2, 404; data 395; during pauses 402–3; keystroke data 397–8, 403; motor plans 398; tracking 395, 404 Faigley, L. 409 Fairclough, N. 48, 111, 329 feedback 31, 34–9, 41, 64–5, 72, 227, 244, 246, 285, 287, 288, 290–1, 326, 328, 370, 374; fear of 243; instructor 268, 371; kinaesthetic 399; marginal comments 288; mental and visual loops 379; model 41; peer 36; practice changes 40; regular 32; social 77; teacher 31, 72–3, 206; visual 399; on writing 372; written 35 Feyerabend, P. 315, 318–19 film 114–15 filter 178; outputs 390 First Amendment 103, 106 first language 260–1; morphological richness 152 first person 138–9, 191; avoidance 137; point of view 362; writing 144 fixation 395–6, 399, 400–1, 403–4; patterns 398; typing-concurrent 397 flip-flop phase 389 follow-up 32, 34–5 formulation 378–9; phases 378, 380 fraudulent claims 342 Frederiksen, C. 410 freedom 87; of academic debate 5; of speech 106 Freire, P. 112, 114 funding 252, 254, 285, 346, 366 Ganobczik-Willams, L. 259 Garfinkle, H. 362
Index 437 generic structures 48, 50; characteristics 53 genre 46–8, 50, 60, 65, 73, 91, 136, 200, 202–3, 341–2, 348, 357, 358, 361, 414, 420; academic 24, 45; analysis 44, 51–2; awareness gains 368; characterization 53; complex 118; concept 356; conventions 279, 423; discipline 260; and discourses 20; ecologies 358; evolution of 11; instructional 50; investigation focus 51; knowledge 278, 280, 373, 429; legal 39; literary 20; methodologies 198; modern practical 12; practices 348; primary and secondary 114–15, 117–18; as social action 353, 356–7; theory 339–40, 362; variety of 222 genre studies 44–6, 48–50, 53 genre systems 353–4, 361; analysis 359; of peer review 344 GNU Free Documentation License 343–4, 346–7 goals 7, 254, 292; of instruction 280 Goody, J. 115–16, 234 grade 316–17, 320, 322, 326; average 316; distribution 42n3; scale 316; system 32, 220 graduate courses 260; in linguistics and education 44–5 grammatical structures 45, 123; analyses 45; dimensions 202, 207, 209; person used 138 graphomotoric level 379 guidelines 285, 287, 326; government 123 Haas, C. 367, 375 Halliday, M. A. K. 48, 298, 311 handwriting 169, 171–2, 177–8; drafts 231; essays 167; practice 167; problems 166 Hasan, R. 298, 300–1 Haswell, R. H. 372 Hawthorne, R. K. 213, 221 Hayes and Flower cognitive model of writing 411, 413 hegemony 115, 117–20 Herrington, J. and Curtis, M. 369, 375 heuristic 212; exploration 302; reading filter 401; writing 24, 181 higher education 124–5, 251, 286; classrooms 325; cultural characteristics 32; oral culture 99; Romanian 98; settings 328; social commitment to 126
Hillocks, George 90 history 38, 41; research 10; studies 12; of writing research 408 horizontal system 11 Hunt, K. 409 hybridism 118, 225 hypertext 407; networks 353 hypotheses generation 366 idea generation 168–70, 174, 177; in developing writers 178; oral 177; process 167 idea organization 215 ideas 172–3, 410; categories 176; development 174, 215–17; dictated 169, 171; limited 170; for publication 243; review 216 ideational function 48; fluency 170 identity 86, 229, 232–3, 336; construction 25; discoursal 334–5; new 228; rehearsal 230 ideology 11, 88, 93, 117–18, 120, 128; appropriation 90; colonization 86; dominant 91; function 46; indoctrination 87 illiterate 13, 116; adults 113, 118, 119 illiterate–literate continuum 113, 115, 118–19 illustrations 136 image 25, 26; global 86; photographic 155; text dependence 142; of voice 234 implicational silence 420, 425–7 indefinite article 151, 153, 156, 164; plural 157 individual 91; initiative 91; production 136; responsibility 148 inferential connections 410 inflectional morphology 153, 154 informal discussion 366, 368–9 information 143, 309; age 13; delivery 310; density 328; domain-specific 151; hidden 420; introduced 146; missing 142; organization 140; personal 23, 141; required 127; retrieval 177; technology 38, 410 initial anchorage 141–2, 144 insinuations 425, 426 institutional 285–6; constraints 292; contradictions 360; mission in higher learning 255; policies 129–30, 131n4; practices 348; pressures 262 Institutional Educational Project 123
438 Index instructional 65, 222, 290; benefits 309; interventions 202; mediation 217, 220; models 127; programs 71; quality 214; texts 212–13; time constraints 231 instructions 212; for authors 345–6; essay writing 169–70; national experiences 284; oral mode 116–17; on revision skills 78; Soviet-style 99 intellectual 102, 115, 328; confidence 330; property 341, 344, 346–7 interactions 144–6, 172; with peers 367 interactive 11, 105; learning 297; network 9 international 252, 408; perspective 252; preliminary survey 259, 262 international publishing 38; team of editors 262 interpersonal function 48 interplay of writing and reading phases 379 interpretation 301, 312, 407; key skills 368; limited 136; in literacy 323; texts 93 intersecting continua 111, 116, 119 intersubjectivity 312, 322; concept 311; concept in literacy 310, 322; links 360; modality 330; processes 313 intertextuality 360–1, 369, 407 intervention 65, 75–7, 129, 183, 185; evaluation 72; mechanisms 79; model 194; variables 390 interview 113, 214, 230–1, 234, 255, 288, 362, 367–71, 373; data 360; in depth 239; online 330, 334; preparation 228; questions 293, 294n3; recorded 224, 228; semistructured 184, 285, 326, 366, 373; spoken 231; with students 66; transcripts 416; TV 313, 314 interviewees 286–8, 291 introduction 193, 277, 279 invisible author 313–18, 320–1 invitational summer institutes 238–9, 246 involvement 114, 115; administrators 261; across disciplines 262 Italian 177; university 314 IText working group 339 Izhikevich, Eugene M. 345 Jewitt, C. 311 journal 252; literature 238 journalism 105, 106
Kalman smoother 390 Kaplan, A. 406 kernel assertion 170–1 keyboard 169, 177, 178, 395, 400; typing 169 keyboard-gazers 397, 398; eye movements 399 keystroke 396, 398, 400, 402; activity 396; data 395; logging 413; motor plans 398; record 416 key terms 269, 277 knowledge 265, 357; access 182, 184; application 266; building process 234; communities 23; developed socially 356; from emerging disciplines 10; making 343, 348; metaknowledge 73–5, 79; multiple domains 374; organization 40; personal 176; sharing 347; system on writing 4; and thought 128; used in writing 408 laboratory manual 268, 279; lab protocols 347 language 44, 48, 88, 111–12, 151–3, 161, 419; actions 46; Arts education 237; developing skills 230; dimensions of 59; first 157, 159; foreign 53; functional aspects 182; group 156; and identity 118; international 256; learning 224; metalanguage 73, 207; metalinguistic consciousness 60; native 87; nonnative 255; patterns 279; poststructuralist theories 49; practices 18; professionals 260–1; with rich morphology 163; as shared capital 20; second 163; specialized 127; spoken 232; of students 316; studies 87, 122; systemic-functional approach 51; textbooks 183; typological distance 152; used at school 61; uses 50 language and literacy 225; contextualized 227; development 232; reading abilities 159 language teachers 260; Netherlands 261 Latin America 126; context 128; universities 292 law 123–4; faculty 39, 41; legal language 39; recent 125; students 39 learner 310; engagement 26; community of practice 326; online academic settings 335
Index 439 learning 26, 34, 39, 357; academic writing 195; assessment 252; computer-mediated 322; constructivist theory 311; difficulties 297; and doing 20; environments 298; impediment 298; inferential 368; interpersonal 104; Management System 38, 41; materials 323; needs 289; opportunities missed 287; practices 21; process 102; reading and writing 63, 68, 123, 260; scenarios 308; traditional environments 362; transfer 373; transformative 107 learning difficulties 79; students 72–7 lecturer 301–3; lectures 36; strategies 304–7 Legacy Study 239 Lévi-Strauss, C. interview 318–19 lexical cohesion 269, 272; analysis 271 lexicogrammar 48, 298; analysis 304; indicators 329; textualization 51 linear filter 385 linguistics 50; analysis 298; applied 44, 50; communication 419; deficiency model 24; metalinguistic activity 115, 148; perspective 200; practices 232; programs 49; register 135; research questions 25; resources 188, 194; strategies 297; studies 59; systemicfunctional theory 48, 298; understanding 201 links 11; hyperlinks 9; intertextual 360–1; speech-writing 152, 159, 163; textual 407 lists 140–3 literacy 111–13, 116, 119, 199, 234, 292, 312; academic 61, 65; academic research 283; activities 90; administrative 8; classes 72; curriculum 92; demands 369; digital era 322; education journals 237; failures 323; functional knowledge of 164; learning research 224; legislative 8; modern practical 12; paths 127; personal 375; poor levels 58; reformers 93; restricted 115, 120; tasks 290, 292; teachers 77; unsupported 292; urban 119; written 201, 255 literacy development 72; key knowledge 313 literacy education 85, 87–8, 91, 123, 286; democratic 93–4; embedding
93; previous 153; state assessments 208 literacy practices 182–3, 285, 362; complex 115 literacy restricted-full continuum 111, 115–20 literacy skills 199, 368, 371; academic 284 literary analysis 46, 215, 218 literary experience 159 literary studies 4, 8, 122; education focus 92 literate 369; ethics 119; exchanges 312 literature 222; reviews 47, 260 locative-derivative 152–3, 156, 164; formation 151, 155, 163 logging program 381, 395 low-pass filters 388, 390 Lunsford study 373, 376 McCarthy, Lucille 366 McLeod, Susan 254; survey 258 manipulative silence 422, 424–5; discreet 423; presuppositional 428; textual 420; topical 421 Mapping Project 253–4; website 255 material 300; pivot 230; readily available 252 Mayer, R. E. 310 meaning 48, 232, 289, 307, 312, 357, 419; adjusted 148; appropriation 302; beyond words 311; explorations of 297; lexical 163; making 233–4, 322; and practices 63; into text 22 media 309–10; genres 50; mass media 5; mainstream news 422; richness lacking 328 mediated activity 225, 307, 354 mediated mastery 97, 99, 106–7, 108n3 mediation 102, 107 mediational triangle 354–5 Mediawiki 339–40, 345; software code 344 memory requirements 176–7; collective 87; metamemorial search 169–70, 174 mental schemata 184, 408; underlying 412 mentor 226, 228, 231, 244, 308, 371 methodology 201, 277, 406, 408, 411; procedures 53, 268 Miller, C. R. 46–7, 353, 357 mimicry 299, 307, 308 minority groups 115; literacies 225 Miraglia, Eric and McLeod, Susan 252
440 Index mirroring 97, 99, 102, 104 modalities 151, 226, 234, 329; modification 161 mode 172; of actualization 20 modeling 78, 245; potential questions 228 modification 159, 161, 170–1; of determiner 157 monitor-gazer 397, 399, 404; experienced 400; fixations 399 monitoring 402, 404; process 379; pressures 335 Moodle learning platform 319 Moroccan children 152–3, 156–7, 159, 161, 163 mother tongue 61; erudition 122 motivation 65, 74–7, 79, 177, 206–7, 243, 292; contradictory 360; loss 6, 39; negative 73 multidisciplinary studies 13; science 267 multimedia 310, 323; authors 313; instructional content 314; learning 311, 322; simulations 361 multimodal 225, 311, 322–3; communication 311; instructional content 314; interactions 233; learning 310, 322; literacies 224, 232; simulations of professional contexts 361 narrative texts 73, 146; implication sequence 305 national goals 260; accountability 126; evaluation survey 35 national liberation 87, 93 National Writing Project 213, 237–40, 246–7; colleagues 244; context 246; influences 239, 243; resources 243, 245 nationalism 101, 103–4; patriotism 4 needs 356; economic and social 9; educational 292; for literacy teaching 292; modern social 9; rhetorical 274; of secondary schooling 200; underlying 130; user assessment 348 network 9, 102, 354, 407; for asking questions 301; cross-institutional 261; international 253; mediated 103; language teachers 129; of writing researchers and teachers 253, 262; opportunities 262; virtual learning 325, 328–9 new literacy 224–6, 233; studies 336 noise process unobserved 387
Nolen, S. 311–13, 319 non-target-like responses 157 North America 292; educational context 77; writing research 353, 356, 358 Norway 32; education 33, 36, 40; students 42n3 number inflection 151–7, 162 number marking 153, 163 number of ideas 172–2, 173; limitation 177 number of quotes 316–18, 321, 322 number of written words 316–17, 320, 321–2 observation 183–4, 408; of teaching writing 200 older children 136–43, 144–9, 171–2; Chinese 157 Ong, W. 98 online 38, 335, 373; dissertations 49; forum 326; lab notebooks 346; multimedia simulations 354; post 332–3; written communication 328 open-access policies 5, 284 OpenWetWare 340, 345–8 oral 174, 312; descriptions 161; dictation 166; discourses 227, 407; generation of ideas 169, 174; interactions 228; language 153, 230, 234, 414; narratives 51; oral–literate dichotomy 312 organization 221; of discourse 51; of meaning 298; of texts 187–8, 191, 194 oral–written continuum 111, 116 organizational structures 193, 212; change 357; components 274; patterns 410; strategy 221 orthography 177 Paivio, A. 310 paragraph 209, 217, 402; accordionparagraph 216, 219; boundaries 400–1; connection 194; generation 214 paratextual features 191–2; cues 187, 195 Paré, A. 341, 344 participant 168, 202, 225, 227, 239, 315, 319, 328, 348, 354–5, 362, 372–3; Chinese 152; congruent 304; distribution 154; foreign 107; interviews 242; observers 326; schools 183
Index 441 participation 90–1, 115, 238, 369; refusal 99; social 374 pause 401–3; duration 400 Paxton, R. 312–13, 319 pedagogy 21, 67, 90–1, 198, 208; benefits 309; changes 32, 34, 41; devices 58, 64; goals 202, 224; implications 181; intervention 63; motivations 65; orientation of classroom 186; restructuring 124; tool 38; totalitarian 98; workplace 354 peer 106, 330, 335; community 266; feedback 268; peer-to-peer learning 59; production 343; response 244; successful 76 peer production 347–8; knowledge 344 peer review 241–2, 339–44, 346–7; anonymous 345; double blind 340; epistemological order 345 peer reviewed publications 341; article 239; encyclopedia 345 performance 72, 76, 156; changes 365; forecasting 391n3; at each task 157; oral 232; pressure 335 personal attitude 93 personal style 92, 310 phases of text production 378; models 390; top-to-bottom 389 phenomenological 362; genre theory 354 photocopies 127–8 plagiarism 299 planning 76, 78, 181–2, 184, 400, 403; formal 74; intervention 76; phases 378, 380; strategies 75 plural 156–7; descriptions 159; expression 151; indefinite article 163; morpheme 163 plurality 155–6; representation 164 Poland 85, 92–3; language textbook 88; literacy education 94; Polish 87–8; real-socialist 90; students and teachers 91 policy 130, 213; directives 58; makers 375 political 86, 90, 348, 406; context 85, 88, 91–3; documents 35; genres 50; guidelines and regulations 123; level 33; re-education 5; reform expectations 40; science course 314; speeches 422; transition 88, 92 Popper, Karl 314–15, 318 Porter, Tara 254, 262 portfolio 34–5, 36; assessment 38; practices 31–3
Portugal 44, 46, 50; classroom 61; education 59; language 58 position 242, 246, 247, 329, 407; revising 148 postmodern 119–20; development 407 post-social 415 poststructural developments 407 post-totalitarian states 98–9; culture 97 power 116, 240; counter-cultural 107; imbalance 243; institutional relationship 283, 334; normalizing 86, 341; relations 336; silence 419 practical concerns 240–1; challenges 243 prediction consequence 306; structures 304; narrative sequence 307 preferences 143, 164 prejudice 120, 342 presuppositional silence 420, 428 primary school 46, 58, 74, 117, 122, 198, 201, 209; teaching 72, 124 print 340; literacy 366 problem-solving 411; research 413 process 412; measures 74, 77–8; perspective 187 processing 166; and planning period 398 process-tracing 412; research 413 production 154; collective 136; contexts 181; focused teaching 77; of knowledge 128; oriented reading 379; of specific genres 59; techniques 24 productivity 73, 76; gains 75 professional development 124, 284; network 237; pathways 359; programs for educators 238; of teachers 68; writing in 24, 239 program 254; interpretation 218; leadership 254; models 262; presentation 215 progression 382; analysis 381, 391n1; graphs 382–4, 385–6, 388 projects 34–5, 219; failures 94; reports 381 psycholinguistics 17; processes 404 psychology 12, 199; cognitive 17, 19, 59, 60; genetic 17, 19; psychoanalytic perspective 26; psychological stability 356 publication 244, 247, 339; article 243; electronic 252; graduate students 262; obstacles 246; preparation 260; rejection 342; samizdat 87; sponsored 242; teacher-authors 238, 240, 242; in traditional journals 345
442 Index public discourse 420, 422, 429; professional 238 public relations 101–2 punctuation 146–7, 148–9; errors 194 qualitative studies 52, 201, 227; analyses 157 quality 125; improvement 166; of publications 342; Reform 31–41; and structure of texts 74; of student writing 62; of texts 77–8 questions 188, 307, 429; categories for asking 300; interrogating concepts 106; posing questions 267, 298 radio broadcast 231; audio file 230; local 228; practices 225; record and edit 232; recording equipment 226; technologies 232, 234 readers 137; struggling 226 reading 45, 226, 289; back through 402; comparison 286; comprehension 311; and editing 234; episodes 401; existing text 395; filter 396; habits in students 72; less 37; and literature 87; objectives 127; order 389; phase 378; practices 123; processes 379; required 288, 315; social interaction 322; sustained activity 401; text 383; source-oriented 379 reading and writing 129, 292, 358; assignments 285; in the disciplines 286, 288; groups 369; in higher education 126; practices 127; REDLEES network 129, 131n5, 131n6; support 284; transferable skills 290; transposition 61 real-socialism 88, 93–4 reasoning 280, 370; predictive 305 recasting of questions 300, 303–4, 307 recognition in society 120; scholarly 341 referees 215, 341, 343, 345, 346–7 reform 35; demands 41; documents 33, 40; in linguistic expression 4; topdown 40 reformulation 25, 188; of instructions 137 registers 195; diverse 120; formal 368; informal-formal 111, 114 relationship 48, 73, 92, 241, 244; author–reader 312; behaviors 328; of college material 370; international 103–6; to language or learning 21;
social 45, 50; teacher–student 61; with texts 329 relevant information 147, 287; omission 428 religious 94, 116; genres 50; topics 123 remedial subject 122 repetition 279; avoidance 144–6 representation 28n3; competing 402; interactions 227; mental 311, 383, 408, 410; modes 322; obstacle 66; self 231; spoken 156; of writing 22 reputation 346; of contributors 344; protection 340–1; ranking mechanisms 347; scholarly 343 research 9, 47, 67, 126, 202, 406; agency 27, 28n1; articles 25, 49; contexts 67; disciplines 17; education 40, 94, 112; empirical 63–4, 266; evaluation study 31; frameworks 13; funding 341; genres 25; intensive institutions 257; labs and groups 346; methods 52, 112; principles 58; projects 127, 309; text-oriented 7; trends 23; socially oriented 414; theoretical perspectives 52; T-unit analysis 409 researchers 22, 44, 124, 265, 416; academic 118; creativity 170; in-training 26; marginalization 342; new 347 research writing 24–5; evolution 22, 25 resistance 86, 99, 108n5, 123, 354 resources 245–6, 261, 311; formulaic 116; human 107; lack of 181; sharing 329, 343, 346, 356 respondents 259–60, 263 reversed-trend phase 389 revision 75, 78, 135–7, 140, 144, 147–8, 181–2, 184, 220–1, 284, 378–9; ability 74; deep-level 78; opportunities 369; phases 378, 380; practice 138; process 139, 143, 146, 149; recorded 169; strategies 378; temporal sequence 381 rhetoric 11, 85–6, 88, 99; epideictic 98, 104; new 47; official 92; social 45–6, 51; studies 4–5 rhetorical 243, 343, 406; analysis 266; challenges 241; concerns 240, 265; descriptors 90, 330; dimension 268; forms 312; indicators 329; knowledge 279, 373, 394; practices 414; purpose 53; spaces 246; strategies 328; structures 410; tasks 182; writing 420; writing choices 368
Index 443 rhizome 407 Rogers, P. M. 373 Rogoff, B. 326 rote learning 116–17 Royal Society of London 340 rural communities 118; culture 11 rural–urban continuum 111, 116, 119 Russell, David 93, 252, 359, 374 saccades 395–6; backward 402; forward 401 sample student work 101, 103, 104, 106, 268 scaffold 77, 208, 230, 279, 290, 306, 370 Scardamalia, M. 316, 365 Scholarpedia 340, 345, 346 scholars 253, 256, 263, 313–14; Brazilian 45, 53; Chinese 5; European 46, 255; focus groups 255; Fulbright 97; Italian 315; North American 47; Polish 86; social sciences 313; writing 5–7 scholarship 262, 283; transnational 93 school 20–2, 46, 50, 67, 74, 86, 117, 122, 124, 207, 376; children 13; curriculum 87; genres 50, 93; learning 91; practices 23, 135; primary 58, 198, 201, 209; secondary 128, 209; teaching 72; writing practice 65 Schutz, Alfred 353, 355–6, 357, 362 science 128; education 265, 279; philosophers 314; a priori 299 scientific 265, 267, 280; genre 268; literacy 8; open information 347; thinking 125 scientific writing 265, 328; social dimension 367 scripts 234; collaboration 230 second language 151–2, 157; acquisition 159, 368; learners 215, 218; research 225; students 297, 299, 308 secondary socialization 106; culturebound 107 seeing oneself as writer 244, 245, 246 self-efficacy 9, 73–7, 79, 220, 330; resistance to intervention 76; as a writer 368 self-related concerns 60, 71–2, 76, 240–2; challenge 241, 243; selfconsciousness 230, 335 semantic 48, 298, 409; aspects 184;
cohesive dimension 269; level 402; organization 409 semiotic practices 27; social 45, 48, 311 Semprini, A. 315, 319 sentence 204, 207; boundaries 277, 401; combining 409; complex 170; construction 202, 207–8; first 272; fluency 208; introductory 279; structuring 201 separate–merge direction 13 sequential skills acquisition 77 Shanahan, T. 311, 313 shifts in meaning 300, 307; word choice 304–5 signal extraction 390; statistical 386 Silverstein, M. and Urban, G. 408 simulations 362; in language 115 singular 156–7; condition 154; expression 151 social 119, 182, 329, 334, 335; activities 46, 102; affordances 247; agency theory 310; alternative orders 348; communication 310; context 323, 369, 415; criticism 114; cues 328; cues reduction 334; dimension 117, 329; discursive actions 46; environment 310; equality 120; exchanges 329; factors 370; information 328; institutions 341; interaction 45, 51, 356, 374, 413; involvement 312, 322; practices 48, 148, 184, 414, 417; presence 313, 319, 322, 335; processes 413–14, 416; psychology 19, 114, 353; reality 119; resources 246; sciences 285, 288 sociocultural 415; environment 59; factors 414; inputs 374; perspective 199, 224; practice 58; theories of literacy 353 socioeconomic factors 366; poverty 98 sociological 17, 353, 362; perspective 25; research 18; studies 23 software code 342–3 software program 343; specialized 230 Sommers, N. 371 sound waves 231 Soviet Union 97–9; epideictic oral culture 102; former 86; psychologists 415; theories 4, 6–7; value system 91 Spack, R. 368 Spain 71 Spanish 50, 122, 138, 230; language textbook 188; students 71; translation 256
444 Index Spanish education 71; curricula 72; law 72–3, 123; primary 75, 79n1; universities 185 speech 51, 111, 117, 312; changes 154; indirect act 425; latched 413; production 399; recognition systems 225 spelling 169, 177, 208; fluency 202, 204; mistakes 191, 196; problems 166 Spinuzzi, C. 358 spoken and written information 87, 154, 163; modifications 161–2 stability 341, 358; institutional 344 Stalinist 87–8; grammatical exercises 90 Standard Modern Arabic 153 Steinke, Erin 256 Step up to Writing 212–15, 217–18, 220–2 Sternglass, M. S. 370–1, 375 strategy 73, 77–8, 371; ambiguous 360; cognitive 75; communal 330; deviant 159, 163 structure 77–8; antagonistic 22; changes 32; institutional 258; logical 307; problem 130 students 35, 64, 104, 113, 128, 207, 213, 267, 286–7, 290, 292, 299, 308, 330, 359; activities 39; autonomy 289, 292; Batchelors 32, 361; capacity 201; Chinese-Malaysian 298; classroom experiences 370, 374; compositions 313; conceptual understanding 220; conference posts 330; construct 359; dissatisfaction 39; explanations 298; French 284; high-achievement 278; Indigenous 206; interaction 76; interaction with teachers 304, 371; language skills 260; learning 38, 213, 215, 280, 362, 371; learning opportunities 217, 221, 307; literacy 287, 289, 291; lowachievement 278; MA 11; mental schemata 188; Mexican immigrant 226; middle school 227; motivation 71; needs 206; non writers 127, 206; North American 284; novice reporters 230; online discussions 325; oral proficiency 233; outcome assessment 99; questionnaires 183; questions 300, 302–3; reaction 315; relationship with school 68n1; responses 65; status differentials 206, 213, 334; success 40; surveys 35; urban 366
student difficulties 22–4, 27, 37, 126, 213, 220, 225, 287, 308; comprehension 127; confusion 297–300, 304, 307; in explaining 303; financial pressures 375; in reading and writing 65, 129 student writers 19, 86, 200, 202, 266; development 371; discussion of word choice 207; support 209, 214; uses of textual structures 187; working alone 77 student writing 31, 33–4, 37–9, 61, 64–5, 102, 256, 260, 262, 265, 268, 280, 288, 336; abilities 367; analysis 204, 207; areas of weakness 199; assignments 31; compositions 316, 322–3; development 369–70; drafts 38; essays 215; experiences 73; improvement 62, 194, 260, 359; notes 123, 219; performance 126, 371; pre-existing abilities 369; problem 359; quality 309; relationship 19, 26, 61, 64–7; samples 100, 104, 201; skills model 336; structuring 40 suffix 153, 163 summarization 218, 410 supervision 35–6 Swales, J. 268, 422 Swanson, H. L. 174 Sweller, J. 310 Switzerland 313; Swiss Public Radio and Television 314 syntax 73, 184, 409, 428; single structure 140 targeted approach 199 target-like utterances 156, 158–9 tasks 60; complex written 209; environment 178 teachers 44, 79, 195, 206, 215, 231, 233, 237–8, 240, 246, 286, 289, 291–2, 311, 335, 359, 369, 419; active role 311; Associations 253, 256; behaviors 374; capacity 198; challenged 245; choices 62, 64–5, 79, 221; comments 371; conflict 37; construct 359; describing writing 187; descriptions 186; didactic options 64–5; discourse 60; flexibility 214, 218; identity 242; influence 212–13; inservice work 238; interventions 149; interviews 183, 201; knowledge building 200, 209; leadership 247; modeling 77;
Index 445 networks 237; opportunities 239; oral communication 290; orientation workshops 124; participants 200; practices 183; presentation 220; publication 237; reflection 204; research 237, 239; responsive 370; student 27, 185, 188; student relations 217, 367; substitute 312; support 291; training institute 97; of writing 238 teachers-teaching-teachers model 238 teacher-writer 239, 244; barriers to publication 237, 239; difficulties 242 teaching 32, 36, 53, 136, 217, 314; active methods 32; assistants 284, 330, 334, 360; change of system 39; contexts 60; description 90; difficulties 22–3; effective 199; human level 360; initiatives 257; language 19, 113; materials 60–1; methods 72, 262; planning strategies 76; practices 58–9, 62, 186, 199; priorities 222, 292; reading and writing 45, 122, 123; requirements 149; sequences 135, 183; Support Computing Service 185 teaching writing 46, 59–60, 68, 73–4, 89, 198–200, 202, 209, 238, 252, 255–6, 261, 283; academic practices 61; in higher education 41; in schools 365; transversally 24; in the United States 18 technical writing 219–20, 222; reports 260; scientific papers 267 technology 225, 348; artefacts 230; critical theories 342, 348; electronic 251, 254; high technology 9; new 127 tension 19, 22–3, 27, 344–5, 354–5, 359–60; building 208; community 112; internal 346; interpersonal 334; personal 26; productive 342; social 334; structural 22 termination 403 tertiary education 299; Argentina 283 testing 97, 99, 103–5; tests 123 text 184, 312, 353, 379, 394, 407, 415–16; analysis 45, 48, 75, 78, 147, 185, 212, 228, 368, 406, 408, 410, 416; classification 147; conceptualized 73; in context 47; descriptive 73; diversifying 59; elaborate 140; emerging 8, 394;
explanatory organizers 187; expository 182, 187–8, 191, 194; generation 174; genres 45–6, 53, 60, 361; inferences 286; initial frame 149; leading line 93; level 380; non-linear 9, 225, 374; opinion 62–3; organization 202–3, 272; oscillations 139–40; planning 378; pre-existing conceptions 367; professional 47; quality 74; representation 411; scientific analysis 7; span 403; structure 73, 76; studies 44, 46, 48–9; superstructure 184; target 420; and writing 93; written 230, 308, 378, 407 textbooks 6–7, 50, 89, 92–3, 122–3, 212, 312; history 422–3; on speech and writing 8, 91 text production 65, 73, 182, 184, 379, 381–2, 391n2, 396; creative 60; human agency 367; phases 380; process 378, 380, 404 textual analysis 11–12, 24, 52, 366, 368–71, 373, 419; linguistics 60; macrostructure 189; modalities 196; nonlinear 46; organizers 188; patterns 7–8; statistical 195; structural 12 textual content 46, 63; discursive aspects 184; features 409; hierarchy 410; pathways 360 textualization 182, 184 textual product 408; diversity 63; production 59–60; representation 415; structure 48; subtype 191 textual silence 419–20, 421, 425, 429; genre-based 422–3; lexical 424 The Problem of Speech Genres 114 The Rhetoric of Reason 98 The WAC Journal 252 THEODILE 19, 28n2 theoretical 277, 304; epistemology 52; general framework 10; knowledge 6–7; methodological constructs 53; orientation 279; perspectives 51; terms 272, 275, 278, 307; writing studies 5 theory 10, 13; of content knowledge 17; of genre systems 47; issues 8; of writing 4, 7, 380 theory-laden 406–7, 414 think-aloud 77–8, 366–7, 408, 412–13, 416 Thiollent, M. 112
446 Index third person writing 138–9, 144, 146, 314 Third World 117; Brazilian 120 time organization 77–8, 381 time series 378; short-term dependency structure 390; structure 388 time to write 240–1, 246; lack of time 38 topic 176–7, 184, 394, 420; ideologized 89; immersion 368; knowledge 202, 203; physical representation 174; planning 378; recurrent 146; research 228; sentence 217; subject matter knowledge 429; writing 89, 93 topical silence 420–2 transcriptions 167, 174, 176–8, 227, 414; conventions 413; interview 408; mode 168–9; recordings 300; skills 166; video 319 transformation 11, 112, 358, 409–10; ideological 87; intrapersonal 104; identity 231; potential 234 transitional appropriation of discourse 306, 307 transitional shifts 307; in learning 307; in understanding 305 transparency 420; lacking 422 trends 388–9; unobserved 387 triangular model 415 trust-building processes 328–9 Tuchman, G. 406 Turner, S. 415 typification 356–9, 362 typing 171, 397; concurrent fixations 400; error 398–9; practice 167 Ulichny, P. and Watson-Gegeo, K. A. 413 undergraduate 289, 290–1; courses 125, 260; working-class 284; writing 287, 372 unintended effects 37, 39 United Kingdom 252, 261–2, 292 United States 97–8, 257; contexts 253; education policies 252; government survey 113; liberal pedagogies 93; literacy education 85, 87; literate culture 117; research 74; trends 91; US/Canada survey 254, 257; universities 44, 97, 99, 124, 129, 251, 256, 359–60, 266, 371–2, 369–71, 373; WAC activity 251; writing studies 85 university 34, 45, 117, 122, 124, 127, 129, 254; accreditation 125;
Brazilian 111–13; Canadian 254; Chinese 11; classroom 307; culture 127; departments 359; elitist 33; French 18, 23, 254; genres 25; graduate programs 49; literacies 18, 23; macro-level policies 37; Malaysian college 298; norms 27; Norwegian 31–3, 38, 41, 42n4; pedagogical practices 23; postgraduate programs 125; private and public 125, 283; reading and writing practices 27, 129; researchers 226; teachers 290; texts 22; traditional disciplines 37; unrestricted entry policy 284; writing 17–18, 24, 26–7 video 315, 362, 412; interviews 314 visible authors 312, 322 vocabulary 329; development 202, 204, 208; specialized 410; teaching 218 voice 204, 225–7, 246; classroom teachers 237; editing 231–2; hearing self speak 234; and identity 234; primary artifact 230; realization 232–3; tenor 48; visualization 228, 231 Voloshinov, V. I. 111, 114, 117 Vygotsky, L. S. 46–7, 104, 298–9, 307–8, 353–7 Wake, B. 302, 304 Wegerif, R. 329 Wenger, E. 415 Western style 93, 98, 107; education 105; methods 99, 106–7; writers 103 Wikipedia 339, 343, 345 wikis 339; academic 348; active academic spaces 340; active research 345; codes 342, 343, 344, 345, 347, 348 word 208; accuracy 395; choice 202, 204, 424, 208; internal elements 152; locations 395 working memory 166–7, 177; theory 310 World Wide Web 361 writers 9, 246; academic 406; adult 395; autonomy 9, 19, 148; beginner 135, 178; developing 174, 177; expert 60, 202; eye movements 394, 403; famous 4; identity 105; intelligence 12; learning 72; novice
Index 447 60, 202; professional 378; proficient 218, 221; reluctant 206; selfdefinition 241; writerly habits 24 writing 4, 6, 12, 41, 64, 74, 116, 123, 207, 226, 228, 251, 256, 289, 368, 383, 390, 417; in academic contexts 182; activities 58, 66, 89, 183, 403; analysis 58; approach 367; assessment 202; assignments 37, 39, 59, 89, 218; attitudes 75; for an audience 246; basic elements of 11; changes reflected 161; cognitive theory 394; computer-based 409; concepts 222; contexts 25, 71–2, 409; conventions 7, 107; crisis 67; culture 102; curriculum 221; descriptive orientation 27; dimensions of 19; drafts 39; experience 369, 371; extracurricular 65, 373; gains 371; guided project 194; habits acquired 244; impact on learning 366; literacy 198; literary 3, 11; log 76, 78–9; maturity 409; models continuum 19–20; notebooks 62–5, 66–7; for others 137; pedagogy 65, 86, 90, 238, 419; performance 73, 372, 391; phases 379, 381, 384–6; practical 9; practices 25, 135; programs 213, 254; progressions 385; protocol segments 170; for publication 243–4; purpose 368; quality 79, 90, 198–9, 316; quantity 65; for self-reflection 369; sequenced instruction 97; skills and abilities 12; social nature 9, 413; space 344; in statistics 27; sub-genres 8; successful 76; summaries 22; support 285; systems 151–2, 159; talking about 417; texts 3, 170; tool for learning 41, 255, 285, 361 writing across the curriculum 251, 265, 283, 359; Clearinghouse website 251–2, 254; directors 258; International Network of WAC Programs 251, 254; obstacles 360; programs 252, 255, 257–9; WAC/ WID initiatives 251–4, 263 writing and reading 60, 380; instruction 291 writing development 45, 177, 204, 366, 368, 373–5; non-classroom factors 374–5; parallel 105 writing difficulties 10, 21–2, 102, 201, 284 writing education 12, 59, 123; classes
6; discipline 6; during training 25; in high school 285; in higher education 366; instruction 71, 73, 77–8, 86, 108n4, 215, 289, 366, 370; instructors 72, 429; practices 31; in primary and secondary schools 27; strategy-focused 79; tasks 75, 177, 218, 360, 371, 411; teachers 33; working in pairs 64, 77; writing-tolearn 410 writing forms 20; appropriate 220; fiction 11; non-interchangeable 220 writing genres 62, 219; in higher education 24 writing intensive 38; courses 37, 255; methods 98; workload spread 35 writing in the disciplines 251–2, 260, 283, 373 writing practices 13, 19–20, 23, 37, 65; changes 40; development 373; functional 21; guided 217 writing processes 11, 59, 62, 66, 71, 74, 77, 79, 135, 198, 200, 383, 389, 411, 417, 391n1, 399; developmental dimensions 375; interaction 167; linear 9, 383; oscillation phase 388; parallel 400; regularities 390; regulation 75; substantive 73 writing research 7–8, 11–12, 58, 85, 93, 198, 362, 378, 390, 408, 417; cognitive 41; community 407; and education 12–13; samples 99, 101, 201; scientific laws 6; studies 6, 9–10, 373 writing researchers 408, 411, 419 writing skills and abilities 7, 71, 78–9, 208, 372; growth 374; improvement 261; social 182; strategies 381; styles 11, 105, 218–19 writing theories 8; interdisciplinary 10; singular model 20, 23; Western 4 writing workshops 59, 62, 65–7, 186; on journal article 246; limitations 68n1 writingology 3–4; basic theories 11; computer 9; modern 5, 10, 12 written language 113, 153, 200; in school settings 148; university practices 18 written production 79, 137, 154, 181, 186; improvement 174; stages of 183 written representation 154, 156, 161, 164 written-spoken link 164; interface 111
448 Index Yerevan State Linguistic University 97, 104; classes 101; classrooms 98–9; teachers 102 younger children 136–43, 144–9, 171–2; Brazilian 113; Chinese 157; Moroccan 159
youth radio 224, 230; activity cycle 226, 228; practices 233–4; program 225, 232 Zawacki, Terry Myers 254–5, 259 zooming-in 201, 208–9