Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Translating and writing processes of adult second language learners Uzawa, Kozue 1994

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1995-983582.pdf [ 3.46MB ]
JSON: 831-1.0054812.json
JSON-LD: 831-1.0054812-ld.json
RDF/XML (Pretty): 831-1.0054812-rdf.xml
RDF/JSON: 831-1.0054812-rdf.json
Turtle: 831-1.0054812-turtle.txt
N-Triples: 831-1.0054812-rdf-ntriples.txt
Original Record: 831-1.0054812-source.json
Full Text

Full Text

Translating and Writing Processes of Adult SecondLanguage LearnersbyKozue UzawaB. A., Sophia University, TokyoDip. App!. Ling., The University of British ColumbiaM. A., The University of British ColumbiaA THESIS SUBMI1TED IN PARTIAL FULFILMENT OFTHE REQUIREMENTS FOR THE DEGREE OFDoctor of PhilosophyinTHE FACULTY OF GRADUATE STUDIESCurriculum and Instructional StudiesWe accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIAAugust, 1994(c) Kozue Uzawa, 1994In presenting this thesis in partial fulfillment of the requirements for an advanced degree atThe University of British Columbia, I agree that the Library shall make it freely available forreference and study. I further agree that permission for extensive copying of this thesis forscholarly purposes may be granted by the Head of my Department or by his or herrepresentatives. It is understood that copying or publication of this thesis for financial gainshall not be allowed without my written permission.Centre for Curriculum and Instructional StudiesThe University of British Columbia2075 Wesbrook PlaceVancouver, CanadaV6T 1WSAugust, 199411AbstractWhile translation in L2 learning/teaching has been viewed negativelysince the 1950s in North America, in the late 1980s a re-evaluation oftranslation has begun (Duff, 1989). The purpose of this research is toexplore text-level translation from the learner’s perspective, as this kind ofresearch, at present, remains quite scarce (Krings, 1987). This study focuseson text-level translation as a useful component of second language (L2)learning/teaching. Adult L2 learners’ translation processes and performanceare examined and contrasted with the same group’s Li and L2 writingperformance.Twenty-two Japanese ESL students studying at a Canadian collegeperformed three tasks individually (translation from Li into L2, Li writing,L2 writing), thinking aloud. Their writing samples were evaluated, andthink-aloud protocols were analyzed, supplemented by interviews and textanalyses.The data were analyzed with attention given to four recent cognitivetheories of language learning: Cummins’ theories (1986) of cross-linguisticinterdependence of cognitive academic skills; Schmidt’s “consciousattention” (1990); Swain’s “i+i output” hypothesis (1985); andMcLaughlin’s “restructuring” (1 990b).Findings: 1) The correlations of the quality of translation, Li writing,and L2 writing of L2 learners (whose Li writing skills are still developing)were not significant. 2) The learners’ conscious attention to language use1111was high in the translation task, but unexpectedly low in the L2 writing.Their language use was more sophisticated in the translation than in the L2writing. 3) Some students preferred translation tasks to L2 writing tasks,expressing their views which were consistent with the 9+1 output”hypothesis. 4) Contrary to general expectation about student translations, thestudents did not translate word for word; they often restructured Li/L2correspondences, and examples of “restructuring” were not limited to theword level.General conclusions: Cross-linguistic interdependence amongtranslation, Li writing, and L2 writing was not confirmed clearly. However,there was evidence that translation processes prompted conscious attention,“i-i-i output”, and restructuring, which some consider to be necessary forsecond language learning. Thus translation in L2 learning deserves a closerlook as it provides potential opportunities for learners to learn a secondlanguage.ivTable of ContentsAbstract.iiList of Tables viiiList of Figures ixAcknowledgements xChapter 1 INTRODUCTION 11.1. Overview 11.2. Purpose of the Study 313. Definition of Terms 4Chapter 2 REVIEW OF LITERATURE 52.1. Text-Level Translation and Second Language Writing as LanguageProficiency 52.1.1. Negative and Positive Views toward Translation 52.1.2. Text-Level Approach to Translation 92.13. Translating and Writing Processes 112.1.4. Translation Models 152.2. Translation and Second Language Writing as Means of Learning a SecondLanguage 192.2.1. Translation in Language Teaching/Learning 192.2.2. Conscious Learning and the “i+1 Output” Hypothesis 232.2.3. Restructuring 2523. Research Questions 282.4. Methodological Issues 302.4.1. Holistic Scoring 302.4.2. Think-Aloud Protocols 32Chapter 3 METHODS 343.1. Preliminary Studies 343.1.1. Case Studies 343.1.2. Pilot Study 363.2. Data Collection 403.2.1. Recruiting Participants 403.2.2. Pre-Writing Sessions 433.2.3. Interview Questions 443.2.4. Writing Sessions 45V3.2.5. Writing Topics .4633. Characteristics of Participants 493.3.1. Backgrounds 493.3.2. Classification of Participants 513.4. Transcribing, Segmenting, and Coding of Protocols 543.5. Evaluation of Writing Samples 563.6. Procedures for Analyses 583.7. Meeting with Instructors 60Chapter 4 RESULTS AND DISCUSSION 624.1. Results for Research Question 1How does the quality ofsecond language learners’ translation correlatewith the quality oftheir Li and L2 writing? 624.1.1. Summary 624.1.2. Analyses 634.1.3. Discussion 674.2. Results for Research Question 2What do second language learners think about translation and secondlanguage writing tasks? 694.2.1. Summary 694.2.2. Analyses 704.2.3. Discussion 734.3. Results for Research Question 3What do second language learners pay conscious attention to whiletranslating and writing? 764.3.1. Summary 764.3.2. Analyses 764.3.3. Discussion 814.4. Results of Research Question 4How often do second language learners pay metacognitive andmetalinguistic attention while translating and writing? 834.4.1. Summary 834.4.2. Analyses 844.4.3. Discussion 894.5. Results of Research Question 5What characteristics are observed in the translating and writingprocesses ofsecond language learners? 94vi4.5.1. Summary .944.5.2. Analyses 944.5.3. Discussion 964.6. Results of Research Question 6How do second language learners restructure L1/L2 semantic andsyntactic correspondences while translating? 984.6.1. Summary 984.6.2. Analyses 994.6.3. Discussion 103Chapter 5 CONCLUSION 1055.1. Summary of Findings 1055.2. Generalizations 10753. Significance for Theories of Second Language Acquisition 1085.3.1.Skill Transfer 10853.2. Conscious Learning 1095.3.3. The “i+1 Output” Hypothesis 1105.3.4. Restructuring 1115.4. Implications for Second Language Education 1125.4.1. College Instructors’ Views 1125.4.2. Text-Level Approach 1155.4.3. Peer Interaction 1175.4.4. Separate Skills 1175.5. Suggestions for Future Research in Translation and L2 Writing 118REFERENCES 120APPENDIX AMemo to the Instructors 13 1APPENDIX B“Volunteers Needed for Educational Project” 132APPENDIX CStatement of Consent 134APPENDIX DQuestionnaire on Background 136APPENDIX ETranslation Task 137APPENDIX FModel Translation 138vi’APPENDIX GTranscribed and Coded Think-Aloud Protocols 139APPENDIX HESL Composition Profile 154APPENDIX IRatings of Writing Samples 155APPENDIX JParticipants’ Writing Samples 158viiiList of TablesTable 1 Coding Scheme Used in the Pilot Study 39Table 2 Proposed Research Design 41Table 3 Preliminary Classification of Participants 42Table 4 Interview questions 44TableS Questions on Tasks 46Table 6 Backgrounds of Participants 50Table 7 Modified Classification of Participants 53Table 8 Coding Scheme 55Table 9 Inter-Rater Reliability on Evaluation 58Table 10 Results of Correlational Analyses 64Table 11 Group Comparisons of Quality of Writing Samples 66Table 12 Raw Scores of Participants’ Answers to Questions on Translation and L2Writing as a Means for L2 Learning and Improvement 71Table 13 Participants’ Reported Conscious Attention in Three Tasks 80Table 14 Metacognitive and Metalinguistic Attention As Measured by Frequency(%) of Utterance 87Table 15 Comparison of Scores on Language Use in Translation and L2 Writing . . .93ixList of FiguresFigure 1 Translation as Third Code .15Figure 2 Translation Model of the Skilled Translator 16Figure 3 Translation as One-Way Act 18Figure 4 Profile of Participants’ Thinking on Translation and L2 Writing 73Figure 5 Participants’ Reported Conscious Attention in Three Tasks 81Figure 6 Participants’ Mean Attention Levels in Three Tasks Measured byPercentage of Utterance 88xAcknowledgementsMany people helped me to go through the long process of completingthis dissertation. Especially, I thank Bernie Mohan, who helped me to getpast an initial psychological obstacle that I had been experiencing. He guidedme in the right direction and, without his insightful advice, this dissertationmay have possibly run aground. I also thank Stephen Carey, who generouslyprovided advice on designing the study and consultation needed to completethis paper. His comments and advice were inspiring. Alister Cumming, nowat OISE, helped me during the preliminary studies.Barbara Siennicki, Sandy Filippelli, Man Miyasaka, and MayumiHasegawa rated participants’ writing samples with their expertise as languageteachers and editors. Jonathan Berkowitz assisted with his professional adviceon statistics. Michael Speier read many drafts tirelessly, making commentsand editing my English.Faculty members at Canadian International College helped me torecruit participants from their school. The faculty’s comments, criticisms,and questions at the discussion session, after completion of the first draft,were very informative and encouraging.Finally, and perhaps most importantly, I thank the 22 Japanesestudents at Canadian International College, who enthusiastically volunteeredto participate in this study. Their involvement made this research possible.1Chapter 1INTRODUCTION1.1. OverviewFrom the late 1950s to the early 1980s, use of the mother tongue insecond language learning was considered ineffective by many researchersand educators in North America (Asher, 1981; Krashen, 1981, forexample). Many Canadian and American language classes prohibited studentsfrom using their mother tongue or translating from the first to secondlanguage (Auerbach, 1993). Sorhus (1975, pp. 2-3) reviews this trend andcomments:After the Second World War, when the Grammar-Translation Methodfell into disrepute, translation lost favour, too. It was believed thatstudents spent too much time learning about the second languagerather than practising it. The application of Behaviourism settled thematter by banishing translation and by focusing on making languagehabits automatic through many drills. Contrastive Analysis put a sealon the banishment of translation by introducing the concept“interference” to explain the effect of the first language upon thesecond. Yet students could still be seen translating, especially at thebeginning of their second-language courses, and even later, wheneverthey had difficulty finding equivalent MEANING in their newlanguages.During this period (late 1950s - early 1980s), research on spokencommunicative competence in second languages was very popular in Canada2and the United States (Chaudron, 1988; Faerch & Kasper, 1983; Krashen,1981; among others), while research in translation and second languagewriting was neglected. Although research in second language writing hasbecome very active in the last decade in North America (e.g., Cumming,1989; Edeisky, 1986; Jones & Tetroe, 1987; Zamel, 1976, 1983), researchon translation has lagged behind (see Krings, 1987).Translation might merit potential as a valuable activity for secondlanguage learners; also, it should not be overlooked as an academic subjectby itself. In light of the recent cognitive approach to second languagelearning (Cummins, 1986; McLaughlin, 1990b; Schmidt, 1990), and thetext-level approach to second language teaching (Brown & Yule, 1983;Carrel!, 1985; Celce-Murcia, 1990; Mohan, 1986; Meyer, 1985), thisdissertation researched the possible value of text-level translation in languagelearning, comparing it with first and second language writing.The dissertation’s five chapters are outlined as follows: theintroductory chapter states the purpose of this study and provides adefinition of terms; the second chapter presents a literature review, someresearch questions and methodological issues; the third chapter describes theresearch methods used in this study; the fourth chapter reports the results ofthe study with detailed analyses and discussion; and the final chaptersummarizes relevant findings and discusses their significance for theories ofsecond language acquisition. Also addressed are implications for secondlanguage education, and suggestions for future research in translation andsecond language writing.31.2. Purpose of the StudySince empirical data on text-level translation in the framework oflanguage learning are still scarce (Krings, 1987), the purpose of this study isto explore text-level translation performed by adult second language learnersand to seek the value of translation for second language learning andteaching. This study researches text-level translation in two ways: i) as acomponent of language proficiency, and ii) as a means to learn a secondlanguage.Four recent cognitive theories in second language acquisition guidethe qualitative analysis of the data. They are: Cummins’ (1986) “linguisticinterdependence of cognitive academic skills” (cognitive skills such asreading and writing developed in the learner’s mother tongue aretransferrable to a second language), Schmidt’s (1990) “conscious learning”(the learner learns a second language by paying conscious attention togrammar, vocabulary, syntax, pronunciation, and pragmatics), Swain’s(1985) “i+i output” hypothesis (the learner has to produce an output whichis a little above the learner’s present level in order to learn a secondlanguage), and McLaughlin’s (1990) “restructuring” (the learner learns asecond language by restructuring Li and L2 correspondences, not by merelyadding new items to the existing L2 knowledge).Learners’ translating and writing processes, performances, skills,and perceptions are the primary focus of this thesis. Analyses include theirconscious attention to language use, their restructuring Li and L2correspondences while translating from their mother tongue into a second4language, and their writing samples in translation, in Li and in L2 writingtasks. Learner& perceptions of translation and second language writing arealso examined through interviews in order to explore language learning indepth.1.3. Definition of TermsBefore proceeding further, the definitions of some terms used inthis study are presented here.Translation: In this research, only written translation of text ina learner’s mother tongue (Li) into second/foreign language (L2) isconsidered for the purpose of comparison with second language writing.Translation is seen in the framework of second language learning andteaching, not as a science for professional translators.Writing: Writing is meant in this research as composing (apassage or an essay). This term does not include calligraphic writing or atype of writing which involves just jotting down a list of words or phrases.Copying text or dictation is not included, either.5Chapter 2REVIEW OF LITERATURE2.1. Text-Level Translation and Second Language Writing asLanguage Proficiency2.1.1. Negative and Positive Views toward TranslationIn university-level foreign language learning, acquiring translationand second language writing skills may be important, especially for peoplewho seek careers in journalism, commerce, international relations, and soforth. Translation itself has a long history. Newmark (1981) comments thatour “intellectual and artistic cultures” are “heavily indebted to translation”.Good translators are valuable in our society. Without good translators,“there would be no summit talks, no glasnost or perestroika, no Cannes FilmFestival, no Nobel prizes, no advances in medicine, science, or engineering,no international law, no Olympic Games, no Hamlet, no War and Peace...”(Duff, 1989, p. 7). Translation is seemingly a necessity in a world of sharedinformation and global relationships. However, although translation itself hasa long history, research concerning translation in a language learningcontext, compared to research concerning second language writing, has beenrelatively scarce (see Krings, 1987).Research of second language writing became active following, thestudent-centred process approach to first language writing in English. Fromthe early 1970s to the 1980s, English educators and researchers shifted their6attention from analyses of written texts to studies which also addressedlearners’ writing processes (Bereiter & Scardamalia, 1987; Britton, et a!.,1975; Emig, 1971; Hillocks, 1986; Langer & Applebee, 1987), andcommented that writing essays and compositions (rather than completingword-level or sentence-level exercises) is necessary for students to think andlearn (Applebee, 1982, 1984a; Emig, 1977; Goodlad, 1983; Langer &Applebee, 1984, 1987; Moffett, 1968). Writing is a personal, self-pacedproblem solving activity. A role for “writing as heuristic” (Emig, 1977) hasbeen alluded to by many thinkers and educators (Bruner, 1971; Moffett,1968; Olson, 1977; Polanyi, 1958; Vygotsky, 1986; among others). Secondlanguage writing researchers found many parallels between first and secondlanguage writing, commenting that second language writing is also apersonal, self-paced problem solving activity (Cumming, 1988; Raimes,1985, 1987; Zamel, 1976, 1982, 1983).Translation in second language learning, however, has beennegatively viewed by many researchers and educators. It has often beendeemed a sentence-level activity, one which may not require deep thinking.Applebee (1982) unfavorably asserts that direct translation from onelanguage to another is “writing without composing” because “the text to betranslated provides the global structure, within which the translator canoperate at a sentence-by-sentence level” (p.371).Other researchers note the negative influence of translation onwriting. Lay (1975) and Sa’Adeddin (1989), for example, report thenegative influence of the learner’s mother tongue on writing English as asecond language (ESL). Similarly, Kaplan’s “contrastive rhetoric”, which hasbeen very influential in the English as a second language research circle(1966, 1967, 1978, 1979, 1987; Andrade, 1990; Connor & Kaplan, 1987;7Hinds, 1980, 1983; Jenkins & Hinds, 1987; Leki, 1991), regards translationas an interference in ESL writing.Translation has also been seen as a simple code-switching used bylanguage learners, a communication strategy that compensates forinsufficient linguistic knowledge in the learner’s second/foreign language.Studies on learners’ strategies for communication in the second languagehave not viewed translation as undesirable; but then they have not viewed itas an academic subject worthy of learning, either (Faerch & Kasper, 1984;Haastrup & Phillipson, 1983; Tarone, Cohen, & Dumas, 1983; Váladi, 1983;among others).In the late 1950s and up until the early 1980s, translation wasparticularly rejected in second language pedagogy as the Grammar-Translation Method in North America. Translation was seen as a definitecause of “interference” in second language learning (see Sorhus, 1975).Even today, translation is ignored as a valid activity for language practiceand improvement largely due to the communicative movement in the 1980s,which put a strong emphasis on spoken communicative competence (seeKrashen, 1981; Chaudron, 1988; Ellis, 1988). Dulay, Burt, and Krashen(1982) comment that the second language is a new and independent languagesystem and that contrasting the Li and the L2, or translation, does not helpthe learner to “acquire” the second language. Krashen. (1981, p. 66) says:• . . studies that report a high amount of first language influence,are mostly foreign and not second language studies, situations in whichnatural appropriate intake is scarce and where translation exercises arefrequent [emphasis addedi.8In spite of the banishment of translation in second language pedagogy,some researchers see value in translation within language learning. Duff(1989), a language educator, comments that translation activities in theclassroom help students to better understand the influence of one language onthe other because translation involves contrast, which enables them “toexplore the potential of both languages - their strengths and weaknesses”. Hecomments that through translation, the learner develops three qualitiesessential to all language learning: accuracy, clarity, and flexibility.Translation, he says, “trains the learner to search (flexibility) for the mostappropriate words (accuracy) to convey what is meant (clarity)” (p.7).SneII-Hornby (1985) also says that translation allows the student to learn tojudge what is equivalent and what is not between Li and L2; hence it is aworthwhile exercise in gaining mastery of, and knowledge about, a foreignlanguage.B. Harris (1977, 1978) views translation as a natural skill, saying thatall bilinguals and students who learn a second language do translate even ifthe method used ignores or forbids it. Similarly, Carey (1991) says thatknowledge acquired in L2 is often translated into Li and stored in one’smemory in Li. Thus, Carey says, French immersion students in Canada,after several years of studying school subjects in French, still perform betterin subject exams written in Li (English) than in L2 (French) (see alsoSamuel, 1990). Cumming (1990) and Uzawa and Cumming (1989) reportthat adult language learners, when they write in .a second language, oftencompare and contrast Li and L2, and translate/back-translate, consciously, inorder to express what they really want to say. Most second languagelearners, thus, have a natural skill for translation. Positive views towardtranslation suggest that learners’ natural skill for translation should be9guided and honed to produce better products of translation (B. Harris, 1977,1978).In summary, translation, as the Grammar-Translation Method, wasdevalued or totally ignored in language teaching from the 1950s up until theearly 1980s in North America. However, researchers in the late 1980s beganseeing the potential value in translation in second language learning, andsome are now treating translation as a worthwhile academic subject. Viewingtranslation from the learner’s perspective, these researchers affirm thelearner as having natural skills of translation and emphasize that these skillsshould be nurtured.2.1.2. Text-Level Approach to TranslationStudents may be asked to translate single sentences (“sentence-leveltranslation”) or they may be asked to translate a whole discourse or text(“text-level translation”). While a text-level approach has been suggested bymany researchers in teaching reading and writing in a second language(Carrell, 1985; Meyer, 1977, 1985; Mohan, 1986), the approach togrammar and translation may be still dominantly sentence-level in manysecond language classrooms in most countries (Applebee, 1982; CelceMurcia, 1990).Some researchers in the late 1980s, however, advocate the text-levelapproach to translation, and take into account the learner’s perspective aswell (Bell, 1991; Butzkamm, 1985; Duff, 1989; Enns-Connolly, 1985;Gerloff, 1987; B. Harris, 1977, 1978; Heike, 1985; Krings, 1987; KupschLosereit, 1985; Snell-Hornby, 1985; Titford, 1985; Tudor, 1987b;10Weymouth 1984). In their approach, translation is positioned as a validsubject in the students’ language curriculum.For instance, Snell-Hornby (1985) emphasizes that translation shouldbe approached as a complex, structured whole, not as a chain of separatesentences, and that in translation, as in writing, coherence, cohesion, focus,and progression are of primary importance. Snell-Hornby further says that“the detailed analysis of a well-written text, in particular one for translationfrom the foreign language, provides a basis for creative writing in thatlanguage, as in composition and essay-writing” (p.24). Similarly, Tudor(1987b) suggests that analyzing the genre, purpose, style, socio-professionalprofile, cultural allusions, and verbal associations of the source text involvedin translation is useful for language learning.In second language pedagogy in the 1980s, the emergence oftranslation as a worthy academic subject was influenced by the rise of text-linguistics (van Dijk, 1972; Kintsch & van Dijk, 1978), and discourseanalysis (Brown & Yule, 1983; Kuno, 1978; Ono, 1990). Brown and Yule(1983) argue that a collection of single sentences does not form a text. Kuno(1978) shows how a single sentence changes its meaning depending on thecontext. Similarly, Celce-Murcia (1990) comments that the sentence-levelapproach to grammar has problems in semantic and/or pragmatic areas. Forinstance, Celce-Murcia (1990) argues that tense-aspect-modality, wordorder, subordination, and complementation in English are often determinedat the discourse level, not at the sentence level (e.g., Indirect objectalternation as in “Peter gave the book to Alice” and “Peter gave Alice thebook” is mostly determined at the discourse level by the relative degree of‘given’ or ‘new’ information in the two objects).11In summary, many researchers in the late 1980s say that grammarand translation exercises in the second-language classroom shouldincorporate a text/discourse level approach. They emphasize that translationor grammar should be approached as a structured whole, not as a collectionof separate sentences, because word choice, word order, meaning, and soforth, depend on the context.2.1.3. Translating and Writing ProcessesIn first language writing research, the writing processes of skilled andunskilled writers have been studied in hopes of improving students’ writing(Bereiter & Scardamalia, 1987; Emig, 1971; Faigley & Witte, 1981; Rower& Hayes, 1980, 1981a, 1981b; Matsuhashi, 1981; Matsuhashi & Gordon,1985; Perl, 1979; Pianko, 1979; Rose, 1980; SeIzer, 1983; Shaughnessy,1977; Sommers, 1980; Witte, 1985; among others). Stimulated by thisresearch, second language writing researchers also studied second languagewriting processes of language learners in order to improve the writing ofESL students (Butler-Nalin, 1984; Cumming, 1988, 1989; Friedlander,1990; Friedlander & Hucking, 1987; Gaskill, 1986; Jones & Tetroe, 1987;Raimes, 1985, 1987; Taylor, 1981; Zamel, 1983).Research on the translating processes of skilled and unskilledtranslators may also prove useful for second language classrooms (Gerloff,1987). Studying how skilled and unskilled translators approach translationmay be necessary in order to evaluate the place of translation in foreignlanguage education properly.Researchers who studied the writing processes of skilled(experienced) and unskilled (inexperienced) writers of English as a mother12tongue (Bereiter & Scardamalia, 1987; Burtis, Bereiter, Scardamalia, &Tetroe, 1983; Emig, 1971; Flower & Hayes, 1980, 1981a; Flower, et a!.1986; Per!, 1979; Pianko, 1979; Rose, 1980; Seizer, 1983; Shaughnessy,1977; Sommers, 1980; among others) say that unskilled writers, in general,are overly concerned with mechanics like spelling, punctuation, andgrammar in the process of writing. They frequently cut short planningbefore writing, and often tend to forget about text organization and theiraudience. Unskilled writers are generally inflexible in planning and revising,and seldom revise their texts beyond the word level.In terms of education, these researchers suggested that inexperiencedand basic writers could improve their writing if teachers were to pay moreattention to, and oversee, students’ writing processes as well as theirwritings. Pre-writing activities and discussions in the classroom (Hillocks,1982, 1986; Troyka & Nudelman, 1975), peer interaction and peer revising(Becker, 1986; Crowhurst, 1981; DiPardo & Freedman, 1988; Hillocks,1986), and one-to-one conferences with the teacher (“instructionalscaffolding”) (Applebee, 1984b) are suggested to have positive effects onstudents’ writing. These activities are meant to help students think throughthe problems of a specific writing task. Conversely, it is generally agreedthat teaching traditional grammar (such as parts of speech, terminology,parsing sentences) does not improve the quality of students’ writing(Braddock, Lloyd-Jones, & Schoer, 1963; Hillocks, 1986).Second language writing researchers (Cumming, 1988; Gaskill, 1986;Jones & Tetroe, 1987; Raimes, 1985, 1987; Zamel, 1983) say that thecharacteristics of first language writing processes are also found in thesecond language writing processes of skilled and unskilled writers. Referring13to Cummins’ theories, they say that writing skills developed in a learner’smother tongue will appear in second language writing.Cummins’ (1979, 1984, 1986) well quoted theories of the cross-linguistic interdependence of cognitive academic skills have been empiricallytested by many researchers in second language writing (Cumming, 1988,1989; Cumming, Rebuffot, & Ledwell, 1989; Jones & Tetroe, 1987; Raimes,1987). Cummins said that cognitive academic skills such as readingstrategies, writing composition skills, and higher-order thinking skillsdeveloped in a person’s first language (Li) would largely transfer to asecond language (L2) provided there is adequate exposure to L2 andadequate motivation to learn L2. Researchers in second language writingagree with Cummins. They found that writing skills in Li largely transfer toL2 writing and that second language general proficiency (mainly speakingskills) is not directly related to L2 writing skills.Cummins’ theories, however, have not been tested using secondlanguage learners’ translation performances. Research on translation in theframework of second language learning has been very inactive, probablybecause, as Krings (1987) suggests, the role of translation in foreign/secondlanguage teaching has always been “a matter of controversy” due to“simplistic theories” concerning the role of the mother tongue in secondlanguage learning (for review of literature, see the previous section).There are some preliminary studies on the translating performancesof second language learners (Gerloff,1987; Höischer & Möhle, 1987;Krings, 1987). These studies examine the translating processes of languagelearners, indicating that we cannot comment on the role of translation insecond language learning when we hardly know about learners’ cognitiveprocesses while translating. One study (Gerloff, 1987) reports that, in14translating a French (L2) text into English (Li), a professional translator inthe study paid more attention to discourse level analyses than studentparticipants did, and that one of the student participants relied heavily onmorphemic and syllabic analyses. Gerloff comments that good translators aredifferent from poor translators in such areas as size of units dealt with,editing styles, and patterns of movement through the text. She says that theprofessional translator in her study went through the whole text more thanonce and moved backward and forward constantly, whereas poor translatorstranslated small units (words and phrases) rather stumblingly and seldomwent through the whole text more than once. The bilingual professionaltranslator, unlike the student translators, maintained “her efforts at analysisnot primarily for comprehension purposes, but in the service of high-level‘production’ goals, that is, for determining the best way to express theoriginal source text in English” (p. 142).In sum, some characteristics of skilled and unskilled translators arevery similar to those of skilled and unskilled writers. It may be that secondlanguage writing and translating performances involve the same cognitiveprocesses, in which case, writing expertise in Li might transfer totranslating performances regardless of a learner’s second languageproficiency. Newmark (1981), a translation theorist, comments: “Atranslator has to know his own language, his subject and the target language-in that order. Excellence in the first requirement often saves him fromhideous mistakes in the second and the third” (p.141).152.1.4. Translation ModelsFrawley (1984) proposes the following translation model (Fig. 1),saying that “the translation itself is essentially a third code which arises outof the bilateral consideration of the matrix and target codes: it is, in a sense,a subcode of each of the codes involved” (p.168). B. Harris (1977) also saysthat the translation is a “third competence” in addition to some competence intwo languages.Figure 1 Translation as Third Code (from Frawley, 1984)Matrix Code >Target CodeI recodification\/New Code (the translation)Frawley’s translation model could be interpreted in Cummins’ (1984)“dual-iceberg” metaphor of bilingual proficiency, in which common cross-lingual proficiencies (cognitive academic skills) such as reading strategies,writing composition skills, and higher-level thinking skills underlie theobviously different surface manifestations of each language. The translationproficiency of the skilled (experienced) translator could be added to thecommon cross-lingual proficiencies (see Fig. 2). According to this concept,although the processes of writing and translating look quite different on thesurface level, they may involve the same kind of cognitive activities.16Figure 2 Translation Model of the Skilled TranslatorSurface Features of Li(Matrix Code) --Surface Features of L2•--->(Target Code)This model is reinforced by a few professional translators’comments: Seleskovitch (1976) says that translation is an interpretation of“thoughts and ideas”, which are “non-verbal”, and that especially inconsecutive interpreting, the translator relies on “semantic memory” ratherthan on “formal memory”, translating “meaning” in one language intoanother. Nida (1976) similarly says that transformational operations between17the surface and deep levels of the sentence are necessary for good translationbecause formal features on the surface level are often not reliable todetermine the meaning of the sentence (see also Chomsky, 1964, 1972).Carroll (1963) indicates that a good translator is “properly aware ofdifferences in linguistic codification” between two languages and takes thisinto account during translation because the structure of how one languageexpresses meaning is markedly different in another (see also Whorf, 1956).These comments are quite similar to those concerning the skilledwriter’s “translating” process in Li writing, which Flower & Hayes (198ia)and Hayes & Flower (1980) proposed. In this model, there are three mainprocesses in composing: planning, translating, and reviewing. Flower andHayes (1981a, p. 373) define the translating process in writing as “theprocess of putting ideas into visible language”. The writer “translates”meaning (non-verbal thoughts) into the verbal form. The process of“translating” in writing often resembles translating one language intoanother, because “there is no rigid correspondence between the units ofthought and speech” (Vygotsky,1934/1986, p.24.9), and “a symbolic functionexists (in mind) which is broader than language and encompasses both thesystem of verbal signs and that of symbols in the strict sense” (Piaget,1964/1967, p. 91). Although the processes of writing and translating lookquite different on the surface level, they do involve the same kind ofcognitive activities.However, translation by the unskilled translator, who may approachthe task word-for-word, or phrase-by-phrase, may not necessarily requirecognitive academic skills. That is, mechanical transposition from a sourceinto a target language is possible. Fig. 3 shows a one-way translation model,18which is based on Frawley (1984, p.161). Translations by the unskilledtranslator may be represented by this model.Figure 3 Translation as One-Way ActSource Language >Target LanguageThe translating performances of the unskilled translator may be, asApplebee (1982) comments, “writing without composing”. The unskilledtranslator may operate at the sentence-by-sentence or word-for-word level,just transposing the source language into the target language.The translation, thus, can be made at different levels ranging fromconcrete to abstract. Biggs and Collis (1982, pp.14-6-147) show that theFrench expression “sa table de null” could be translated as “his table of thenight”, “his table for the night”, “his table at night”, “his night table”, and“his bedside table”. They comment that the first translation (“his table of thenight”) shows a “direct one-to-one correspondence between words in twolanguages without consideration of any factor”, and that the last one (“hisnight table”) shows the “extended abstract” level, “considering not only theEnglish idiomatic expression, but also applying a relevant abstraction to givethe most appropriate translation of the phrase within the context”.In sum, the translating processes of skilled and unskilled translatorsmay be quite different. The translation processes of the skilled/trained19translator might be quite similar to the writing processes of the skilledwriter. The skilled translator does pay attention to the discourse level aswell as the word and phrase level while translating, negotiating between thesource and target texts.This research will study Cummins’ theories empirically, exploringhow translating and first/second language writing proficiencies areinterrelated. Since not many studies have dealt with the translating processesof second language learners by comparing them with L1/L2 writingprocesses, this exploratory study of second language learners’ translating andwriting processes may stimulate studies on translation in the framework oflanguage learning.2.2. Translation and Second Language Writing as Means ofLearning a Second LanguageThis study will also address translation and second languagewriting from a different perspective, underscoring the specific values textlevel translating and second language writing activities may have inenriching second language acquisition, and possibly widening the scope ofthe communicative approach to language learning.2.2.1. Translation in Language Teaching/LearningIn second/foreign language teaching, using translation as a methodwas thought ineffective and labelled as the “Grammar-Translation Method”20throughout the 1950s - 1980s (reviewed in Odlin, 1989; Stern, 1983).Second language pedagogy emphasized spoken communication during thisperiod, considering translation problematic, the cause of “interference” ofthe first language with the second. Many researchers and educators tried toeliminate translation from second language classrooms, or they totallyignored it (Asher, 1981; Chaudron, 1988; Dulay, Burt, & Krashen, 1982;Ellis, 1988; Krashen, 1981,1984; Terrell, 1977; among others). Auerbach(1993) reviews the “English only” belief in second language classrooms inthe United States from social and cultural perspectives, and confirms that thisbelief is still strong among monolingual ESL (English as a second language)teachers. She says (p. 10):Even official TESOL publications lend support to this (“Englishonly”) view with the publication of articles like a recent one(Weinberg, 1990) extolling the virtues of fining students for usingtheir Li. The author humorously tells her students, “This is anEnglish-only classroom. If you speak Spanish or Cantonese orMandarin or Vietnamese or Russian or Farsi, you pay me 25 cents. Ican be rich”.However, it is also true that quite a few researchers in the late 1980sand early i990s have expressed positive attitudes toward translation, urgingeducators to re-assess the potential contribution which translation could maketo language learning (Auerbach, 1993; Duff, 1989; Gerloff, 1987; Krings,1987; Sorhus, 1975; Titford & Hieke, 1985; Tudor, 1987a, 1987b;Weymouth, 1984; among others). Catford (1965, p. viii) commented in the1960s that “the chief defect of the now almost universally condemned“Grammar-Translation Method” was that it used bad grammar and badtranslation-- translation is not a dangerous technique in itself provided itsnature is understood, and its use is carefully controlled.” Catford (1981)21further claims that “translation, itself, is a valuable skill, and an importantmeans of refining one’s knowledge of a foreign language at an advancedstage of learning” (cited in Krings, 1987, P. 160).Snell-Hornby (1985), Titford (1985), and Tudor (1987a) alsocomment that translation is useful for advanced level learners. Titford(1985) says that translation is an activity that is “usefully engaged in after thebasic L2 communicative skills have been taught” (p.74). He says thattranslation is basically “consolidatory and facilitative”. Similarly, Tudor(1987a) says that translation is good for advanced students in order toexpand their L2 resources. His study shows that the inclusion of Li,compared to the use of L2 only, encouraged the students to use“achievement” strategies, which have positive effects on language learning.Tudor says that in order to present a speech in L2 (English), the studentswho worked “within their existing L2 competence, which naturally was notalways adequate for the expression of more or less complex ideas in aprecise way,” delivered speeches which were “clearly marked by a lesserdegree of precision and clarity” (p. 272) than those of the students who usedtheir Li (German) for the preparation stage. Uzawa and Cumming’s study(1989) also reports that intermediate level learners of L2 (Japanese), whenwriting in L2, often used their Li (English) for pre-writing planning, back-translating, vocabulary checking, and so forth, in order to keep up to thestandard of their Li writing.Butzkamm (i985) comments that the usefulness of translation is notrestricted to advanced learners. He says that “formal translation equivalentsare an elegant and economical way of helping the learner to see throughunaccustomed foreign language structures” (p.12). He believes thatutterances acquired as “prefabricated patterns” are unproductive, and that22only if the internal structure of a phrase is made transparent can the learneruse the language actively (see also Schmidt, 1983). Butzkamm thus says thattranslation is helpful in order to make the beginner see and manipulate theL2 structure more clearly. Sorhus (1975), too, comments that the use oftranslation by a teacher is helpful as “an important avenue to meaning” forbeginners.Edelsky’s (1982, 1986) one-year study on Spanish-speaking childrenlearning English as a second language reports that the children she studiedacquired second language writing skills using their mother tongue in abilingual setting. Auerbach (1993) argues that the “English only” policy inmany second language classrooms in the United States is often based on thesocial and cultural power structures (“English linguistic imperialism”),rather than on the learner’s needs and/or second language learning theoriesand research.Ways of using translation in the classroom, from the learner’sperspective, are described by some researchers. Weymouth (1984)emphasizes the importance of group/cooperative/peer discussions on texts ofnon-literary, up-to-date topics. He says that the “linguistic spin-off fromtranslation” should not only improve the learners’ L2 grammar but alsoteach them text structure and appropriateness of style to the social context inL2. Duffs (1989) book is written as a resource manual for instructors whoteach translation in the language classroom. His learner-centred approach totranslation emphasizes small group and pair activities in which translation isactively and positively designed to be used by learners in the classroom. Thebook helps the learner, through group and pair interactions, to perform oraland written translation tasks, emphasizing appropriate L1/L223correspondences not only in words and structures but also in register andcontext for better communication using the four language skills.In summary, the re-emergence of translation in language teachingmight have potential value for language learning at all levels of secondlanguage proficiency. Translation tasks for the advanced and/or intermediatestudents are good for refining, improving, and expanding their linguisticknowledge. Using translation at the beginner’s level is good for associatingand reinforcing clearly the L2’s structure and meaning. Translation activitiesenable the learner to use the target language generatively and creatively, andmake him/her aware of correct L1/L2 correspondences not only in wordsand structures but also in register and context.Vygotsky’s classical (1934/1986) views on foreign language learningalso seem to support translation activities as a means to learn a secondlanguage. He said that, in learning a new language, one does not repeat pastlinguistic developments; rather, a second language is learned by using one’sfirst language as a “mediator”. And that the advanced knowledge of one’sown language plays an important role in the study of the foreign one becauseforeign language learning uses the “semantics” of the native language as itsfoundation.2.2.2. Conscious Learning and the “i+l Output” HypothesisTranslation and second language writing may be very conscious waysof learning a second language. And this concept of “consciousness” has beena key issue in second language learning research since the 1980s. Schmidt(1990, 1993, 1994) says that conscious “attention is necessary for theconversion of input to intake” (p. 17, 1994). Schmidt (1990) reviews24theories on conscious and unconscious learning, and claims that “subliminallanguage learning is impossible, and that intake is what learners consciouslynotice. This requirement of noticing is meant to apply equally to all aspectsof language (lexicon, phonology, grammatical form, pragmatics)” (p. 149).Schmidt’s (1983) case study of an adult who was learning English as a secondlanguage through natural communication without formal learning shows thatthe adult’s acquisition of grammar and vocabulary through naturalcommunication alone was very limited. He just accumulated many“prefabricated patterns” or “formulaic expressions” without acquiring theability to use the language generatively and accurately.Schmidt’s view contrasts with Krashen’s (1981) theories, whichdistinguish second language “learning” and second language “acquisition”.According to Krashen, conscious “learning” does not promote language“acquisition”, which is unconscious. However, many researchers (B ialystok,1981, 1982; Ellis, 1986, 1994; McLaughlin, 1987; Tarvin & Al-Arishi,1991; Wenden & Rubin, 1987) disagree with Krashen, saying that thedistinction between conscious learning and unconscious acquisition is notnecessary and that “conscious” learning (such as learning grammar rules,contrasting Li and L2, paying attention, controlled processing, learningstrategies, reflection) helps the adult learner to acquire/learn a secondlanguage. Yalden (1975) also comments that “analytic study of the forms andstructure of the target language” is necessary in the teaching of secondlanguages so that adult learners are able to express their thoughts in L2 in “acareful, literate way, whether orally or in writing” (p. 338). Bialystok(1982) also says that analyzed linguistic knowledge,which becomesautomatic, is necessary for the language learner in order to reach the nativespeaker’s fluency level. Schmidt (1990) says that “those who notice most25learn most, and it may be that those who notice most are those who payattention most, as a general disposition or on particular occasions” (p. 144).Ellis (1993, 1994) similarly comments that “consciousness-raising” iseffective for language teaching.The concept of conscious learning may be reinforced by Swain’s “i+loutput” or “pushed output” hypothesis (1985). Swain argues that the learnerneeds to have chances to produce words, expressions, and syntax which are alittle higher than the learner’s present level (“i+i level output”) in order tolearn and improve the second language. She says that comprehensible inputor “i+l input” (Krashen, 1981) is not the process of translating and L2 writing, the learner is oftenforced to pay conscious attention to all linguistic areas such as grammar,vocabulary, syntax, register, and so forth, and tends to produce a piece ofwriting which uses more difficult words and expressions than in his/herspeaking. The concept of conscious attention in the context of the “i+loutput” hypothesis seems to be promising in second language learning. Iftranslation and second language writing are effective in learning a secondlanguage, this may be due to the learner’s conscious attention to L1/L2linguistic aspects in the process of producing an “i+l output”.2.2.3. RestructuringAlong with the concept of conscious learning and the “i+l out”hypothesis, the concept of “restructuring” is important in recent cognitivetheories of second language learning. Cognitive theories claim that the“reorganization” of grammar rules acquired wrongly and/or partially, and“restructuring” syntactic and semantic correspondences between Li and L2,26are necessary processes in language development (Ellis, 1986; Ijaz, 1986;Karmiloff-Smith, 1986; Lightbown, 1985; McLaughlin, 1987, 1990b).Cheng (1985) comments, from an information processing perspective,that improvement in performance in cognitive tasks such as solvingarithmetic problems (e.g., “Find the sum of ten 2s.”), can be “due to arestructuring of the task components” (p.414), rather than simpleautomaticity. That is, the learner performs better in the task by restructuringperceptual and cognitive units (from adding ten 2s to multiplying 2 by 10).McLaughlin (1990b) also comments as follows: “Restructuring ischaracterized by discontinuous, or qualitative, change as the child movesfrom stage to stage in development. Each new stage constitutes a newinternal organization and not merely the addition of new structural elements”(p.117).Similarly, in second language acquisition, when the learnerencounters new forms, he/she often restructures the whole system, ratherthan merely adding new forms to the existing system (Lightbown, 1985).Ijaz (1986) comments that “in many instances, a lexical item in the L2 cannotbe directly mapped onto a concept existing in the Li “; therefore, “the L2learner has to restructure existing Li concepts or develop a new concept thatcorresponds to a lexical item in the L2” (p. 405). In terms of translation,examples from Biggs and Collis (1982) may illustrate the restructuring ofL1/L2 correspondences. That is, the French expression “sa table de nuit”was translated as “his table of the night” or “his bedside table”. In the secondtranslation, “his bedside table”, there seems to be a restructuring of semanticand syntactic L1/L2 correspondences. The first one, however, shows notrace of restructuring. It is a direct one-to-one correspondence between27words in two languages. There is a qualitative difference between the twotranslations.Translation and second language writing may have value for learninga second language when the learner pays conscious attention to grammar,vocabulary, syntax, text, register, sociolinguistic aspects, and so forth, andtries to produce an “1+1 output”, and when this process leads to restructuringsemantic and syntactic correspondences between Li and L2. Butler-Nalin’s(1984) case study shows that ESL (English as a second language) learnersoften search for the words and the proper grammatical forms to expresstheir messages while writing (L2). This process may be facilitative forlearning a second language, although it may disturb the writing process.One of her participants comments:Sometimes I change some sentence, reorganize it. But I not sure somewords . . . I not sure its meaning. So I’m not sure it’s proper to use inthis sentence so I have to look at dictionary and check it.----Li, grade 9 (p. 129)Cumming (1990) also says that composition writing in L2 wouldhave value in second language learning if learners put effort into searchingout and assessing appropriate wording, while they compare their thoughtsand uses of language cross-linguistically (see also Ringbom, 1987).Cumming found that the frequency and quality of metalinguistic andideational thinking of adult second language learners while they write (L2) issignificantly related to learners’ writing expertise in their mother tongue. Hespeculates that the frequency and quality of cross-linguistic analyses whilewriting in a second language would have good effects for incidental learningof the second language.28Recent cognitive theories which advocate conscious learning andrestructuring support this study’s underlying assertion that translation andsecond language writing tasks are helpful for language learning. Translationand second language writing often involve conscious cross-linguisticattention to vocabulary, grammar, syntax, and discourse in order to producean “i+l output”, and this conscious focus may lead to a qualitative change inthe learner’s semantic and syntactic correspondences between Li and L2.Raimes (1983) says that second language writing helps students learnthe language because writing reinforces the grammatical structures, idioms,and vocabulary that have been taught in the classroom. It seems, however,that writing (and translation as well) often provides more than simplereinforcement in the language learning process. These tasks providelearners chances of producing a conscious “i+ 1 output” and restructuring ofLi and L2 correspondences, which may lead to second language learning.This exploratory study seeks to determine, by analyzing learner&conscious metalinguistic awareness and restructuring, how translation andsecond language writing tasks act as a means for second language acquisition.What learners think about translation and second language writing exercisesas a means of learning a second language, and how their positive or negativethinking is related to their translating/writing skill levels will also be ofinterest.2.3. Research QuestionsAs mentioned previously, this study explores translation and secondlanguage writing in two ways: 1) as components of language proficiency and2) as a means of language learning. The preceding sections have reviewed29the literature on translation and second language writing in the frameworkof language learning, highlighting these two perspectives of proficiency andacquisition.In the field of translation and second language writing as languageproficiency, there are many studies on skilled and unskilled writers insecond language writing (Cumming, 1988; Jones & Tetroe, 1987; Raimes,1985, 1987; Zamel, 1983; among others), but few on translation. Detailedstudies on the learner’s translating processes are scarce (Gerloff, 1987), andthere is a gap between theories of translation and studies on the learner’stranslating processes. Thus many questions may arise: How do languagelearners translate? How are translation skills related with first and secondlanguage writing skills? What views do language learners have ontranslation?Translation and second language writing from the perspective ofsecond language learning has seldom been studied empirically althoughresearchers and educators in the late 1980s touch on the prospects oftranslation and second language writing for language learning (Butzkamm,1985; Cumming, 1990; Duff, 1989; Raimes, 1983; Titford & Hieke, 1985).Concepts of conscious learning, “i+ 1 output”, and restructuring may leadtoward ascertaining why translation and second language writing may fosterlanguage learning, but no empirical studies on this subject are evident. Thusmany questions remain. Specifically, how often do learners pay consciousattention while translating and writing in Li and L2? How does theirconscious attention relate to their translating and writing skills? How dolanguage learners restructure L1/L2 semantic and syntactic correspondenceswhile translating? What do language learners think about translation andsecond language writing as a means of second language learning?30Considering these questions, the following six research questionswere formulated.1. How does the quality of second language learner& translationcorrelate with the quality of their Li and L2 writing?2. What do second language learners think about translation andsecond language writing tasks?3. What do second language learners pay conscious attention to whiletranslating and writing?4. How often do second language learners pay metacognitive andmetalinguistic attention while translating and writing?5. What characteristics are observed in the translating and writingprocesses of second language learners?6. How do second language learners restructure Li/L2 semantic andsyntactic correspondences while translating?(Regarding research question #4, see 3.1.2. for the definition of“metacognitive” and “metalinguistic”.)2.4. Methodological Issues2.4.1. Holistic ScoringTo answer research question #1, holistic scoring will be applied.Holistic scoring has been used extensively for evaluation in English writingeducation and research, and its strengths and limitations have been discussed(Charney, 1984; OdelI & Cooper, 1980; White, 1985). It is a highlyreliable procedure for sorting or ranking samples of writing qualitatively31within a reasonably short time, and it is suitable for measuring growth inwriting ability, or judging a student’s achievements in writing in reference toa test group (Diederich, 1966; White, 1985). However, it is not suitable fordiagnostic purposes, and it cannot represent an absolute value in itself (Odell& Cooper, 1980; White, 1985). Since the purpose of evaluating samples ofwriting for the present study is to find the relative levels of translation andcomposition proficiency of each participating student in relation to the groupof participants, and not to find diagnostic information or an absolute value ofeach participant’s translating and writing abilities, holistic scoring isappropriate here.The validity and reliability of holistic ratings have been challengedbecause holistic ratings are often influenced by superficial features such ashandwriting, spelling, lexical choice, grammar, and so forth, while contentand organization are not duly evaluated (Charney, 1988; W. H. Harris, 1977;Santos, 1988; Vann, Meyer, & Lorenz, 1984), and evaluators’ differentexperiences and disciplines may cause discrepancies in ratings (Mendelsohn& Cumming, 1987). In order to overcome these drawbacks in holisticscoring, some revisions have been made in the present study: (1) content,organization, and language use are evaluated separately using the scale forrating ESL compositions developed by Jacobs, et al. (1981); (2) chosenevaluators do not follow their own evaluation criteria (for the evaluationprocedures, see the next chapter); and (3) all the samples of writing aretyped so that participants’ handwriting does not influence the evaluation.322.4.2. Think-Aloud ProtocolsRegarding research questions, #2 to #6, the think-aloud techniquecombined with interviews, observations, and text analyses will be used inorder to analyze learners’ thinking and cognitive activities while translatingand writing.In order to analyze cognitive processes during translation, someresearchers (Gerloff, 1987; Hölscher & Möhle, 1987; Krings, 1987) havediscussed uses of think-aloud protocols, following research on writingprocesses (Hayes & Flower, 1983). They found that think-aloud protocolsare suitable for analyzing the cognitive processes involved in translation aswell.Think-aloud protocols (Ericsson & Simon, 1980, 1984) have beenused by many writing researchers, and pros and cons have been discussed.The merits of think-aloud concurrent verbal reports (compared toretrospective reports) are that they are relatively accurate and reliable,because they are less altered by the participant’s elaboration, evaluation,generalization, or memory loss than are retrospective reports (Ericsson &Simon, 1980, 1984; Hayes & Flower, 1983; Krings, 1987), and that theyare useful to detect some of the conscious strategies a writer uses to solveproblems (Bereiter & Scardamalia, 1987; Flower & Hayes, 1981b; Ode!!,Goswami, and Herrington, 1983). Hayes and Flower (1983) say thatconcurrent verbal reports can provide “rich” and “direct” evidence aboutcomposing processes.It is generally said that concurrent verbal reports do not affect theperformance of the task if the information is already in verbal form. Andeven if the information is in non-verbal form (e.g., visual images, motorimages), verbal reports do not change the quality of the performance; they33may just make the performance of the task slower (Ericsson & Simon, 1980,1984; Hayes & Flower, 1983).The limitations of concurrent verbal reports may be that “tacit” and“automated” knowledge (such as spelling, grammar, word choice in Liwriting) is usually not commented on in these kinds of reports (Ericsson &Simon, 1984; Hayes & Flower, 1983; Odell, Goswami, & Herrington,1983), and that not everyone feels comfortable doing two things (reportingand writing) at the same time. Krings (1987) comments that his “data showsthat individual differences between subjects with regard to their willingnessto verbalize might be greater than Ericsson and Simon seem to assume” (p.167). This tendency has been also detected in our preliminary studies asreported in the next chapter.Therefore, think-aloud protocols will be combined with otherprocess-tracing methods (interviews, observations, and text analyses) in thisstudy in order to overcome some of these drawbacks (see Odell, Goswami,& Hernngton, 1983). Also, a session of demonstration and practice ofthinking-aloud before the experiment will be provided for each participant(see Cumming, 1988).34Chapter 3METHODSThis chapter is divided into seven sections. The first section presentspreliminary studies, performed in order to field test the research tools. Thesecond section details the data collection procedures. The third sectionprofiles the characteristics of the participants. The fourth and fifth sectionsdescribe how transcribing, segmenting, coding, and rating of data wereperformed. The sixth section summarizes procedures for statistical anddescriptive analyses used to answer the research questions. And the finalsection reports the meeting with instructors at the College where participantswere recruited for this study.3.1. Preliminary Studies3.1.1. Case StudiesThree case studies on translation and second language writingprocesses were carried out from 1988 to 1990 at UBC (University of BritishColumbia) in order to field test the research design and give the tools forthis dissertation a trial run. All data collected in the case studies werederived from the think-aloud technique combined with interviews,observations, and text analyses, stressing the learner’s thinking and cognitivestrategies while translating and writing. The think-aloud data were collected35individually in a small classroom on the UBC campus. The presentresearcher (Uzawa) sat beside the participant, taking observational notes.The think-aloud utterances were tape-recorded, and then transcribed by theresearcher for analyses.The first case study involved four English-speaking university studentswho were studying Japanese as a foreign language at UBC. They were askedto write one English and one Japanese essay on the same topic, thinkingaloud in English. Eventually, this case study was published (for details, seeUzawa & Cumming, 1989). Within this study, it became clear that theamount of verbalization while writing varied more than expected from oneparticipant to another, as Krings (1987) suggested. Thus, the interviewsperformed right after the writing tasks in this case study have provenvaluable in order to grasp the participant’s thinking and cognitive strategiesperhaps not fully expressed in the think-aloud protocols.The second case study involved a Japanese graduate student at UBC,who was an experienced translator. He translated an English passage intoJapanese, and wrote an English short letter, thinking aloud in Japanese. Heseemed to be able to verbalize easily what he was thinking while translatingand writing, but commented later that he was able to verbalize just a smallportion of his thoughts. Here, this case study also reinforces the idea that afollow-up interview right after the writing session may indeed be useful inorder to supplement think-aloud protocols. The writing tasks in this casestudy (translation from L2 into Li, letter writing in L2) were admittedlyproblematic, and translation from Li into L2 may be a more appropriatetask to compare with L2 writing.The third case study was performed using two Cantonese-speakinguniversity students who were learning Japanese as a foreign language at36UBC. They wrote a short letter in Japanese and translated a short Englishpassage into Japanese, thinking aloud in English and Cantonese. The twoparticipants were not able to verbalize well due to a language gap. That is,the researcher did not understand their mother tongue, and English, ourlingua franca, was their second language. This case study showed that usingthe participant’s mother tongue, one which the researcher is able tounderstand, is important for a think-aloud task. As for the translation task,translating their second language (English) into their third (Japanese) wastoo complicated to analyze.In order to widen the possibilities of the think-aloud method, thepresent researcher asked several students who were studying Japanese atUBC in 1991 to write a Japanese composition on an assigned topic at homeover the weekend, thinking aloud and tape-recording their utterances. Thisexperiment was not successful because most participants summarized on tapewhat they did after having written the composition. Some students said thatthey usually generate ideas for a composition while taking a shower ordrinking coffee, when they write at home. Thus it was very difficult to tape-record “real time” thinking-aloud utterances in the natural environment.Consequently, the controlled environment of fixed-time writing in aclassroom with the researcher in attendance seemed a more direct route incollecting think-aloud data for this study.3.1.2. Pilot StudyA small scale pilot study using six Japanese students who werestudying English as a second language at the ELI (English LanguageInstitute), UBC, was performed in spring, 1992. The participants were all in37their twenties, and had been in Canada for less than one year. Six individualsessions (three hours each) were held on the UBC campus over two months.In the pilot study, some drawbacks of the think-aloud technique,which became clear in the case studies, were modified. For instance, ampletime to practise thinking aloud was provided for those who were not good atverbalizing their thoughts; semi-structured interview questions wereprepared and asked orally, in Japanese, immediately after each task; and theuse of the participants’ mother tongue for thinking-aloud was encouraged.Interview questions were modified after the pilot study in order to addressthe research questions more directly. In this way, the think-aloud techniquewas fine-tuned for the full-scale study.The participants were asked to do three tasks (translation from Liinto L2, Li writing, and L2 writing) while thinking aloud in their mothertongue. The topics of the three tasks seemed to be appropriate in the pilotstudy in terms of time, content, and levels of difficulty. All the participantsfinished the three tasks within the scheduled time, saying that the topics werenot too difficult. The three topics used in the pilot were kept for the full-scale study. (For the topics, see 3.2.5.)In the pilot study, it was extremely difficult to get participants whowere experienced translators. (There was only one experienced translatoramong the six participants.) In order to answer the research questions, itwas important to include several students who, if not experienced translators,were at least taking translation courses. Thus, students at CanadianInternational College (CIC) in North Vancouver, where translation ëoursesare offered, were asked to participate in the present study. It was alsoextremely difficult to find professional writers among the ESL students;38therefore, the research questions were modified so that a factor in Liwriting skill levels would not become a major focus.Our coding was roughly based on Gerloff’s (1987, P. 141) codingscheme (Table 1) in order to analyze participants’ metalinguistic attentionswhile translating and writing. However, after the pilot study, Gerloffslevels 1-5 were combined as one linguistic level because the number of ourparticipants’ utterances in each level was too small to calculate statistically,especially in the protocols of the Li and L2 writing tasks.The “metacognitive” attention was taken from Applebee (1982, p.372). Metacognitive attention includes strategies for retrieving andorganizing information and experiences relevant to the writing topic, as wellas strategies for narrowing the subject to be discussed.Table 1 shows the coding scheme used in the pilot study. Examplesare taken from Cumming (1988, 1990), in which French-speaking studentsare writing in their second language (English). Their utterances arereclassified according to our coding scheme. (Utterances enclosed in “ “ arethe created text.)In the pilot study, the participants’ utterances which could beclassified into more than one level were double or triple coded. For instance,the example presented in level 6 (discourse level attention) (see Table 1)could be coded as level 2 (word level attention) as well.39Table 1 Coding Scheme Used in the Pilot StudyLevel 1 -- morphemic or syllabic level attention (M)Breakdown or expansion of a word into syllables or morpheme unit. Also, attention tolcanji writing is included here. Example: “To fulfill”? “Fulfilling”? Ah, “to fulfill”? “Tofill”. “To fulfill.”Level 2 -- word level attention (W)Treatment of a word as a complete unit. Example: I’m looking for a word. “Not even arethey.. .“ “Are they. . .“ I was going to say “enable.” “Able”? Ah. I don’t want to say“able”. I don’t want to say “efficient”. Ah, I’m looking for words. “Are they.. .“ “Noteven are they able”? Ah. How do you say someone who has the qualification of doingsomething? Are they qualified? “Qualified” would be too much. No, no, that’s not what Iwant to say. “Capable”, yah, “capable”. Well, is it the same than “able”? Yah, okay.Okay.Level 3 -- phrase level attention (P)Processing of a group of words constituting a phrase. Example: No, not “the majority”.“What some people”? No. Okay, “what the majority”. No, not “the majority”.”Thepublic”? No. Okay, I’ll change it to “the traditional way”.Level 4 clause level attention (C)Processing of words in units containing a subject and verb alone; or subject and verb,plus complements. Example: “Courses who I prefer. . .“ I don’t know if I can say that.Des cours quej’aime. “The courses, who, that! really like are...”Level 5 -- sentence level attention (S)Processing a complete sentence as an entire unit, without breaking it down into smallerunits of analysis. Example: “But the couple. . . “Le couple doitparler. “The couple musttalk”.Level 6 -- discourse level attention (D)Clearly processing two or more sentences together, either by referring back to somethingread previously in the text while decoding another unit; by skipping ahead to anothersentence or paragraph in order to decode the unit being processed; or by reading two ormore sentences consecutively, without significant pausing. Example: “About the.. .“I’mtrying to find a way to start the other paragraph. Like, “about the other course”, or “as...“ I want to introduce the other paragraph. I can’t find the right word. I know in French,but I switch it in English, and it doesn’t fit there. Mm. How can I put it? Ah. Okay, I’lltry to imagine it’s in French. Pour l’autre cours. “For the other one”? No, it doesn’tsound. .. A propos? “About”? Um. “In concern”? “What concerns”? En ce qui concerne.I want to say en ce qui concerne “the other one” I’ll put “about”.Level 7-- metacognitive attention (MC)Strategies for retrieving and organizing information and experiences relevant to thewriting topic, as well as strategies for narrowing and focusing the subject to bediscussed. Example: So, I have to describe my courses, unha. Three points. If I enjoythe courses, what I learn in class, and other things I want to explain to you.403.2. Data Collection3.2.1. Recruiting ParticipantsCanadian International College (CIC), where the participants for thepresent study were recruited, is for Japanese high-school graduates, andacademic courses such as environmental studies, international business,academic writing, translation/interpretation, and so forth are offered inEnglish. The College has two campuses: one in North Vancouver and theother in Nelson, B. C. The first-year students have to study basic English atthe Nelson campus before coming to the North Vancouver campus.By recruiting participants from CIC (North Vancouver campus), thebackground of the participants was controlled naturally. That is, ages,educational and cultural backgrounds, emotional maturity, motivation, andlengths of residence in Canada were controlled. Furthermore, training intranslation was controlled as well because there were students who had beentaking the translation/interpretation course, and those who had never takenthe course. (For details about the characteristics of the participants, see 3.3.)A research proposal was submitted to the faculty of CIC in order toobtain the approval to recruit students at the College. In October, 1992,participants were recruited from six classes. These six classes had beenselected carefully beforehand with the assistance of the dean of the Collegeso that both students who had been taking the translation/interpretationcourse and those who had not could be recruited equally. (Memo to theinstructors is in Appendix A.) Fourth and second-year students weretargeted; all the first-year students were at the Nelson campus. Some thirdyear students were included in the fourth-year group because the enrollmentof fourth-year students at CIC was very small and, according to the faculty,41there was not much difference in English proficiency between the third andfourth-year students.Japanese was used to explain the project in the class, with oneexception, where the researcher spoke English and two students interpretedit simultaneously into Japanese as a part of the class activities. This lastedhalf an hour. (The instructor of the class and the researcher had arrangedthis plan beforehand.) Copies of recruiting letters in English and Japanese(Appendix B) were handed out to students in each class. The names andtelephone numbers of students who were interested in participating in theproject were taken from the six classes. Thirty-three students (6 male, 27female) volunteered altogether. The proposed research design for this studyneeded 24 participants (12 male, 12 female) as in Table 2, so 33 volunteersseemed to be sufficient.Table 2 Proposed Research DesignTraining in TranslationTrained Not Trained4th Year 6 (m=3; f=3) 6 (m=3; f=3)En2lish Proficiency2ndYear 6(m=3;f=3) 6(m=3;f=3)The researcher telephoned each student later to confirm his/herparticipation, and to arrange times for individual writing sessions. Twostudents cancelled. Twenty-two students (6 male, 16 female) were selectedout of the 31. Fourth-year students not trained in translation were difficult42to find because the number of fourth-year students at the College was verylimited, as mentioned earlier. It was also hard to balance the male vs. femaleproportion of the participants because the number of male students at theCollege was very small. Table 3 shows a preliminary classification of theparticipants. This classification was modified a little after having checked theparticipants’ backgrounds in detail. (For the final classification of theparticipants, see Table 7.)Twenty-two controlled background participants seemed to besubstantial enough for the think-aloud data collection; in the past most secondlanguage writing studies which used think-aloud protocols involved less than10 participants, and their backgrounds were hardly controlled (see alsocomments in Cumming, 1988).Table 3 Preliminary Classification of ParticipantsTrained in Translation Not TrainedFourth Year 6 (m=3, f=3) 4 (m=2, f=2)English ProficiencySecond Year 6 (m=O, f=6) 6 (m=1, f=5)The participants’ English proficiency level was judged by the amountof time spent learning English (see Stern, 1985), measured as the length ofresidence (LOR) in Canada (see Cummins, 1984, 1986; Cumming, 1989)plus the length of formal learning of English. Thus, their years at the43College (fourth and second year) were used to establish two levels ofproficiency. None of them had TOEFL scores.3.2.2. Pre-Writing SessionsAfter arranging writing sessions for each participant, a smallclassroom on the campus in North Vancouver was booked for the 22individual three-hour writing sessions. The researcher met with eachparticipant in the room during October and November, 1992. Sessions wereundisturbed by anyone else.Each participant was first asked to fill in a consent form to ensurehis/her willingness to participate (Appendix C). The content of the consentform was explained in Japanese and all participants received a Japaneseversion of the consent form. They were then asked to answer a questionnaireon their educational background (Appendix D). This whole procedure tookabout 10 minutes.Next, four questions on translation and second language writing taskswere asked orally, in Japanese, and each participant answered using a five-point scale. (For details, see the next section, 3.2.3.) This question-and-answer session also took about 10 minutes.Each participant was then asked to perform three tasks, thinkingaloud. Before the tasks, the researcher demonstrated how to think-aloudwhile writing for each participant, with the participant first supplying a topic(such as “my hobbies”) for the researcher. Then the participant practicedthinking-aloud while writing on the topic given by the researcher (“OurCollege”). The practice session took about 10 minutes.443.2.3. Interview QuestionsThe following prepared questions (Table 4) were asked verbally inJapanese in the pre-writing session. Also, open-ended questions were asked,in Japanese, regarding translation and L2 writing tasks for language learningby developing and extending the prepared questions (e.g., Why do you thinkthat second language writing tasks are very helpful/not so helpful forlearning a second language?). The distinction between “learning” and“improving” in these questions was explained to the participants so that“learning” meant beginners to intermediate level acquisition, and“improving” meant advanced level acquisition in language learning.Table 4 Interview Questions1. Generally speaking, translation tasks are helpful for learning a second language. Howstrongly do you agree or disagree with this statement?1 2 3 4 5strongly disagree strongly agree2. Generally speaking, translation tasks are helpful for improving a second language. Howstrongly do you agree or disagree with this statement?1 2 3 4 5strongly disagree strongly agree3. Generally speaking, second language writing tasks are helpful for learning a secondlanguage. How strongly do you agree or disagree with this statement?1 2 3 4 5strongly disagree strongly agree454. Generally speaking, second language writing tasks are helpful for improving a secondlanguage. How strongly do you agree or disagree with this statement?1 2 3 4 5strongly disagree strongly agree3.2.4. Writing SessionsThree tasks (translation, Li writing, L2 writing) were assigned. (Fortopics, see the next section.) They were randomized so that the order wouldnot affect the performance. Each participant was told to write a first “draft”while thinking aloud. All were encouraged to use dictionaries while writingand translating. One hour each was allocated for the translation and for theL2 writing tasks, and half an hour was given to Li writing. The suggestedlength for the Li and L2 writing was one-page each, double-spaced. Moststudents finished their three tasks in less than the allocated time. Nobodyexperienced “writer’s block” in the writing session. Each participant wasasked not to tell other participants about the topics of the three tasks.The researcher was sitting just beside the participant, takingobservational notes, while he/she was performing the three tasks. Think-aloud utterances were recorded using a small tape-recorder. The researcherbasically remained silent, just nodding from time to time, responding to theparticipant’s comments and questions. With some participants who tended toforget to verbalize, the researcher occasionally reminded them to verbalizeor asked the participants questions (e.g., “What are you thinking now?”).The following semi-structured, open-ended questions (Table 5) wereasked in Japanese right after each task in order to supplement the think-aloud46protocols. The questions are overlapped intentionally in order to maximizeeach student’s answers. This question session took several minutes.TabJe 5 Questions on Tasks1. Where did you pay attention most in this task?2. What do you think is the most important aspect of this task?3. What was the most difficult aspect of this task?4. If you were asked to revise this text for publication later, how and what would you like torevise?In order to express gratitude for the students’ participation, anhonorarium (gift certificate) was paid to each participant after the writingsession, and a copy of the short summary of the study (10 pages in typedJapanese) was sent to each participant in January, 1993, when data collectionand analyses were completed.3.2.5. Writing TopicsThe following are topics for the three writing tasks. (They weretested in the pilot study prior to the present study.) A length of one-page(double-spaced) was suggested for each composition.471) Translation (Japanese into English)An article from the Nihongo Journal (April, 1990), which is forlearners of Japanese as a second/foreign language, was used for this task(Appendix E). This article was selected because the text is easy to read formost native speakers of Japanese, while translating the text into English ischallenging due to some words and expressions which are hard to translateinto English. In addition, the content is general and understandable withoutspecific background knowledge, and the length is appropriate for a one-hourtask. The publisher’s English translation (Appendix F) is attached to the text,so this was used as a guide/model translation for evaluating the students’translations.2) Li writing (Japanese)The topic was taken from Jones and Tetroe (1987): Describe the mostdifficult adjustment that you have had to make living in Canada. In theirresearch, this topic had been used as an Li (Spanish) writing task forSpanish-speaking ESL students in Canada. The present study selected thistopic because it would be contextually appropriate for Japanese ESL studentsin Canada.3) L2 writing (English)The topic was also taken from Jones and Tetroe (1987) and modifieda little according to our needs. (In the present study, the one word“Venezuelan” in Jones and Tetroe was changed to “Japanese”): What is themost important difference between Canadian and Japanese society? In theirresearch, this topic had been used for the L2 (English) writing task for thesame Spanish-speaking students mentioned above, who were from Venezuela.48This topic seemed appropriate for the present study as well because ourparticipants had been studying in Canada for 1.5 to 3.5 years, and seemed tobe familiar with Canadian society.These last two topics were chosen from topics which Jones and Tetroeused for their research on the second language writing processes of Spanish-speaking ESL university students. These topics are, as Jones and Tetroeindicate, generally used for conventional writing tasks usually given to ESLstudents in their writing courses or on English-proficiency tests.In this study, different topics were intentionally assigned for the twowriting tasks in order to avoid some unwanted influences of ordering. Thatis, if the same topic is used, the first task may facilitate the second task. It isconsidered that the difference in the two topics for the Li and L2 writingwould not greatly affect the participants’ writing performance, based onCumming’s (1988) comment that the level of cognitive demand (e.g.,argument vs. personal letter writing), rather than simple difference in topics,affects the learner’s L2 writing performance. Since both topics in the presentstudy require “descriptive” writing and a specific background knowledge isnot necessary, the level of cognitive demand in the two writing tasks (ofcourse, L2 writing is more demanding than Li writing, but as far as thetopics are concerned) seems equal. Likewise, the content of the translationtask is also “descriptive”, and does not require any specific knowledge.Thus, the level of cognitive demand in the three tasks (again, as far as thetopics and content are concerned) is similar.493.3. Characteristics of Participants3.3.1. BackgroundsTwenty-two Japanese students were selected from the thirty-one whovolunteered to participate in the study. All of them were learning English asa second language at the same College through studying academic subjects inEnglish. They stated that studying English would enable them to get jobs inJapan or enter Japanese universities. Relevant aspects of participant&backgrounds are presented in Table 6. Pseudonyms are used throughout thisstudy in order to maintain the confidentiality of the participants.All of the participants had studied in Japan for 12 years (up to thehigh-school level) before coming to Canada. Five students had studiedEnglish two more years in Japan at college, university, language school, orprep school. One student (Takako) had studied in Argentina under a highschool exchange program for 10 months. None of them had taken TOEFL atthe time of this study.The participants ranged in age from 19 to 23, and had been in Canadafor 1.5 to 3.5 years. None of the participants had professional experience inwriting or translation. Nor did they have work experience, writing ortranslating letters and memos.50Table 6 Backgrounds of ParticipantsParticipant Gender Year Age LOR Educational Length of Trainingat CIC in Canada background in Translation at CIC(years) (years)1. Takako F 4 23 3.5 high-school 2.52. Takeshi M 3 21 2.5 high-school 1.53. Takuya M 3 21 2.5 high-school 1.54. Tami F 4 21 3.5 high-school 2.55. Taro M 3 23 2.5 prep school 1.56. Toshie F 4 22 3.5 high-school 2.57. Saburo M 3 21 2.5 high-school none8. Sachiko F 4 22 3.5 high-school none9. Satomi* F 2 22 1.5 2-year college none10. Sawako F 4 23 2.5 2-year college none11. Setsuko* F 2 22 1.5 university (2 years) none12. Shiro M 4 22 3.5 high-school none13. Shizuko* F 2 22 1.5 language school (1 month)14. Madoka F 2 19 1.5 high-school 0.515. Makiko F 2 19 1.5 high-school 0.516. Man F 2 20 1.5 high-school 0.517. Miki F 2 20 1.5 high-school 0.518. Momoko F 2 20 1.5 high-school 0.519. Naoko F 2 20 1.5 high-school none20. Nami F 2 20 1.5 high-school none21. Nobuo M 2 20 1.5 high-school none22. Noriko F 2 20 1.5 high-school none*Moved from the second-year group (See details in 3.3.2.).513.3.2. Classification of ParticipantsThe participants were classified initially, as in Table 3 (section3.2.1.), according to their training in translation at CIC and their Englishproficiency (LOR in Canada and years at the College). However, afterchecking their backgrounds more closely, the preliminary classification wasmodified a little: three students in the second year (Setsuko, Satomi, andShizuko) were moved to the fourth-year (not trained) group because theyreported that they had studied English in college, university, and languageschool, respectively, for two years after high-school (See Table 6 in the,previous section). Thus, years at the College do not necessarily correspondto the years counted in the modified classification. (In the case of Shizuko,she reported at the time of recruiting that she was taking CIC’s first-yeartranslation course, but she had just started the course in September while theother second-year “trained” students had been taking this course since April.She was therefore disqualified as a “trained” student.) This newcategorization seemed appropriate as the three students’ mean scores in thethree tasks were closer to the mean scores of the 4N group than to the 2Ngroup.Participants in the “trained” group had been taking thetranslation/interpretation course at the College for some time at the point ofbeing interviewed (October, 1992); the students in the fourth-year group hadbeen studying translation and interpretation for 1.5 to 2.5 years, and thesecond-year group for half a year. All the “trained” participants, however,had never undertaken translation professionally. The participants in the “nottrained” group had never taken the course at the College and none had anyprofessional experience doing translation.52All the participants in the fourth-year (trained) group mentioned inthe interview that they wanted to use the training in translation andinterpretation as a skill for the workplace, although they had no intention oftranslating or interpreting, professionally, in the future. All the second year(trained) students said that they selected the course because the coursedescription stated that students would get to know Canadian people andsociety through training in translation and interpretation. They said they wereinterested more in interpretation than in translation. Most students in the“not-trained” group said that they did not enroll in the course because theyhad no desire to be translators or interpreters.Table 7 is the modified (final) classification of the participants.Pseudonyms used here correspond to the classification:Names start from “T” belong to the fourth-year, trained in translation (4T);“S” “ fourth-year, not trained in translation (4N);second-year, trained in translation (2T);second-year, not-trained in translation (2N).53Table 7 Modified Classification of ParticipantsTrained in Not-Trained inTranslation Translation4th-year 6 (m=3) 7 (m=2) I(f=3) (f=5)2nd-year 5 (m=O) 4 (m= 1) I(f=5) (f=LITrained Not TrainedTakako Saburo*Takeshi* SachikoTakuya* Satomi4th- Year Tami SawakoTaro* SetsukoToshie Shiro*_________________ShizukoMadoka NaokoMakiko Nami2nd-Year Man Nobuo*Miki NorikoMomoko*male student543.4. Transcribing, Segmenting, and Coding of ProtocolsAll the tape-recorded think-aloud protocols (approximately 50 hours)were transcribed by the researcher during the two months from October toNovember, 1992. However, Setsuko’s think-aloud protocols of thetranslation task were erased by mistake; this was treated as missing data inthe statistical calculation.Segmentation was roughly based on previous studies (Cumming,1988;Ericsson & Simon, 1984; Flower & Hayes, 1981b): pause (a silence of fiveseconds or more before and after the statement), intonation, and meaningwere used in this study in order to distinguish stretches of verbalization fromone another.Coding was performed using the coding scheme (Table 8) developedfor the present study, revised from the coding scheme used in the pilot study.The five linguistic levels (morphemic, word, phrase, clause, and sentencelevels) in Gerloff’s study (1987, p. 141) were combined into one as thelinguistic level (L). Her discourse level was retained as the discourse level(D). The category “Personal comments” (P) was added to this study becauseparticipants often expressed their personal comments (metacomments) whileperforming the task (e.g., “I’m sorry, my handwriting is very awful.” “Oh,I’ve forgotten to double-space.”) The metacognitive level (M) was, asmentioned in the section about the pilot study, taken from Applebee (1982).Examples of utterances in Table 8 are taken from the data collectedfor the present study, in which our students were speaking in their mothertongue (Japanese). Examples of Japanese utterances are enclosed with < >,55and the English translations by the researcher which follow the Japaneseutterances are presented in ( ). Participants’ created text (not necessarilywritten down) is indicated by” “. (More examples of transcribed and codedthink-aloud protocols are in Appendix G.)Table 8 Coding SchemeMetacognitive level attention (M) -- content generation, attention to audiencee.g. <“kanadajin to nihonjin no shakai no chigai”, “kanadajin wa chotto individual,nihonjin wa group ishiki” tte kaite...(writes down key words)> (“difference betweenCanadian and Japanese society”., “Canadian people are a little individualistic, Japanesepeople have a group consciousness”, let’s write this.... (writes down key words)) -- Shiro,English taskDiscourse level attention (D) -- attention to text organization, logical flow,attention to more than two sentences at the same timee.g. <kono tsuzuki doo nagarete ikoo kana’tte omotte....> (how to continue from here, Iwonder...)-- Madoka, English taskLinguistic level attention (L) -- attention to single sentences, clauses, phrases,words, spelling, punctuatione.g. <“regard” to “respect” kana, kono chigai’tte yuuno naka naka wakari nikui...>(“regard” and “respect”, it’s hard to tell the difference of the meaning...) -- Nobuo, EnglishtaskPersonal comments (F) -- metacomments, questions, personal chatting, etc.e.g. <jisho tsukatte ii desu ka’?> (may I use the dictionary?) -- Sawako, Translation taskThe inter-rater reliability assessment on coding was performed by anindependent judge on 182 randomly selected utterances (10 percent of the56protocols). The reliability was very high (.92). The intra-rater reliability on118 another randomly selected utterances was also high (.9 1).3.5. Evaluation of Writing SamplesThe quality of the participants’ written texts was evaluated by twoindependent judges (in each language) who were native speakers andexperienced language teachers / evaluators / editors, using holistic scoringand the ESL Composition Profile developed by Jacobs, et al. (1981) (seeAppendix H). In the present study, however, the ESL Composition Profilewas modified a little: a 4-point scale was used and the three categories of“vocabulary”, “language use”, and “mechanics” were combined into one as“language use” (see also Cumming, 1988). That is, two judges evaluatedwriting samples using a 4-point scale (max=4, min=1) in three categories(content, organization, language use). In the case of the translation, only onecategory (language use) was evaluated because content and organization weregiven by the text assigned. The English evaluators evaluated participants’translations referring to a model translation in the Nihongo Journal(Appendix F). The sum of scores (max=8, min=2) produced by two judgesfor each task was used for statistical calculation.Two independent evaluation sessions (one for each language) wereheld. The English language evaluators gathered in a classroom on the UBCcampus, and the Japanese language evaluators at the residence of one of theevaluators. The two judges in each language had practice sessions (about 30minutes) before evaluating whole samples. (Two sets of writing samples ofthe three tasks had been prepared beforehand for the two judges in each57language.) They first read 10 randomly chosen papers quickly and rank-ordered them; then they evaluated two randomly selected papers out of theten, using the modified ESL Composition Profile. Any ambiguous pointregarding the evaluation criteria, etc. was clarified during the practicesession. The inter-rater reliability on 6 items (2 papers x 3 categories) was.94 in English and .89 in Japanese in the practice session.The two judges in each language then evaluated all of the writingsamples as follows: 1) first, they sorted the writing samples into four groups(high / mid-high / mid-low / low), judging the quality of the samples byreading them quickly, estimating that each group should have 25% of thesamples; 2) they then evaluated the samples in three categories (content,organization, and language use) using the modified ESL CompositionProfile. They did a re-evaluation when the evaluation discrepancy betweenthe two evaluators was greater than 1. The two judges discussed thedifferences in rating before the re-evaluation. About 15 papers were reevaluated in Li writing, 8 in L2 writing, and 2 in Translation. The wholesession of evaluation (practice, evaluation, discussion, and re-evaluation)took about two and half hours for the English session (translation and L2writing samples), and one and half hours for the Japanese session.Each participant’s name and his/her language fluency level were notknown to the evaluators, and all the papers had been typed by the researcherbefore the evaluation so that handwriting would not influence the evaluation.Evaluators had been informed, before evaluating the samples, thatparticipants created only first “drafts” in the three tasks while thinkingaloud. Thus the evaluators knew that the participants in this study producedfirst, and not final, drafts and that the writing situation was very artificial.This consideration was necessary because students’ composing behaviours58and quality of writing may vary according to writing situations (Hall,1991).The inter-rater reliability on evaluation assessed by the Pearsonproduct-moment correlation coefficient (r) on 22 papers in each task ispresented in Table 9. (Note: As mentioned earlier, the translation taskinvolved only one category (language use) because content and organizationwere given by the text assigned.)Table 9 Inter-Rater Reliability on EvaluationTranslation Li Writing (Japanese) L2 Writing (English)r r rContent--- .903 .697Organization--- .887 .696Language Use .678 .959 .773Total*--- .952 .835(*Total is the sum of scores on Content, Organization, and Language Use.)3.6. Procedures for AnalysesThe data collected for this study (writing samples, think-aloudprotocols, interviews, observational notes) were analyzed in the context ofthe research questions. Not only the four groups (4T, 4N, 2T, and 2N) werecompared, but also group pairs (second vs. fourth year; trained vs. not59trained) and the whole group of 22 participants was compared across threetasks, depending on the research questions, as described below.In order to answer the research question #1 (How does the quality ofsecond language learners’ translation correlate with the quality of their Liand L2 writing?), the rank order correlation of the quality of the 22students’ writing samples involving three tasks (Translation vs. Li;Translation vs. L2; Li vs. L2) was calculated using the Pearson product-moment correlation coefficient (r) and the Spearman correlation coefficient(rho). Raw scores produced by the judges were then converted intopercentages and the four groups (4T, 4N, 2T, 2N) were compared in eachtask using ANOVA (analysis of variance).The research question #2 (What do second language learners thinkabout translation and second language writing tasks?) was answered by theprofile analysis using the participants’ answers (scores on a 5-point scale) tothe questions discussed in 3.2.3. The parallelism among the profiles for thefour groups and difference among response means were calculated. Theprofile analysis was complemented with a qualitative analysis of theinterviews.The research question #3 (What do second language learners payconscious attention to while translating and writing?) was answered byanalyzing the 22 participants’ answers to the questions asked directly aftereach task (For the questions, see Table 5, Section 3.2.4.). The answers werecategorized into four writing aspects (content, organization, language use,and audience) and how many students of the 22 paid attention to these aspectsin the three tasks was counted.For the research question #4 (How often do second language learnerspay metacognitive and metalinguistic attention while translating and60writing?), the think-aloud utterances were analyzed. This question wasapproached using a factorial analysis of variance with four factors: Attention(4 levels) x Tasks (3 levels) x Training (2 levels) x Year (2 levels).The research questions #5 and #6 were answered descriptively. Theobservational notes were analyzed to answer research question #5 (Whatcharacteristics are observed in translating and writing processes of secondlanguage learners?). For #6 (How do second language learners restructureL1/L2 semantic and syntactic correspondences while translating?), theparticipants’ translations were analyzed referring to the Japanese text and themodel translation of the assigned text in order to examine the restructuringin detail. Their transcribed think-aloud protocols were also analyzed.3.7. Meeting with InstructorsAfter analyzing the data, a 15-page summary was submitted to theCollege faculty in January, 1993, requesting their comments, opinions,and/or questions. It was important to ask instructors’ views on the researchfindings in order to counterbalance the researcher’s views, which are basedstrictly on the created experimental setting. (see Guba & Lincoln, 1989). Thefaculty kindly arranged a one-hour discussion session, which was held on theCIC campus in May, 1993, and was attended by about 15 faculty members.The session was tape-recorded with the permission of the attendinginstructors. Because of their insights based on classroom experience andobservation, their comments and opinions were very informative. Findingsin this study, which were limited in the experimental setting, were thenexpanded.61The instructors’ comments and opinions on findings in this study arereported in the next chapter along with the researcher’s interpretations of thefindings.62Chapter 4RESULTS AND DISCUSSIONIn this chapter, the results of descriptive and statistical analyses arejuxtaposed with the six research questions guiding this study. Findings foreach research question are reported in separate sections. In every section, themain findings are summarized first, and this is followed by detailed analysesand discussion of findings.4.1. Results for Research Question 1: How does the quality of secondlanguage learners’ translation correlate with the quality of their Li and L2writing?4.1.1. SummaryThe statistical analyses for research question #1 were not significant.However, in the case of the quality of the students’ language use, although itwas not significant, there were weak correlations between the translation andL2 composition, and between the Li and L2 compositions. As well, groupcomparisons of mean scores of the quality of the three tasks showed that the4T (fourth-year, trained in translation) students performed better than thestudents in the other three groups in the translation and L2 writing tasksalthough the difference was not strong enough to reach statisticalsignificance.63In most studies of second language writing in which professional andstudent writers are compared, it has been said that skills in Li writinglargely transfer to L2 writing. However, in this study, in which second andfourth-year college students, not professional writers, were compared, theevidence for this skill transfer was not clear.4.1.2. AnalysesAs reported in section 3.5, two independent evaluators in eachlanguage (English and Japanese) evaluated the participants’ writing samples.They did not know the participants’ English proficiency levels andbackgrounds. The English evaluators evaluated the translations and L2compositions.The raw scores on each participant’s writing samples in the threecategories (content, organization, and language use) are presented inAppendix I. The scores are out of 8 (max=8, min=2). That is, the sum of thetwo raters’ scores (max=4, min=1) are used. (Some of the students’ writingsamples are attached in Appendix J.)The raw scores in Appendix I were calculated for correlationalrelations (Translation vs. Li; Translation vs. L2; Li vs. L2), using thePearson product-moment correlation coefficient (r) and Spearman (rho).The results are summarized in Table 10. As mentioned before, thetranslations were evaluated only on language use because content andorganization are given by the assigned text.64Table 10 Results of Correlational AnalysesPearson Spearmanr rhoContent (Li vs L2) .282 .174Organization (Li vs L2) .174 .126Language Use(Li vs L2) .472* 448*(Trans vs Li) .004 .045(Trans vs L2) .365* .510**weakly correlatedAccording to this table, correlational relations between Li and L2Writing (content and organization) and Li writing and Translation (languageuse) of the participants are very low and not significant. However, thecorrelations between Li and L2 Writing (language use), and Translation andL2 (language use) are relatively high, although they are not significant.Because the correlational relations of the quality of the individualparticipant’s performances in the three tasks were found to be very weak andnot significant, the raw data (Appendix I) were examined from a differentperspective in order to seek possible interrelations among the four groups.The raw scores were converted into percentages so that the quality ofwriting (content, organization, and language use, combined) in the threetasks could be compared. The proportions for the three categories (i.e.,content, organization, language use) were weighted as follows (see Jacobs, etal., 1981; Kikuchi, 1987).Translation = Language Use 100%Li Writing = Content 50%, Organization 20%, Language Use 30%L2 Writing = Content 30%, Organization 20%, Language Use 50%65The proportions for the three categories in L2 Writing were based onJacobs, et al. (1981) (see Appendix H). This proportion seems to beappropriate because most second language learners often feel that the mostdifficult part in L2 writing is vocabulary (Silva, 1992; Uzawa & Cumming,1989). In the case of Li Writing, the proportion was decided by the twoJapanese evaluators and the researcher. That is, considering differences inLi and L2 writing, the weight on language use in L2 Writing (50%) wasreduced to 30% for Li. The translation task was evaluated on language useonly; thus 100% was based on language use.The scores in percentages were examined by ANOVA. The differencein the quality of translation and writing in the four groups (4N, 4T, 2T, 2N)was not strong enough to show a statistical significance, as shown in Table11. However, there was a tendency for the fourth-year students (4T and 4N)to perform better than the second-year students (2T and 2N) in the Liwriting task (p = 0.0589). As well, there was a tendency for the trained (intranslation) groups (4T and 2T) to perform better than the non-trainedgroups (4N and 2N) in the L2 writing task (p = 0.0778). In the translationtask, the 4T group performed slightly better than the other three groupsalthough the difference was not significant. Overall, the fourth-year trained(4T) group performed better than the other three groups in the three tasksalthough the difference was not strong enough to be significant. The fourthyear not-trained (4N) group did best in the Li writing task, but did notperform well in the translation and L2 writing tasks.66Table 11 Group Comparisons of Quality of Writing SamplesTranslation Li Writing U Writing( of Students Means Means MeansTrainingTrained ii 70.98 % 6539 % 67.39 %Not-trained 11 69.57 63.73 59.89Year4th 13 70.65 70.18 66.062nd 9 69.74 56.45 60.15Year x Training4T 6 74.08 69.65 71.152T 5 67.26 60.28 62.884N 7 67.70 70.63 61.702N 4 72.85 51.66 56.74Total 22 70.28 64.56 63.64Translation:Training x Year: F (1, 18) = 1.78 p = 0.20Training: F (1, 18) = 0.01 p = 0.93Year: F(1,i8)=0.03 p=O.8SLi Writing:Training x Year: F (1, 18) 0.47 p = 0.50Training: F (1, 18) = 0.30 p = 0.59Year: F (1, 18) = 4.07 p = 0.0589*L2 Writing:TrainingxYear: F(1, 18)=0.16 p=O.7OTraining: F (1, 18) = 3.50 p = 0.0778*Year: F (1, 18) = 2.52 p = 0.13*marginally significant674.1.3. DiscussionA difference of one or two years in school learning, as in this study,may not show a difference in the quality of translation and L2 writing if wefollow Cummins’ theories. Contrary to face-to-face communicativecompetence in L2, cognitive academic skills such as L2 writing and readingskills develop very slowly, taking several years (Cummins, 1984).Specifically, in the context of academic writing, Cummins’ hypotheses(1979, 1984, 1986) of cross-linguistic interdependence of cognitive academicskills were not confirmed clearly in this study. Cumming (1988) providesevidence that Li and L2 writing skills are interdependent, comparingprofessional and student writers. He claims that writing skills in Li largelytransfer to L2, saying that professional writers’ expertise in paying attention,simultaneously, to more than one aspect of writing and problem-solvingwhile writing is transferable from Li to L2 writing. No student in this studyhas had professional experience and expertise in Li writing. Thus theirtranslation, Li writing, and L2 writing skills might not have shown strongcorrelations, therefore appearing modular and independent.In addition, the target language here (English) is linguistically verydifferent from the learners’ mother tongue (Japanese). Learning a languagequite different from Li may take three to four or five times longer thanlearning a language similar to Li (e.g., Indo-European languages similar toone another) (Odlin, 1989; Ringbom, 1987; Stern, 1985). As well, it maytake additional time to acquire translational competence as “translationalcompetence is an interlingual competence, it is clearly marked off from thefour traditional monolingual skills: listening, speaking, reading, and writing”(Wilss, i976, p. 120).68The levels of cognitive demand within the tasks may not have beenappropriate to differentiate the skill levels of the participants: Cumming(1988) reports that the cognitively demanding tasks in his study (argumentand summary writing) produced a great difference in the quality of writingbetween the expert and student writers, whereas in the cognitively lessdemanding task (letter writing), the difference in the quality of writingbetween expert and student writers was small in his study.It is likely that the three tasks in the present study were cognitivelydemanding enough for the participants. The participants’ average scores inthe three tasks were less than 80%, indicating they were not easy tasks forthe participants.Technically speaking, the number of participants may not have beenlarge enough for a statistical analysis of score ratings. However, in thisstudy, each student did not necessarily gain consistent scores across the threetasks, and the standard deviation was high. For instance, scores of the six 4Tstudents in the Li and L2 writing (content) were, as shown here, notconsistent across the two tasks with high standard deviation.Li (content) 4.2 7.0 8.0 4.3 3.5 7.0 (means = 5.7; s = 1.88)L2 (content) 4.0 5.0 7.0 5.5 6.5 6.0 (means = 5.7; s = 1.08)Thus, even if more participants were added, the correlational relations andthe difference in the four groups for the three tasks might still have been thesame; i.e., it might not have produced a significantly different profile.Some instructors at the College mentioned in the discussion sessionthat class hours spent for actual translation exercises (Li into L2) at theCollege were very limited: one hour a week, or less than a few hours aweek, at most, over 33 weeks (one academic year). Therefore, “trained” and69“not-trained” distinctions in translation (especially for the second-yearstudents) seemed inappropriate.The instructors further said that the English proficiency of newstudents from Japan entering the College was improving each year becausethe reputation of the College had become known in Japan, thus attractingmore “good” students. The difference in English proficiency between thefourth and second-year students in this study, therefore, may have actuallybeen very slight.In sum, correlational relations among the translation, Li and L2writing performance of college ESL students who were not professionalwriters and translators were not significant. Likewise, group comparison ofperformance in the three tasks showed no significant difference among thefour groups.4.2. Results for Research Question 2: What do second languagelearners think about translation and second language writing tasks?4.2.1. SummaryThe results of research question #2 showed that the participants in thisstudy generally regarded both translation and L2 writing tasks as relativelyhelpful for learning and improving a second language. The participants wereasked four questions about their thoughts on translation and L2 writing tasksas a means of learning and improving a second language, and their answerswere recorded on a five-point scale. The profile analyses based on theseanswers showed no significant difference in thinking among the four groups.70However, in the interviews, the students in the 4T group expressedvery positive views toward translation tasks, while most students in the otherthree groups (4N, 2T, 2N) preferred L2 writing tasks to translation. Inaddition, many students indicated that speaking practice is also necessary forlearning and improving a second language.4.2.2. AnalysesAs stated in 3.2.3., the participants were asked four questionsregarding translation and L2 writing tasks as a means for language learningas follows (partially duplicated here from Table 4).1. Generally speaking, translation tasks are helpful for learning a secondlanguage. How strongly do you agree or disagree with this statement?2. Generally speaking, translation tasks are helpful for improving a secondlanguage. How strongly do you agree or disagree with this statement?3. Generally speaking, second language writing tasks are helpful forlearning a second language. How strongly do you agree or disagree withthis statement?4. Generally speaking, second language writing tasks are helpful forimproving a second language. How strongly do you agree or disagree withthis statement?The distinction between “learning” and “improving” was explained tothe participants: “learning” meant beginners to intermediate levelacquisition, and “improving” meant advanced level acquisition in languagelearning.The participants answered on a five-point scale (5=very helpful,3=neutral, 1=not helpful at all) and explained in the interview their reasons71for choosing the particular number. These answers on a five-point scale aresummarized in Table 12 (raw scores) and Figure 4 (group means of theiranswers for each question in graph form).The statistical analyses showed no significant difference among theprofiles for the four groups. The multivariate test of parallelism among theprofiles for the four groups (4T, 4N, 2T, 2N) gave a p-value of 0.12. Aswell, there was no difference among response means. In addition, the groupeffect was equal (p=O.3l) over all four questions.Although it was not statistically significant, the 4T (fourth-year,trained in translation) group seemed to have a somewhat different viewtoward translation and L2 writing, compared to the other three groups, asshown in the graph (Figure 4). The students in the 4T group thought thattranslation is very helpful (means = 4.67 - 4.25) for learning and improvinga second language, and that L2 writing is less helpful (compared totranslation) for learning and improving a second language (4.33- 3.50).Inversely, the students in the other three groups (4N, 2T, 2N) thought thatL2 writing is more helpful (means = 3.70- 4.20) than translation (3.30 -3.75) for learning and improving a second language.Statistically speaking, however, the students in the four groupsgenerally regarded both translation and L2 writing tasks as relatively helpfulfor learning and improving a second language. Their positive answerstoward using translation for language learning might be related to theparticipants’ educational background. In Japan, translation exercises inEnglish classrooms are very common. If participants are from othercountries, where translation exercises are seldom performed, the results mayhave been quite different.72Table 12 Raw Scores of Participant& Answers to Questions on Translation and L2Writing as a Means for U Learning and Improvement5 very helpful; 4 helpful; 3 = neutral; 2 = not helpful; 1 not helpful at all.Not-Trained inTranslationTranslation for L2 ImprovementTrained in Not-Trained inTranslation Translation4th Year 445555means 4.67 3.64 4.25 3.502ndYear 3 5 2 43.5 3 3 33 2.5 4 3.3 3 3.5 54 4means 3.30 3.38 3.30 3.75L2 Writing for L2 Learning L2 Writing for L2 ImprovementTrained in Not-Trained in Trained in Not-Trained inTranslation Translation Translation Translation4thYear 4 5 4 54 4 2 45 4 4 54.5 3 3.5 35 4 4 43.5 4 3.5 42 2means 4.33 3.70 3.50 3.862ndYear 4 4 3 33.5 4 4 44 4 4 54.5 3 4 4.55 4means 4.20 3.75 3.80 4.13Translation for L2 LearningTrained inTranslation54.53433343554.54435422.5473Figure 4 Profile of Participants’ Thinking on Translation and L2 WritingLegend: 4th-Year, Trained in translation (4T) A — A4th-Year, Not-trained in translation (4N)2nd-Year, Trained in translation (2T) * — *2nd-Year, Not-trained in translation (2N)5 = very helpful; 4 = helpful; 3 = neutral; 2 = not helpful; 1 = not helpful at all54TA—43..21Translation Translation L2 Writing L2 Writingfor learning L2 for improving L2 for learning L2 for improving L24.2.3. DiscussionAlthough the statistical analyses did not show a significant differenceamong the four groups, the results of the interviews show that students in the4T group have a very positive attitude toward translation as a means oflearning and improving a second language. For example, Takeshi said that*0/74translation is more helpful than L2 writing for improving expressions in L2because one has to use new words and expressions in addition to those whichone already knows in order to translate a text, whereas in L2 writing, onecan write using only very basic words. Toshie and Tami said that translationis more helpful than L2 writing because one has to think about Li/L2correspondences in translation tasks. Taro said that he often experienced intranslation tasks new discoveries in language use. Takuya said that he couldlearn differences in culture and expressions in Li and L2, throughtranslation. Takako, who was somewhat negative toward translation, said thattranslation and L2 writing may be helpful, but speaking skills cannot beimproved by translation and writing exercises.Thus five out of six students in the 4T group said in the interviews thattranslation tasks are very helpful for learning and improving a secondlanguage. Their comments might be summarized as follows: in L2 writing,one can usually write essays using only words and expressions which he/sheknows well, without checking dictionaries. Thus there is not muchopportunity for improving the second language, whereas in translation, aperson has to use words and expressions which are beyond his/her level, andalso has to consider the L1/L2 differences in language use and cultures inorder to translate the original text. This process is helpful for learning andimproving a second language.These comments are consistent with Swain’s (1985) “pushed output”hypothesis in language learning. Swain argues that the learner needs to havechances to produce words, expressions, and syntax which are a little higherthan the learner’s present level (“i+i level output”) in order to learn andimprove the second language. She says that comprehensible input orinput” (Krashen, 1981) is not enough.75Swain (1985) compared L2 speaking and L2 writing, and commentedthat French immersion students in her study performed better in writingFrench than speaking it because the students had often produced “i-i-i output”in L2 writing. For the 4T students in this present study, however, L2writing, compared to translation, was often perceived as an “i -1” level task.Other students, especially second-year students, however, said that L2writing is more helpful than translation. For instance, Saburo (4N) said thatan assignment such as diary writing is more helpful than translation tasks forlearning L2; translation tasks are good for improving accuracy after havingacquired speaking fluency. Noriko (2N) said that checking dictionaries whiletranslating often leads to mechanical and direct translations; L2 writing isbetter because one can think in English directly. Nami (2N) said that goodtranslations do not necessarily mean improvement in L2. Momoko (2T) saidthat translation is not efficient for L2 learning because words andexpressions used for translations are not often used in everydayconversation.For the not-trained (in translation) students, translation often meansusing the dictionary all the time, whereas in L2 writing, they can expresswhat they really want to say in their own words. For the not-trainedstudents, translation may have often meant an “i+2” level task.Many students mentioned that practice in speaking is also important toimprove second language skills. Nobuo (2N) said that writing and speakingare separate skills. Nami (2N) indicated that writing itself is not helpful forimproving speaking. Miki (2T) said that words in speaking and writing aredifferent. Marl (2T) said that writing may be helpful for speaking, but thattranslation is not because content is already provided. Even the 4T studentssaid that translation practice is not enough for language learning, if speaking76practice is not included. These comments are very astute; the students haveinsights into language learning. They know that writing practice alone willnot facilitate speaking, and vice versa. Their answers are consistent withCumming’s (1988) comments that speaking and writing are indeed separateskills.4.3. Results for Research Question 3: What do second languagelearners pay conscious attention to while translating and writing?4.3.1. SummaryThe results of research question #3 were analyzed by classifying theparticipants’ answers to interview questions into four categories: content,organization, language use, and audience.Contrary to our expectation (see Cumming, 1990; Uzawa andCumming, 1989), half of the participants in this study did not pay consciousattention to “language use” in L2 writing. However, all of the participantswere conscious about “language use” in the translation task. Attention to“content” and “organization” was very high in the writing tasks (both Li andL2), and relatively high in the translation task. The number of participantswho paid attention to “audience” was very small.4.3.2. AnalysesIn order to answer research question #3, the following questions(presented here from Table 5, section 3.2.4.) were prepared. Each studentwas asked these questions in Japanese following every task.771. Where did you pay attention most in this task?2. What do you think is the most important aspect of this task?3. What was the most difficult aspect of this task?4. If you were asked to revise this text for publication later, how andwhat would you like to revise?These questions were overlapped intentionally in order to maximizeeach student’s response: i.e., even if a participant missed giving thoroughanswers to one question, he/she still had a chance to give answersadditionally within the framework of other questions.For example, Takako’s (4T) answers for these questions after thetranslation task were as follows:1. I tried to translate the meaning of the text, and tried not to translateword for word.2. To translate the point, i.e., to translate what the writer wants to sayis important.3. It was difficult to find appropriate English words for Japanesewords like zazen and dantai seikatsu because English equivalents forthese words listed in the dictionary did not seem to fit into thiscontext.4. I would like to revise my grammar (especially prepositions andtense), and also would like to ask a native speaker to check whethermy English expressions are natural.These answers were categorized using four writing aspects (content,organization, language use, and audience), which are commonly recognizedin writing pedagogy. Takako’s answer to question 1 was categorized into78“content”; question 2, “content”; question 3, “language use”; and question4, “language use”. Altogether, Takako’s conscious attention in thetranslation task was judged to be directed to “content” and “language use”.Similarly, Noriko’s (2N) answers in the translation task were:1. I tried to transfer the content/information accurately.2. To convey information as much as possible.3. Vocabulary. Although I checked the dictionary, it was not clear.Oftentimes, I had to create my own English equivalents for someJapanese words, and this bothered me.4. I would like to check for grammar, words (whether they aretranslated accurately), and accuracy of content.Noriko’s answers were categorized as follows: question 1, “content”;question 2, “content”; question 3, “language use”; question 4, “languageuse” and “content”. Noriko’s conscious attention in the translation task wasalso directed to “content” and “language use”.Regarding the L2 writing task, Takako’s response was:1. I paid attention whether what I wanted to say was organized well; Itried to organize what I wanted to say logically.2. Content generation.3. I was not able to generate examples well. I should have generatedenough examples and ideas before writing.4. I would like to add some more concrete examples. Also, I wouldlike to check my grammar.79Takako’s answers were categorized as follows; question 1, “organization”;question 2, “content”; question 3, “content”; question 4, “content” and“language use”. Takako’s conscious attention in the L2 writing task wasdirected to “content’, “organization”, and “language use”.Noriko’s answers after the L2 writing task were:1. I paid attention to whether the content was appropriate, i.e., I triedto answer the topic directly and faithfully.2. Faithfulness to the topic.3. To generate ideas was difficult, because I lack information aboutthe Japanese educational system.4. I would like to check the structure (introduction, body, conclusion),and the appropriateness of my examples, and comparisons.Noriko’s response to question 1 was categorized into “content”; question 2,“content”; question 3, “content”; and question 4, “organization” and“content”. Noriko’s conscious attention in the L2 writing task was directed to“content” and “organization”.Likewise, other participants’ answers to questions posed after theyfinished the three tasks were analyzed and categorized using the samemethod. All participants’ answers concerning each task are summarized inTable 13. This table shows how many students reported to have paidattention to certain writing aspects during the three tasks. It indicates thestudents’ beliefs about where their writing’s attentions should be, whiletranslating and writing. This does not mean that the students themselves paidattention to these aspects. Table 13 is presented graphically in Figure 5.80Table 13 Participants’ Reported Conscious Attention in Three Tasks(Figures are out of 22 students.)Translation Japanese (Li) English (L2)Content 16/22 19/22 16/22Organization 13/22 20/22 22/22Language 22/22 17/22 11/22Audience 9/22 7/22 5/2281Figure 5 Participants’ Reported Conscious Attention in Three Tasks# of Students Legend: * * * Translation+++ Li Writing22 0 * 0 0 o L2Writing*20 + *+ + *+ + *+ +*+ + *±15 + *+*+ + *+*+ *+ * +*+ * +*+ * + *+10 *+ * +*+ * + *+ ** + *+ ** + *+*+ * + *+ *+5 * * + *+ *°*+ * + *+ *+0*+ * + *+*+ * + *+0 *+ * + *+ *+°Content Organization Language Use Audience4.3.3. DiscussionIn this study, it was very surprising that the number of students whopaid attention to “language use” in L2 writing was fewer than in Li writing.Most students in the interviews said that they wrote the English compositionusing very easy words and expressions. Indeed, they seldom used dictionarieswhile writing in English. In the previous section (research question #2),many participants commented that they usually wrote essays in English using82only words and expressions which they knew well, without checkingdictionaries. Thus, they would appear to write by using an “avoidance”strategy which sidesteps difficult vocabulary and structures (see Schachter,1974).For language acquisition, conscious attention to language is important(Schmidt, 1990). Cumming (1990) and Raimes (1983) suggest that L2writing is facilitative for language learning due to the learner’s consciousattention to language use. However, our data show that half of theparticipants did not pay conscious attention to “language use” in L2 writing.It was in the translation task that all of the students paid conscious attentionto “language use”. For research question #2 (in the previous section), our4T students maintained that compared to translation, L2 writing is not aneffective exercise for improving a second language. They said that usingwords and expressions which are beyond one’s level, and considering theL1/L2 differences in language in order to translate the original text are bothhelpful for learning and improving the second language. If language learnerswho have attained some face-to-face oral communicative skills habituallywrite L2 compositions using only words and expressions which they knowwell, there seems to be little opportunity to improve the L2.Regarding “content” and “organization”, it is often said that studenttranslators do not pay attention to content and organization in translationbecause these are given by the text to be translated (Applebee, 1982).However, the data indicate that more than half of the students were veryconscious about content and organization, even in the translation task. (Forinstance, Noriko (2N) answered that she tried to transfer thecontent/information accurately in the translation task. See the previoussection.) This table does not contend that they actually paid attention to these83aspects while translating, but it shows that at least half of the participantsbelieved that it is necessary to pay attention to content and organizationwhile translating.Regarding “audience”, more than half of the participants did not payattention to the audience in the three tasks, and those who paid attention tothe audience were mostly the fourth-year students. However, even thestudents who said that the reader was important were not able to givesatisfactory answers to such questions as how to take care of the reader. Lackof conscious attention to audience and lack of “procedural knowledge”(Anderson, 1983) about audience seem to characterize the translating andwriting processes of the student writers who do not have much writingexperience.4.4. Results of Research Question 4: How often do second languagelearners pay metacognitive and metalinguistic attention while translating andwriting?4.4.1. SummaryResearch question #4 was approached by analyzing the think-alouddata using a factorial analysis of variance.The results were that the attention patterns of the participantsdepended on the task. That is, the participants’ attention patterns in the Liand L2 writing tasks were very similar, but they were quite different in thetranslation task. The participants’ metacognitive attention in the writing tasks(Li and L2) was relatively high, but it was very low in the translation task.The participants’ attention to language use in the translation task was very84high, but it was unexpectedly low in the L2 writing task. Their consciousmetalinguistic attention may be related to the quality of their language use;the participants’ scores on the language use in the translation task werehigher than in the L2 writing.4.4.2. AnalysesResearch question #4 was approached by collecting ‘think-aIoud”speech data. Each student was asked to talk aloud while translating andwriting, and their utterances were tape-recorded. Taped utterances weretranscribed, segmented, and coded by the researcher according to theprocedures described in section 3.4. (Transcribed and coded think-aloudprotocols are in Appendix G.) The coding scheme used in this study appearsin Table 8 in section 3.4. (A simplified version is presented here.) The inter-rater reliability was high (.92 on 182 randomly chosen utterances), and theintra-rater reliability was also high (.91 on another 118 randomly chosenutterances).M: metacognitive level attention (e.g., content generation, attention toaudience)D: discourse level attention (e.g., attention to text organization,attention to more than two sentences at the same time)L: metalinguistic level attention (e.g., attention to single sentences,clauses, phrases, words, spelling, punctuation)P: personal comments (e.g., metacomments, personal chatting, questions,etc.)85The utterances of each participant in each attention level across threetasks were counted separately and converted into percentages. The use ofpercentages was necessary for the statistical analyses because the number ofutterances was different from participant to participant. The number of eachparticipant’s utterances in one level in one task was divided by the number ofhis/her whole utterances in one task. The participants’ whole utterances inthree tasks ranged from 8 to 67 (means 27.1) in the translation task, 4 to 44(means 17.1) in the Li writing, and 4 to 54 (means 18.3) in the L2 writing.The data include four factors: Attention (4 levels: M, D, L, P), Task(3 levels: Translation, Li, L2), Training (2 levels: Trained, Not-trained),and Year (2 levels: 4th-year, 2nd year). In order to answer researchquestion #4, the data were analyzed using analysis of variance with fourfactors (4 x 3 x 2 x 2). The results are shown in Table 14. Whenutterances could be coded into more than one category, double or triplecoding was used. Thus, the total is more than 100%. The data in Table 14are presented in graph form in Figure 6, in which differences of attention inthe three tasks are shown visually by presenting means of the four attentionlevels (metacognitive, discourse, metalinguistic, metacomments) of the 22participants.According to the statistical analyses, there were very significantinteractions in Attention x Task (p = 0.0000), Attention x Year (0.0223), andAttention x Training x Year (0.0 147). However, there were no significantinteractions in Attention x Task x Year (p = 0.2 194), and Attention x Task xTraining (p = 0.1385). As well, an interaction in the four factors (Attentionx Task x Training x Year) was not significant (p = 0.458). That is, the 22participants’ attention patterns in the three tasks were quite similar. The86participants, regardless of their training and year, paid attentiondifferentially depending on the tasks. The participants’ attention patternswere very similar in the Li and L2 writing tasks, but significantly differentin the translation task. Every participant in the four groups paid significantlyvery high metalinguistic attention but low metacognitive attention in thetranslation. (This tendency corresponds to the results of research question #3in the previous section.) Figure 6 visually shows the participants’ very highmetalinguistic attention and low metacognitive attention in the translationtask, and the similar patterns of attention in the Li and L2 writing.87Table 14 Metacognitive and Metalinguistic Attention As Measured byFrequency (%) of UtteranceM = metacognitive level attentionP = discourse level attentionL = linguistic level attentionp = personal commentsFigures are percentages (%). Group means are shown.meansTranslation Li Writing L2 WritingSource of variation df Sum Of squares Mean square F pAttention 3 12646.96 4215.66 18.05 0.0000Attention x Task 6 25716.23 4286.04 18.35 0.0000Attn x Task x Train. 6 2300.83 383.47 1.64 0.1385Attn x Task x Year 6 1954.81 325.80 1.39 0.2194Attn x Task x Train x Yr 6 1337.36 222.89 0.95 0.4580Attention x Training 3 1720.35 573.45 2.46 0.0649Attention x Year 3 2300.57 766.86 3.28 0.0223Attn x Train. x Year 3 2526.94 842.31 3.61 0.0147M D L P M U L P M U L P2N 0 9.8 77.2 15.5 47.0 24.1 28.3 28.5 32.1 35.7 21.1 43.42T 0 13.9 62.2 33.5 35.5 18.2 24.5 59.1 31.6 14.7 35.2 39.34N 0 10.3 76.1 17.6 39.6 23.0 28.7 39.0 55.4 23.0 17.1 29.84T 8.7 19.7 71.6 15.6 32.3 20.8 40.4 24.3 47.6 19.6 35.8 27.78.7 13.7 71.7 20.4 38.0 21.8 30.8 37.7 43.6 20.5 27.1 33.888Figure 6 Participant& Mean Attention Levels in Three Tasks Measured byPercentage of Utterance100%Translation: A—A Li Writing: L2 Writing: * — *90%80%70%60%50%40%30%*A0%metacognitive discourse level metalinguistic personallevel attention attention level attention comments894.4.3. DiscussionThe participants’ attention levels measured by proportion of frequencyof utterance relating to the four categories are quite similar in the Li and L2writing, but not in the translation. In the Li and L2 writing, the percentagesof the utterances on the metacognitive level (M) are high because everyparticipant talked before writing about the content to be written. In thetranslation task, however, the metacognitive level is very low. (Only onestudent in the 4T group talked about the audience of the translated text.) Bycontrast, the percentages of the participants’ discourse level (D) utterancesare almost the same in the three tasks. Participants paid attention to theparagraphs and the groups of sentences in the translation task as well as inthe writirig tasks. The participants’ personal comments or metacomments aresimilar in the three tasks, although they are a little lower in the translationtask. The personal comments, such as comments on handwriting, writinghabits, personal experiences, and so forth, however, are not directly relatedto research question #4, but make up a part of the percentages.The percentages of the utterances on the linguistic level (L) isextremely high in the translation, and it is lower in the L2 writing than inthe Li writing. In the L2 writing, most students wrote English very quicklywithout checking their dictionaries, although their English vocabulary andgrammar were very limited. They certainly had a writing fluency in L2, buttheir language use was not very articulate. Also, they did not elaborate ordevelop their ideas fully; thus, most students just wrote a half page, finishingwriting in half an hour. By contrast, in the translation task, most participantsspent one full hour translating the assigned text, examining words andexpressions which they did not know well. Nobody skipped sentences in the90translation task. (For the students’ writing and translation samples, seeAppendix J.)Cumming (1988) reports that expert writers (such as journalists andnovelists) in his study paid “conspicuous” attention to words while writing inL2 in order to overcome their lack of second language proficiency. Ourstudents did not pay enough attention to language use in the L2 writing task,probably because they lacked the necessary Li writing experience and theirwriting skills are still developing. In the present study, it was in thetranslation task that the participants paid conspicuous attention to words andgrammar in order to express the meaning of the text assigned (Li) in thesecond language.The participants’ conscious attention to language and grammar in thetranslation task seems to be related to the quality of their language use. Thetwo English evaluators remarked that the translations were much better thanthe English compositions when they first read several translation papers forevaluation, after having finished evaluating the L2 compositions. And theresults of the evaluators’ actual evaluation of the quality of the students’language use was 10.88% higher on average in the translation than in the L2writing task (Table 15). This is a statistically significant difference (p =0.0006; s = 12.01). Particularly, the students in the non-trained groups (4N& 2N) up-graded their language use in the translation task more so than thetrained groups (4T & 2T) as the scores of L2 writing of the non-trainedgroups were lower than the trained groups.In general, the participants’ English expressions were more vivid andcolorful in the translation than in the L2 writing; details there wereexpressed with varied vocabulary and syntax. An English native speaker,91whom the researcher asked to read the students’ writings, similarlycommented as follows:The translations were, overall, far more precise and logical than theL2 writings. There was a lot more structure to the translation piecesand that writing sounded like it had a purpose. However, most of theL2 writing rambled and digressed and was confused by lack ofstructure. The L2 writing was plagued by too many undevelopedideas.The results of this study are consistent with some studies ontranslation and L2 writing. Kobayashi & Rinnert (1990), comparingcomposition via translation and direct composition, say that Japanese EFL(English as a foreign language) students in their study produced significantlybetter L2 (English) compositions by writing via translation (from Japanese)than by writing directly in L2. The students in the low proficiency group intheir study, when compared to the students in the high proficiency group,especially benefitted from translation.Similarly, Tudor (1987a) reports that the work of the students whoworked within their existing L2 (English) competence was “clearly markedby a lesser degree of precision and clarity” than that of the students who usedtheir Li (German) for the preparation of a speech in L2 in his study.In these studies, the same topic was used for direct writing andwriting via translation. In the present study, however, two different topicswere used for the translation and L2 writing tasks. By assigning differenttopics, the present study was able to avoid an obvious influence of the firsttask on the second task. It was also able to avoid the problem of translatingmaterial which was far beyond the participants’ English proficiency level92(translating one’s own Li composition is often too difficult for lowproficiency level second language learners).While the degree of difficulty between the two tasks in this study maynot have been exactly the same, this fact does not significantly affect theresults. Cumming (1988) shows that it is the level of cognitive demand (e.g.,argument vs. personal letter writing), rather than simple difference in topics,that affects the learner’s L2 writing performance. Since the content of thetranslation task was “descriptive” and the topic of the L2 writing alsorequired “descriptive” writing, there is reason to believe that the level ofcognitive demand in the two tasks in this study was not significantlydifferent.In the interview, some students said that the translation task was moredifficult than the L2 writing due to the number of many unknown words.Some other students said that the translation task was easier than the L2writing task because they were not required to generate their own ideas intranslation. However, most students said that they did not see muchdifference between the two tasks and topics.Generally speaking, in the translation task, the participants were freedfrom the cognitive activities of generating and organizing ideas, and wereable to concentrate on linguistic activities. Thus, their use of language andgrammar may have been qualitatively better in the translation than in the L2writing.93Table 15 Comparison of Scores on Language Use in Translation and L2 Writing(Scores are sum of two evaluators: max =8, mill = 2)L2 Writing Translation Difference L2 Writing Translation Difference(Trans. minus L2 Writing) (Trans. minus L2 Writing)4T 5.0 5.8 +10.0% 4N 5.5 6.7 +15.0%5.0 5.5 +6.25% 4.0 4.8 + 10.0%6.0 5.7 -3.75% 3.0 6.0 +37.5%6.0 6.5 +6.25% 5.5 6.4 +11.25%5.5 6.05 ÷6.88% 3.5 4.5 + 12.5%6.0 6.0 0 4.5 5.5 + 12.5%5.0 4.0 -12.5%means +4.27% +12.3%2T 4.5 5.0 +6.25% 2N 3.5 5.5 +25.0%5.5 4.0 -18.75% 4.0 5.3 +16.25%5.5 6.5 +12.5% 5.5 7.0 +18.75%4.5 5.8 ÷16.25% 4.0 5.5 +18.75%4.0 5.6 ÷20.0%means +7.25% +19.7%944.5. Results of Research Question 5: What characteristics are obsen’edin the translating and writing processes ofsecond language learners?4.5.1. SummaryResearch question #5 was approached by observing the 22 participantsindividually while they were translating and writing (Li and L2). Thefollowing characteristics were noted. 1) Many participants used the“sentence-by-sentence” approach in the translation task without reading thewhole assigned text before translating. As well, they did not re-read theirtranslated texts after having finished translating. 2) In the writing tasks (bothLi and L2), most students used the “what-next” approach. That is, beforewriting, they generated ideas, but they did not organize these in any way ordevelop them further to form a unified text before actually writing. In sum,the students may have had “theoretical” knowledge, but lacked “practical”knowledge, of translation and (LI and L2) writing.4.5.2. AnalysesThe observational notes taken while each participant was performingthe three writing tasks were mainly used to answer research question #5.(For the procedural details, see section 3.2.4.) Recorded observations ofparticipants’ writing behaviors were then analyzed.In the translation task, i2 of the 22 students started translating thetext sentence by sentence without reading the whole text first. After havingtranslated the passage, 13 students read back the whole translated Englishtext quickly, but only three students took time to revise. Many students inthis study believed that paying attention to “content” and “organization” is95important in translation (see the results of research question #3, section4.3.), but this was not put into practise well in the actual translating process.Their translating processes may be summarized as “writing withoutcomposing” (Applebee,1982), or “translation as a one-way act” (Frawley,1984, see section 2.1.3.).When an experienced translator translated the same text for thepreliminary study, she read the assigned text very slowly, underlining someexpressions, considering the possible audience, and then translated the text in10 minutes. However, after translating the text, she spent 50 minutes revisingthe translation and said that, if possible, she would like to revise it further.She paid a great deal of attention to the audience of the translated Englishtext, indicating that words and phrases as well as the text organization of theEnglish text should be natural, comprehensible, and clear to a reader whomay not know the cultural background of the original text.In this study, however, only one fourth-year, trained student paidattention to the likely reader of the translated text, and consequently insertedsome explanatory phrases not in the original text, in order to make the piecesensible to a reader who might not know Japanese social customs. Payingattention to the reader, both in the translation and writing tasks, was rare inthe participants’ actual processes as well as in their thinking patterns (see theresults of research question #3).The results of the thinking aloud in the previous section (4.4) showthat the percentages of the participants’ metacognitive attention in the Li andL2 writing was quite high in comparison with the translation task. However,the real picture was that most students generated ideas, haphazardly, aboutwhat they wanted to say, only for a few minutes before writing, withoutthinking of strategies: i.e., how to organize generated ideas or how to argue96a point. In this study, they did not think beforehand about logical flow. Bothin the Li and the L2 writing, the so-called “what-next” approach waswidespread. Oftentimes, “introduction body-conclusion”, “topic sentence”,“thesis statement”, and so forth were mentioned, but were not implementedas concepts by which to organize and develop the text.The revising pattern was similar in the Li and L2 writing as well.After writing, most students did not reread their own text. Some students reread, but only once, revising some minor points, such as spelling. Somestudents said that they were afraid of rereading because they did not knowwhat to do if they discovered errors.One instructor at CIC commented that their students, when assignedessay writing in L2, would often finish their writing very quickly, withoutspending enough time generating ideas, organizing and re-organizing thestructure, reflecting upon the topic, and revising their own text. Thus, itseemed that the general characteristics of the participants’ writing processesrevealed in the think-aloud session did not greatly differ from their usualwriting behaviors in the classroom.4.5.3. DiscussionIn general, the characteristics mentioned above are quite similar to theprofile of the “basic” or “inexpert” writer documented in many researchpapers (Bereiter & Scardamalia, 1987; Emig, 1971; Flower & Hayes, 1980;Per!, 1979; Sommers, 1980; for example). Bereiter and Scardamalia (1987)say that expert writers usually can “transform” their knowledge as they writeby rethinking and restating ideas until they become fully developed thoughts,but inexpert writers just “tell” their knowledge as they write by simply97stating ideas without planning or goal-setting. Most students in this study,through their knowledge of composition, knew terms like “brain-storming”,“outlining”, “thesis statement”, “topic sentence”, “introduction-body-conclusion”, and so forth, and mentioned these often during writing, but theyused these concepts as formulaic prescriptions and were not actually able to“transform” their ideas using these concepts. This knowledge gap is alsosimilar to Ryle’s (1949, pp. 25-61) distinction between “knowing how” and“knowing that”, Anderson’s (1983, p. viii) distinction between “declarativeknowledge” and “procedural knowledge”, and Mohan’s (1986) “theoreticalknowledge” and “practical knowledge”. The participants in this study mayhave “declarative” or “theoretical” knowledge about composition, or knowfacts about writing, but they lack “procedural” or “practical” knowledge”,or knowledge of how to perform the task. In short, they lack writingpractice.As shown in this study, second language learners who lack experiencein writing cannot write as they might wish, even in Li, due to the gapbetween “declarative” and “procedural” knowledge. Their writing skills arestill “developing” (see Mohan and Lo, 1985). Raimes (1987) also commentsthat second-language writers, in addition to lacking linguistic proficiency inL2, might also lack writing ability in Li, lack knowledge of the conventionsof L2 written products, and lack practice in generating and organizing ideasin L2 for an L2 reader.Some ESL researchers may think that second-language writers’writings are influenced by cultural thought patterns, as Kaplan’s “contrastiverhetoric” (1966, 1967, 1978, 1979, 1987; Hinds, 1980, 1983) may suggest.According to Kaplan (1966), “Oriental” (he means Korean and Chinese)students’ way of writing/thinking is circular and does not approach a98conclusion directly. He says that the cultural thought patterns of non-Englishstudents may affect their English (L2) writing. Similarly, Hinds (1983)argues that a Japanese rhetorical pattern (ki-sho-ten-ketsu: introduction-development-development with twist-conclusion) is radically different fromthe English rhetorical pattern, and that this may have a negative influence onJapanese students’ English (L2) composition.However, both Kaplan and Hinds compared published articles in Li(translated into English) and students’ L2 (English) compositions. Theirstudies are not based on the direct comparison of students’ actual writingsand writing processes in Li and L2. The present study, by comparing thestudents’ actual translation and (Li and L2) writing processes and texts,failed to confirm Kaplan’s and Hinds’ claims. It was hard to conclude that thestudents’ thought patterns were fixed in cultural thought patterns, as such.4.6. Results of Research Question 6: How do second language learnersrestructure L]/L2 semantic and syntactic correspondences while translating?4.6.1. SummaryIn order to answer research question #6, the participants’ written texts(translations) were analyzed descriptively, and compared with the originaltext and model translation. The participants’ think-aloud protocols were alsoanalyzed descriptively. Contrary to the findings from the observationsreported in the previous section (4.5), where the participants’ translatingprocesses were described rather negatively as “one-way acts” or “writingwithout composing”, the text analyses of the participants’ translations99revealed that most of their lexical choices were not simple word-for-wordmappings across Li and L2, and instances of restructuring were frequentlydiscovered. In the L2 writing task, however, because most students wrotedirectly in English using vocabulary and sentence structures which theyknew well, and there was no “model” or “original” text, examples ofrestructuring were not able to collect.4.6.2. AnalysesExamples of participants’ restructuring in the translation task areanalyzed here as follows. The sentences in italics are from the modeltranslation in the Nihongo Journal. (The whole translation is in Appendix F,and the Japanese text is in Appendix E.)(1) “April is the month for joining companies. It’s a time when many newemployees come into being.”In this model translation, the underlined expression “come intobeing” is a translation of a Japanese expression, “umareru” (lit. “are born”).Most students in the study avoided literal translation and rephrased theJapanese expression so that their translated English expression became“appear”, “start to work”, “go to companies”, “we can see ....“, and so on.Some examples of student translations:a) April is the month when job start. At that time there are a lot of newemployees. (Shiro, 4N)b) April is the month to starting jobs. It’s also the time when many peoplebecome freshmen. (Takako, 4T)100(2) “Some companies request a lecture from a famous person. Others teachthe proper use ofpolite language. Sometimes, employees will go to acamp; this helps them adjust to working in a group.”In this case, the sentence subject in each sentence in the Japanese textis omitted because it is clear from the context. Therefore, when translatinginto English, it is necessary to supply the subject, as shown in the modelsentences (underlined), due to the English structural rule. Almost all thestudents in the study supplied the sentence subject by saying “companies”,“some companies”, “people”, “they”, and so on. However, the subject of thethird sentence is different from the subject of the first and the second.Unfortunately, many students failed to distinguish this point clearly in theEnglish translation. They used “they” carelessly and made the translationambiguous. Otherwise, their restructuring of the Japanese sentence bysupplying the sentence subject was successful.Examples of student translations:a) The side of training always ask a famous person for lecture. For example,new employee have to learn the way of speaking right words, to go campfor accompanying with other coworkers and they have to sit in religiousmeditation. From the religious mediation, they can obtain the habit of thecool and concentrated minds. (Setsuko, 4N)b) The company asks to a famous person to have a lecture. They teach themwording expression. They lodge together to get used to a group life. Also,they sit in religious contemplation at a temple. They make quiet, calmmind to do it. (Nami, 2N)(3) “Recently, a certain publishing company put together a textbook onemployee education. The textbook, which is targeted to a generation ofyoung people raised on comic books, (consists of comics from cover to101cover. In this book, a senior employee) tells a new worker -- in easy-to-understand terms -- the ins and outs of the adult world.In the present study, the underlined part was omitted inadvertentlyfrom the text to be translated. The result, however, turned out to beunexpectedly rewarding; The researcher was able to observe each studentstruggling to understand the whole paragraph and supplying words in orderto make the translated text logical. Many students guessed that the textbookwas a comic book by reading the last sentence in the paragraph (“The comicis truly considerate.”), and several students restructured the first sentence byinserting information (“comic”) although it was not in the sentence of theassigned (Japanese) text. For example, one student restructured this as“....published a comic textbook”, and another, “.... published a textbook,which is written in comic.”Other examples of student translations are as follows.a) Recently, a certain publishing company made a textbook for young peoplewho are in a period of comic to teach the knowledge of the worker easily.It was written by comic. (Miki, 2T)b) Recently, one publishing company published a easy cartoon textbookwhich shows some experiences of business man for young generation whogrow with cartoon. (Takuya, 4T)The participants’ think-aloud protocols also revealed that they oftenconsciously restructured Li (Japanese) phrases and syntax according to L2(English). Following is some of Madoka’s (2T) think-aloud protocols onrestructuring in the translation task. (All of her think-aloud protocols are inAppendix I.) She was speaking in Japanese, but for ease of reading, only theEnglish translation (made by the researcher) is presented here.102• “no, not ‘education’, ‘kyooyoo’ (education, culture), ‘shitsuke’(training, drill, discipline)”• “it’s’training’!”(Madoka avoided direct translation (kyooiku = education), listing upsynonyms in Japanese.)• “in Japanese, it says ‘kono hitotachi ga yoi sham ni naru yoonikaisha wa kyooiku o hajimeru’ (lit. in order that these peoplewould become good workers, companies start training), but Iwould restructure this sentence this way, ‘company starttraining for them to be a good staff’(Madoka restructured the Japanese sentence according to the Englishsentence structure.)• “and it lists various examples, so I inserted ‘for example’ “(She inserted “for example” in the English translation although it was notin the Japanese text.)• “there is no sentence subject, so....’a company’, ‘a company’, isit odd?”(Madoka supplied a sentence subject which is omitted in the Japanesesentence.)• “in this case, I will make the person or something the sentencesubject, I may use ‘it’ or....”(Here again, she supplied a sentence subject for the English translation.)• “I’m gonna connect these sentences with a comma...”(She connected two short sentences in the Japanese text for the Englishtranslation.)103“suiteru densha’ (lit. ‘a train that is empty’) means ‘man’in janai densha’ (‘a train that is not crowded’)”(Madoka rephrased the Japanese expression according to the Englishusage.)“to yuu iken de aru’ (‘these are their opinions’), this phrasebelongs to the above sentence,.... let’s insert it here”(Madoka moved the translated English phrase around so that the translatedtext would become logical.)4.6.3. DiscussionContrary to the findings from the observations reported in theprevious section (4.5), where the participants’ translating processes weredescribed rather negatively as “one-way acts” or “writing withoutcomposing”, the text analyses of the participants’ translations and “thinkaloud” protocols revealed that most of their lexical choices were not simpleword-for-word mappings across Li and L2, and instances of restructuringoccurred throughout.Repeatedly, it is asserted that direct or literal translation has negativeeffects on language learning (Dulay, Burt, & Krashen, 1982; Krashen,1981). However, as shown here, direct translations were often avoided andrestructuring was not limited to only the sentence level. Although finishedtranslations may oftentimes be unsuccessful due to the lack of Englishproficiency, the second language learners in this study took the time in theirtranslating tasks to, as Ijaz (1986, p.4-05) said, “restructure existing Liconcepts or develop a new concept that corresponds to a lexical item in theL2”.104Restructuring during translation and paying conscious attention tovocabulary and grammar across Li and L2 seems related to an increase inthe scores on language use for the translation task as reported in section 4.4.Translation often involves conscious cross-linguistic attention to vocabulary,grammar, syntax, and discourse; and this conscious attention may lead to thequalitative change in the learner’s L1/L2 semantic and syntacticcorrespondences.Raimes (1983) says that L2 writing is helpful for language learningdue to the reinforcement of the grammar and vocabulary which studentslearn in the classroom, but it seems that translation tasks could give morethan simple reinforcement of the grammar and vocabulary. Translation tasksprovide learners with good opportunities for restructuring between Li andL2 linguistic correspondences by paying conscious attention to L1IL2grammar and expressions.105Chapter 5CONCLUSIONThis concluding chapter presents, first, a summary of the researchfindings. The second section develops these findings into three generalizedpoints of relevance, leading to conclusions. The subsequent three sectionsdescribe the significance of this study in terms of second language acquisitiontheory, including implications for second language education and suggestionsfor future research in translation and L2 writing.5.1. Summary of FindingsThe research findings described in the previous chapter aresummarized as follows.1. Correlations between the quality of second language learners’ (the secondand fourth-year college students’) translation and that of their Li and L2writing were unclear. Other studies, such as Cumming (1988), Gaskill(1986), and Jones and Tetroe (1987), compared student writers andprofessional or semi-professional writers, and claim that second languagelearners’ Li writing skills are transferable to their L2 writing skills.However, in this study, only student writers participated. Due to theirsimilar levels of Li writing skills, correlations between the quality of106translation and that of their Li and L2 writing were not able to be clearlyconfirmed.2. The second language learners in this study generally thought that bothtranslation and L2 writing exercises are helpful for learning andimproving a second language. As well, they expressed the view thatspeaking practice is important. The fourth-year students in this study whowere trained in translation maintained, however, that translation tasks aremore helpful than L2 writing for learning and improving a secondlanguage. Their comments were consistent with Swain’s (1985) “pushed”output hypothesis for language learning.3. The second language learners’ self-reported conscious attention paid to“language use” in the translation task was very high, as we expected, butvery low in the L2 writing task. Similarly, their conscious attention paidto vocabulary and grammar while performing the task was very high inthe translation, but unexpectedly very low in the L2 writing.4. The two English evaluators found that the quality of the second languagelearners’ language use was significantly better in the translation than inthe L2 writing.5. The second language learners, both second and fourth-year collegestudents, showed the characteristics of unskilled writers’ translating andwriting processes described by many other research papers. The secondlanguage learners in this study had “declarative” or “theoretical”knowledge, but lacked “procedural” or “practical” knowledge. That is,they lacked practice in writing and translation.6. The text analyses revealed that the second language learners did nottranslate the text word for word, contrary to our expectation. They oftenavoided direct translations and restructured Li expressions and syntax107according to L2. In their translating processes, instances of restructuringnot just at the word level were observed frequently.5.2. GeneralizationsBased on the research findings stated in the previous section, threegeneralizations are presented here:First, the present research findings indicate that second languagelearners who have obtained basic spoken communicative competence in L2,yet whose Li writing skills are still developing, tend to write L2compositions using only words and expressions readily accessible to them.Both translation and L2 writing tasks may be perceived by learners to behelpful for learning/improving a second language, but translation tasks maybe more helpful for improving learners’ language use in L2. The learners inthis study frequently paid conscious attention to language use in thetranslation task but did not pay enough attention to language use in the L2writing task. The quality of the students’ language use in the translation taskwas significantly better than in the L2 writing task.Second, translation tasks can encourage or push second languagelearners to produce an ‘Ti+i output” by keeping the level of translation tasksslightly higher than the learners’ present level. The fourth-year students inthis study who were trained in translation expressed views which wereconsistent with Swain’s “i+i output” hypothesis.Third, translation tasks may provide good opportunities for“restructuring” L1/L2 correspondences for second language learners. Secondlanguage learners often restructure Li and L2 corresponding phrases and/108sentences while translating. The research findings indicate that secondlanguage learners seldom translate word for word; they often considersynonyms, and restructure Li phrases, sentences, and paragraphs accordingto the L2, although the translated text may not necessarily be successful dueto the students’ particular lack of linguistic resources.Thus, conclusions from this study can be stated as follows: Text-leveltranslation in L2 learning deserves a closer look as it provides potentialopportunities for learners to learn a second language 1) by paying consciousattention to language use, 2) by producing an “i+l output”, and 3) byrestructuring Li and L2 correspondences at word, phrase, sentence, andparagraph levels.5.3. Significance for Theories of Second Language Acquisition5.3.1. Skill TransferCummins’ (1979, 1984, 1986) theories of the cross-linguisticinterdependence of cognitive academic skills have been tested by manyresearchers in second language writing (Cumming, 1988, 1989; Cumming,Rebuffot, & Ledwell, 1989; Jones & Tetroe, 1987; Raimes, 1987).Comparing inexperienced or basic student writers and professional oradvanced writers, they found that writing skills in Li largely transfer to L2writing and that second language general proficiency (mainly speaking skills)is not directly related to L2 writing skills.The present study tested Cummins’ theories using second languagelearners’ translation performances, but was not able to confirm therelationships among translation, Li and L2 writing under his theories. The109learners’ Li writing skills showed only a weak correlation with translationand L2 writing. Probably this is because the learners’ Li writing skills arestill developing. Unless a person’s Li writing skills are professional or of ahigh level, this skill transfer may not occur. If “skill transfer” does notoccur unless the skill acquired is extremely developed or professional,second language learners whose Li writing skills are still developing mayneed training in both Li and L2 writing, as well as translation (see Edelsky,1982, 1986).5.3.2. Conscious LearningThe generalizations on translation in language learning stated in thisstudy may not be entirely consistent with some views of second languagepedagogy. As described in the literature review, translation has generallybeen ignored as the “Grammar-Translation Method” in second languagepedagogy since the late 1950s in Canada and the United States. Even today,many researchers and educators think that translation or use of Li isineffective in second language education.Generalizations from this present research, however, are in concertwith cognitive theories of second language acquisition which advocateconscious learning. Schmidt (1990) comments that “those who notice mostlearn most, and it may be that those who notice most are those who payattention most, as a general disposition or on particular occasions” (p. 144).Many researchers (Bialystok, 1981, 1982; Ellis, 1986, 1993; McLaughlin,1987; Wenden & Rubin, 1987) say that conscious learning is facilitative foradult learners learning/acquiring a second language.110Translation is often done in a very conscious way, but few researchershave studied translation in the framework of conscious learning. This study,in a sense, expands the concept of conscious learning by exploring secondlanguage learners’ translating processes in this context. Examining thelearners’ conscious attention patterns in the translation and L2 writing tasks,this study suggests that translation tasks may be more helpful than L2 writingtasks for language learning due to the learner’s frequent conscious attentionto language use while translating.Translation tasks may be particularly helpful for those whose Liwriting skills are still developing. The learners’ conscious metalinguisticattention in the L2 writing task was unexpectedly very low in this study. Theinexperienced student writers in this study consciously avoided the use ofdifficult expressions and syntax in the L2 writing task. As Cumming (1988)says, L2 writing may be “more beneficial for those with greater writingexpertise” (p. 190).5.3.3. The “i-i-i Output” HypothesisMost students in the study indicated that both translation and L2writing tasks are helpful for learning and improving a second language.However, the fourth-year students trained in translation answered thattranslation is more helpful than L2 writing because students are forced to usewords and expressions which are slightly beyond their levels when theytranslate.Their comments are consistent with Swain’s (1985) “pushed” outputhypothesis. Swain argues that the learner needs to have chances to producewords, expressions, and syntax which are a little higher than the learner’s111present level (“i+i level output”) in order to learn and improve the secondlanguage. Swain considered that L2 writing, compared to L2 speaking, is atask that forces students to produce an “i+ 1” output, after she had examinedFrench immersion students in Canada, who performed better in written thanin spoken French. In this case, she did not pay attention to translationexercises.However, translation exercises controlled at the “i+i” level may bemore effective than L2 writing exercises for second language acquisitionbecause learners have to use words, expressions, grammar, and syntax whichare a little beyond their present level in order to translate the Li text intoL2. Second language learners may write L2 compositions at the “i - 1” level,using only words and expressions readily accessible to them. They often tendto “avoid” expressions and syntax which they feel too difficult (see Schacter,1974). Or they may write at the “i+2” level, trying to achieve a higher levelthan expected, and end up writing an incomprehensible text or giving upcompletely (see Uzawa & Cumming, 1989).Thus, translation exercises controlled at the “i+l” level by theinstructor may be effective in encouraging the learner to produce a “pushed”output without necessarily pushing the learner too far.5.3.4. RestructuringRecent cognitive theories in second language learning suggest thatlearners learn an L2 by restructuring syntactic and semantic correspondencesbetween Li and L2, and not by merely adding new L2 items to their existingL2 systems (see Cheng, i985; Ellis, 1986; Ijaz, 1986; Karmiloff-Smith,1986; Lightbown, 1985; McLaughlin, 1987, 1990b).112The concept of “restructuring” in language learning, however, hasnever been studied in the process of translation. This study found that“restructuring” often occurred in the process of translation of secondlanguage learners; it revealed that the second language learners, even if theywere not good at translation, often avoided a word-for-word translation,contrary to general expectations. The text analyses of their translations andthink-aloud protocols showed that the learners frequently rephrased Liexpressions so that the L2 might sound natural and grammatical. Instances of“restructuring” were abundant in the learners’ translating processes, andwere not limited to the word level.Translation and any use of Li in language learning has been thought tohave a negative influence on L2. However, here there is reason to believethat translation tasks are in concert with recent cognitive theories oflanguage learning.5.4. Implications for Second Language Education5.4.1. College Instructors’ ViewsAfter analyzing the data, a short preliminary summary was submittedto the College faculty, requesting their comments, opinions, and/orquestions. It was necessary to ask instructors’ views on the research findingsin order to counterbalance the researcher’s views, which are based strictlyon the experimental setting. (see Guba & Lincoln, 1989.) The facultyarranged a one-hour discussion session, which was held on the CIC campus,and was attended by about 15 faculty members. The session was taperecorded with the permission of the attending instructors.113First of all, many instructors expressed their surprise at one of theresearch results, which showed no significant difference in the quality of thestudents’ translation, Li and L2 writings among the four groups. That is,the quality of the “4th-year trained” and “2nd-year not-trained” students’translations in this study was relatively similar. The researcher explainedthat the number of the participants was very small for statistical analyses ofratings of writing samples, and that, although not significant statistically, the4th-year trained students did best in translation and L2 writing tasks amongthe four groups.Regarding the students’ performance, two instructors in the translationand interpretation courses mentioned that the number of class hours spentfor actual translation exercises was very limited (one hour a week, over 33weeks a year). Thus, one instructor said that the term “trained” used in theresearch was not completely appropriate.Another instructor mentioned that most students in the College, whoare high-school graduates, do not get enough training in writing in theirmother tongue (Japanese) at the College. That is, the students’ chances towrite in Japanese are limited to personal letter writing. She suggested thattraining in Li writing seems to be necessary to teaching text-leveltranslation.One instructor expressed her doubts about translation, saying thatusing a first language in the language classroom is inefficient becausestudents are learning a second language, not a first. This type of view againsttranslation in second language learning, actually, seemed to be still verystrong among language instructors in general.Generally speaking, however, the instructors, especially those whoteach the translation and interpretation courses at the College, expressed114similar views toward translation as the researcher’s. For example, anexperienced instructor of translation at the College commented as follows:“. . . once they [students] have done a few exercises in translation, theyquickly realize they can’t translate literally word for word.Oftentimes, students are forced to use more complicated sentencestructures, vocabulary they are not used to; they seem to draw moreupon a kind of “passive” vocabulary, rather than just rely on theirmore limited “active” vocabulary, to express ideas and concepts thatare there in the Japanese; they [students] probably would not try toexpress [these] in English, without translation requiring them to do soIn a sense, they have to express themselves in more complicatedways...• . . It stretches them, and at times stretches them too much and youget something that doesn’t make sense, but then, if you give them thestructure and vocabulary they need, or at least direct them to wherethey can find that, then it’s another step in the writing process .whereas in L2 writing, they [students] might be content to usesentences they are familiar with . . . and therefore not really takechances.”In sum, the instructor indicates that students tend to use L2 structuresand vocabulary which they normally don’t use more often in translationexercises than in L2 writing, and that this type of taking chances is good forlanguage learning. This tendency observed in the classroom was observed inthe present research.Although the students’ performance in general was not good enough tosatisfy the instructors, most instructors were favorable towards views thattranslation could be an important component in second language learning,and a worthwhile academic subject.1155.4.2. Text-Level ApproachTranslation (especially Li into L2) exercises in second/foreignlanguage classrooms used to be predominantly sentence-level ones, and onlygrammatical mistakes were checked by the instructor. Text-level exerciseswere rare.One instructor at CIC commented that most Japanese students educatedin Japan are generally familiar with text-level translation because theyusually translate English textbooks into Japanese in most English classes.However, text-level translation exercises from Japanese into English are veryrare in high school English classrooms in Japan. Most participants in the 4Nand 2N groups said that this study was their first experience of translating awhole Japanese text into English. The present researcher’s experience oflearning English as a foreign language at junior and senior high-schools inJapan was similarly dominated by sentence-level translation (Li into L2)exercises.The concept of translation in the language classroom needs to berethought. Text-level translation exercises should be incorporated into thecurriculum. A collection of single sentences does not form a text (see Brown& Yule, 1983); the text has a structure, organization, or wholeness as a text.The choice of vocabulary and syntax depends on the context, i.e., text.Similarly, for teaching grammar, the sentence-level approach has problemsin semantic and/or pragmatic areas (Celce-Murcia, 1990). Celce-Murcia(1990) argues that tense-aspect-modality, word-order, subordination, andcomplementation in English are often determined at the discourse level, notat the sentence level.116Translation exercises, as Weymouth (1984) says, should teach thestudent text structure and appropriateness of style in the sociolinguisticcontext in L2. Text structures such as description, contrast, and argumentcan be taught explicitly in text-level translation exercises using Mohan’s(1986) concept of “knowledge structures”, which examines academicdiscourse at practical and theoretical levels (see also Carrel!, 1983, 1985;Meyer, 1977, 1985). For example, the text structures of “description” andlanguage register in L2 can be discussed in lectures first (giving theoreticalknowledge). Then through the actual practice of translating short stories,comic strips, letters (personal and formal), and so forth, the learner canlearn how structures of “description” and language register in L2 are interrelated, and how they are similar or different across Li and L2.In terms of text organization, the present study showed that theparticipants lacked “procedural” or “practical” knowledge of translation andL1/L2 writing. They often mentioned terms such as “introduction-body-conclusion”, “thesis statement”, and “topic sentence” while writing, but werenot able to “transform” their knowledge in their actual writing. They lackedactual practice in translation and writing. In general, teaching “theoreticalknowledge” (e.g., grammar rules, writing rules, translation rules) seems tobe more emphasized than “practical knowledge” (actual writing andtranslation) in many classrooms. In fact, some instructors at the College saidthat their students normally do not have enough time for actual translation(Li into L2) practice (only one hour a week over 33 weeks in one academicyear) because students have to study theories of translation as well (four tofive hours a week). Keeping a balance between the “declarative” and the“procedural’, or the “theoretical” and the “practical”, in classroom teachingis therefore important.117In the present study of second language acquisition, translation tasksseemed more effective than L2 writing tasks for second language learnerswith developing Li writing skills, but L2 writing exercises are also verynecessary. Translation exercises may not necessarily encourage students togenerate and organize ideas independently. Students need text-level practiceboth in translation and in L2 writing.5.4.3. Peer InteractionResearchers and educators in writing in the 1980s have suggested“peer” interaction as a way of improving students’ writing skills (e.g.,Crowhurst, 1981; DiPardo & Freedman, 1988; Hillocks, 1986). Researchersin translation have also mentioned the importance of “peer” interaction(Duff, 1989; Weymouth, 1984). Weymouth (1984) says thatgroup/cooperative/peer discussions on texts involving non-literary, up-to-date topics in the translation classroom are facilitative. Paying attention tothe reader and revising accordingly was very rare among the participants inthis study. Peer discussions and peer revising in the translation classroommight encourage the students to pay attention to the reader and to acquireITpractical knowledge” and real experience in translation.5.4.4. Separate SkillsMany students in the study indicated that speaking practice is alsonecessary for learning and improving the second language. Speaking andwriting are separate skills (Cumming, 1988; Fodor, 1983). It is doubtful that) the learner can acquire speaking skills only through translation and L2118writing. Similarly, as the present study suggests, it may be that Li writingskills do not transfer to L2 writing, or L2 writing skills do not transfer totranslation unless the learner’s skill levels are extremely high. It wouldappear that each skill needs to be learned and practiced. In the classroom,listening, speaking, reading, writing, translating, and interpreting activitiesshould all be planned (see Duff, 1989).To this end, team teaching (with two instructors, at least one bilingualand biliteral) may be ideal for L2 language teaching in the next century, inorder to satisfy this kind of dynamic model of language learning.5.5. Suggestions for Future Research in Translation and L2WritingReviewing the present research, the following five points may besuggested for future research in translation and L2 writing.1. In this study, Cummins’ theories were not clearly confirmed. IfCummins’ hypotheses are to be examined, comparisons betweenprofessional and student writers/translators, as in Cumming’s (1988)study, would be ideal. However, it is very hard to find a number ofprofessional translators and/or writers with similar backgrounds.2. In the present research, only Japanese (Li) and English (L2) werecompared. Thus, findings are limited. Other language combinations, suchas Korean and Japanese, English and Chinese, French and Spanish, and soon might produce some other interesting results.1193. The number of participants should be large enough for statistical analysesif the participants’ translation or writing ratings are to be compared. Thenumber of participants in this study (22 students) was large enough onlyfor the statistical analyses of their think-aloud utterances, not for theratings of writing samples.4. In the present study, only one translation task was assigned. The level ofcognitive demand was not low, but it may not have differentiated the skilllevels clearly. More than two kinds of translation (cognitively demandingand less cognitively demanding translation tasks, e.g., argument vs.personal letter writing) may be useful in order to check how tasks affectthe participant’s performance depending on his/her skill levels.5. The use of the “think aloud” technique for data collection in order toexamine the learner’s translating and writing processes was appropriate inthis study. However, there were some participants who tended to forget toverbalize their thoughts, or whose utterances were extremely limited, evenif they practiced thinking aloud before performing three tasks. Thus, bothinterviews right after the performance and text analyses were very usefulin order to compensate for the weaknesses of the “think aloud” technique.Triangulation of data collection methods (combining at least three types ofdata collection as in this study) is necessary./120REFERENCESAnderson, J. R. (1983). The architecture ofcognition. MA: Harvard University Press.Andrade, M. R. (1990). Review of Connor & Kaplan, Writing across languages: Analysis ofL2 text. TESOL Quarterly, 24(4), 724-727.Applebee, A. (1982). Writing and learning in school settings. In M. Nystrand (Ed.), Whatwriters know: The language, process, and structure ofwritten discourse (pp365-38 1). NY: Academic Press.Applebee, A. (1984a). Writing and reasoning. Review ofEducational Research, 54(4), 577-596.Applebee, A. (Ed.) (1984b). Contextsfor learning to write. NJ: Ablex.Asher, J. (1981). Fear of foreign languages. Psychology Today, August, 52-59.Auerbach, E. (1993). Reexamining English only in the ESL classroom. TESOL Quarterly,27 (1), 9-32.Becker, H. (1986). Writingfor social scientists: How to start andfinish your thesis, book,or article. Chicago: University of Chicago Press.Bell, R. T. (1991). Translation and translating. New York: Longman.Bereiter, C. & Scardamalia, M. (1987). The psychology of written composition. NJ:Lawrence Eribaum.Bialystok, E. (1981). The role of conscious strategies in second language proficiency.Modern Language Journal, 65, 24-35.Bialystok, E. (1982). On the relationship between knowing and using linguistic forms.Applied Linguistics, 3, 181-206.Biggs, J. & Collis, K. (1982). Evaluating the quality of learning: The SOLO Taxonomy(Structure of the Observed Learning Outcome). NY: Academic Press.Braddock, R., Lloyd-Jones, R., & Schoer, L. (1963). Research in written composition.Urbana, IL: National Council of Teachers of English.Britton, J. et al. (1975). The development ofwriting abilities (11-18). London: Macmillan.Brown, G. & Yule, G. (1983). Discourse analysis. Cambridge: Cambridge UniversityPress.Bruner, J. (1971). The relevance ofeducation. NY: Norton.121Burtis, P. J. & Bereiter, C., Scardamalia, M., & Tetroe, J. (1983). The development ofplanning in writing. In B. Kroll & G. Wells (Eds.), Explorations in the developmentofwriting (pp.l53-l’74.). Toronto: John Wiley.Butler-Nalin, K. (1984). Revising patterns in students’ writing. In A. Applebee (Ed.),Contextfor learning to write (PP. 121-133). NJ: Ablex.Butzkamm, W. (1985). The use of formal translation equivalents in the teaching of foreignlanguage structures. In C. Titford & A. E. Hieke (Eds.), Translation in foreignlanguage teaching and testing (pp. 87-97). Germany: Gunter Narr Verlag Tubingen.Carey, S. (1991). The culture of literacy in majority and minority language schools.Canadian Modern Language Review, 47(5), 950-976.Carrell, P. (1983). Three components of background knowledge in reading comprehension.Language Learning, 33(2), 183-207.Carrell, P. (1985). Facilitating ESL reading by teaching text structure. TESOL Quarterly, 19(4), 727-752.Carroll, J. B. (1963). Linguistic relativity, contrastive linguistics, and language learning.International Review ofApplied Linguistics in Language Teaching, 1 (1), 1-20.Catford, J. C. (1965). A linguistic theory of translation. London: Oxford University Press.Celce-Murcia, M. (1990). Discourse analysis and grammar instruction. Annual Review ofApplied Linguistics, 11, 135-151.Charney, D. (1984). The validity of using holistic scoring to evaluate writing: A criticaloverview. Research in the Teaching ofEnglish, 18(1), 65-81.Chaudron, C. (1988). Second language classrooms: Research on teaching and learning.Cambridge: Cambridge University Press.Cheng, P. W. (1985). Restructuring versus automaticity: Alternative account of skillacquisition. Psychological Review, 92 (3), 414-423.Chomsky, N. (1964). Aspects of the theory of syntax. Cambridge, MA: MIT Press.Chomsky, N. (1972). Language and mind. NY: Harcourt Brace Jovanovich.Connor, U. & Kaplan, R. (Eds.) (1987). Writing across languages: Analysis ofL2 text.Reading, MA: Addison-Wesley.Crowhurst, M. (1981). Managing marking: The writing workshop. In 1. Pringle & A.Freedman (Eds.), Teaching, writing, learning (pp. 51-58). Ottawa: Canadian Councilof Teachers of English.Cumming, A. (1988). Writing expertise and second language proficiency in ESL writingperformance. Unpublished doctoral dissertation. University of Toronto.Cumming, A. (1989). Writing expertise and second language proficiency. LanguageLearning, 39(1), 81-141.122Cumming, A. (1990). Metalinguistic and ideational thinking in second language composing.Written Communication, 7(4), 482-511.Cumming, A., Rebuffot, J., & Ledwell, M. (1989). Reading and summarizing challengingtexts in first and second languages. Reading and Writing: An Interdisciplinary Journal,2, 201-219.Cummins, J. (1979). Cognitive/academic language proficiency, linguistic interdependence,the optimum age question and some other matters. Working Papers on Bilingualism,19, 197-205.Cummins, J. (1984). Bilingualism and special education: Issues in assessment andpedagogy. Clevedon Avon: Multilingual Matters.Cummins, J. (1986). Language proficiency and academic achievement. In J. Cummins & M.Swain (Eds.), Bilingualism in education (pp. 138-16 1). London: Longman.Diederich, P. B. (1966). How to measure growth in writing ability. English Journal, 55,435-449.DiPardo, A. & Freedman, S. W. (1988). Peer response groups in the writing classroom:Theoretic foundations and new directions. Review ofEducational Research, 58(2),119-149.Duff, A. (1989). Translation. Oxford: Oxford University Press.Dulay, H., Burt, M. & Krashen, S. (1982). Language two. NY: Oxford University Press.Edeisky, C. (1982). Writing in a bilingual program: The relation of Li and L2 texts. TESOLQuarterly, 16(2), 211-228.Edeisky, C. (1986). Writing in a bilingual program. NJ: Ablex.Ellis, R. (1986). Understanding second language acquisition. Oxford: Oxford UniversityPress.Ellis, R. (1988). Classroom second language development. NY: Prentice Hall.Ellis, R. (1993). The structural syllabus and second language acquisition. TESOL Quarterly,27(1), 91-113.Ellis, R. (1994). The Study ofSecond Language Acquisition. Oxford: Oxford UniversityPress.Emig, J. (1971). The composing processes of twelfth graders. Urbana, IL: NCTE.Emig, J. (1977). Writing as a mode of learning. College Composition and Communication,28(2), 122-128.Enns-Connolly, E. E. (1985). Translation as interpretive act: A narrative study oftranslationin university-levelforeign language teaching. Unpublished doctoral dissertation.University of Toronto.123Ericsson, K. A. & Simon, H. A. (1980). Verbal reports as data. Psychological Review, 87(3), 215-251.Ericsson, K. A. & Simon, H. A. (1984). Protocol analysis: Verbal reports as data.Bradford, MA: MIT Press.Faerch, C. & Kasper, G. (Eds.) (1983). Strategies in interlanguage communication. London:Longman.Faerch, C. & Kasper, G. (1984). Two ways of defining communication strategies.Language Learning, 34(1), 45-63.Faigley, L. & Witte, S. (1981). Analyzing revision. College Composition andCommunication, 32 (4), 400-414.Faigley, L., et a!. (1985). Assessing writer’s knowledge and process of composing. NJ:Ablex.Rower, L. & Hayes, J. (1980). The cognition of discovery: Defining a rhetorical problem.College Composition and Communication, 31, 21-32.Flower, L. & Hayes, J. (1981a). A cognitive process theory of writing. CollegeComposition and Communication, 32 (4), 365-387.Flower, L. & Hayes, J. (1981b). The pregnant pause: An inquiry into the nature of planning.Research in the Teaching ofEnglish, 15(3), 229-243.Rower, L. et a!.. (1986). Detection, diagnosis, and the strategies of revision. CollegeComposition and Communication, 37, 16-55.Fodor, J. A. (1983). Modularity of mind. Cambridge, MA: MIT Press.Frawley, W. (1984). Prolegomenon to a theory of translation. In W. Frawley (Ed.),Translation: Literary, linguistic, andphilosophical perspectives (pp. 159-175).Newark: University of Delaware Press.Friedlander, A. (1990). Composing in English: Effects of a first language on writing inEnglish as a second language. In B. KroIl (Ed.), Second language writing: Researchinsights for the classroom (pp. 109-125). Cambridge: Cambridge University Press.Friedlander, A. & Huckin, T. N. (1987, April). Composing in ESL: Language acquisitionand the retrieval of topic knowledge from memory. Paper presented at the AmericanEducational Research Conference. Washington, D. C.Gaskill, W. H. (1986). Revising in Spanish and English as a second language: A process-oriented study ofcomposition. Unpublished doctoral dissertation, University ofCalifornia, Los Angeles.Gerloff, P. (1987). Identifying the unit of analysis in translation: Some uses of think-aloudprotocol data. In C. Faerch & G. Kasper (Eds.), Introspection in second languageresearch (pp.135-158). Philadelphia: Multilingual Matters.Goodlad, J. (1983). Aplace called school. NY: McGraw-Hill.124Guba, E. & Lincoln, Y. S. (1989). Fourth generation evaluation. Newbury Park, CA: Sage.Haastrup, K. & Phillipson, R. (1983). Achievement strategies in learner/native speakerinteraction. In C. Faerch & G. Kasper (Eds.), Strategies in interlanguagecommunication (pp. 140-158). London: Longman.Hall, E. (1991). Variations in composing behaviors of academic ESL writers in test and non-test situations. TESL Canada Journal, 8(2), 9-33.Harris, B. (1977). The importance of natural translation. Working Papers on Bilingualism,12, 96-114.Harris, B. (1978). The difference between natural and professional translation. CanadianModern Language Review, 34,417-427.Harris, W. H. (1977). Teacher response to student writing: A study of the response patternsof high school English teachers to determine the basis for teacher judgement of studentwriting. Research in the Teaching ofEnglish, 11, 175-185.Hayes, J. & Rower, L. (1980). Identifying the organization of writing processes. In L. W.Gregg & E. Steinberg (Eds.), Cognitive processes in writing (pp. 3-30). NJ:Lawrence Eribaum.Hayes, J. & Flower, L. (1983). Uncovering cognitive processes in writing: An introductionto protocol analysis. In P. Mosenthal, L. Tamor, & S. Walmsley (Eds.), Research onwriting: Principles and methods (pp. 207-220). NY: Longman.Hieke, A. E. (1985). Transliteration as a modest alternative. In C. Titford & A. E. Hieke(Eds.), Translation in foreign language teaching and testing (pp. 99-115). Germany:Gunter Narr Verlag Tubingen.Hillocks, G. (1981). The interaction of instruction, teacher comment, and revision inteaching the composing process. Research in the Teaching ofEnglish, 16(3), 261-278.Hillocks, G. (1986). Research on written composition: New directions for teaching. Urbana,IL: National Conference on Research in English.Hinds, J. (1980). Japanese expository prose. Papers in Linguistics: International Journal ofHuman Communication, 13 (1), 117-158.Hinds, J. (1983). Contrastive rhetoric: Japanese and English. Text, 3 (2), 183-195.Hölscher, A. & Möhle, D. (1987). Cognitive plans in translation. In C. Faerch & G. Kasper(Eds.), Introspection in second language research (pp.1 13-134). Philadelphia:Multilingual Matters.Ijaz, I. H. (1986). Linguistic and cognitive determinants of lexical acquisition in a secondlanguage. Language Learning, 36(4), 401-451.Jacobs, H. L. et al. (1981). Testing ESL composition: A practical approach. Rowley, MA:Newbury House.125Jacobs, S. (1982). Composing and coherence: The writing ofeleven pre-medical students.Washington, D. C. : Center for Applied Linguistics.Jenkins, S. & Hinds, J. (1987). Business letter writing: English, French, and Japanese.TESOL Quarterly, 21(2), 327-349.Jones, S. & Tetroe, J. (1987). Composing in a second language. In A. Matsuhashi (Ed.),Writing in real time (pp.34-57). NJ: Ablex.Kaplan, R. (1966). Cultural thought patterns in inter-cultural education. Language Learning,16(1&2), 1-20.Kaplan, R. (1967). Contrastive rhetoric and the teaching of composition. TESOL Quarterly,1 (4), 10-16.Kaplan, R. (1978). Contrastive rhetoric: Some hypotheses. ITL, 39/40, 61-72.Kaplan, R. (1979). Contrastive rhetorics: Some implications for the writing process. In A.Freedman, I. Pringle, & J. Yalden (Eds.), Learning to write: First language/Secondlanguage. London: Longman.Kaplan, R. (1987). Cultural thought patterns revisited. In U. Connor & R. Kaplan (Eds.),Writing across languages: Analysis ofL2 text (pp. 9-20). Reading, MA: Addison-Wesley.Karmiloff-Smith, A. (1986). Stage/structure versus phase/process in modelling linguistic andcognitive development. In I. Levin (Ed.), Stage and structure: Re-opening the debate(pp. 164-190). Norwood, NJ: Ablex.Kikuchi, Y. (1987). Sakubun no hyoka hoho ni tsuiteno ichi shian [Procedures forevaluating writing: A proposal]. Nihongo Kyoiku/Journal ofJapanese LanguageTeaching, 63, 87-104.Kintsch, W. & van Dijk, T. A. (1978). Toward a model of text comprehension andproduction. Psychological Review, 85 (5), 363-394.Kobayashi, H. & Rinnert, C. (1990, March). Effects offirst language on second languagewriting: Translation vs. direct composition. Revised version of the paper presented atthe 24th Annual TESOL Convention, San Francisco.Krashen, S. (1981). Second language acquisition and second language learning. Oxford:Pergamon.Krashen, S. (1984). Writing: Research, theory, and applications. Oxford: Pergamon.Krings, H. P. (1987). The use of introspective data in translation. In C. Faerch and G.Kasper (Eds.), Introspection in second language research (pp.159-176). Philadelphia:Multilingual Matters.Kuno, S. (1978). Danwa no bumpo. [Discourse grammar]. Tokyo: Taishukan.Kupsch-Losereit, S. (1985). The problem of translation error evaluation. In C. Titford & A.E. Hieke (Ecis.), Translation in foreign language teaching and testing (pp. 169-179).Germany: Gunter Narr Verlag Tubingen.126Langer, J. & Applebee, A. (1984). Language, learning, and interaction: A framework forimproving the teaching of writing. In A. Applebee (Ed.), Contextsfor learning towrite (pp. 169-181). NJ: Ablex.Langer, J. & Applebee, A. (1987). How writing shapes thinking: A study ofteaching andlearning. Urbana, IL: NCTE.Lay, N. D. S. (1975). Chinese language interference in written English. Basic Writers, 1(1), 50-61.Leki, 1. (1991). Twenty-five years of contrastive rhetoric: Text analysis and writingpedagogies. TESOL Quarterly, 25(1), 123-143.Lightbown, P. (1985). Great expectations: Second-language acquisition research andclassroom teaching. Applied Linguistics, 6(2), 173-189.Matsuhashi, A. (1981). Pausing and planning: The tempo of written discourse production.Research in the Teaching ofEnglish, 15(2), 113-134.Matsuhashi, A. & Gordon, E. (1985). Revision, addition, and the power of the unseen text.In S. W. Freedman (Ed.), The acquisition ofwritten language: Response and revision(pp. 226-249). NJ: Ablex.McLaughlin, B. (1987). Theories ofsecond-language learning. London: Edward Arnold.McLaughlin, B. (1990a). “Conscious” versus “unconscious” learning. TESOL Quarterly, 24(4), 617-634.McLaughlin, B. (1990b). Restructuring. Applied Linguistics, 11 (2), 113-128.Mendelsohn, D. & Cumming, A. (1987). Professors’ ratings of language use and rhetoricalorganizations in ESL compositions. TESL Canada Journal, 5(1), 9-26.Meyer, B. (1977). What is remembered from prose: A function of passage structure. In R.0. Freedle (Ed.), Discourse production and comprehension, v. 1 (pp. 307-336).Norwood, NJ: Ablex.Meyer, B. (1985). Prose analysis: Purposes, procedures, and problems. In B. K. Britton &J. B. Black (Eds.), Understanding expository text (pp. 269-304). Hilisdale, NJ:Lawrence Eribaum.Moffett, J. (1968). Teaching the universe ofdiscourse. NY: Houghton Mifflin.Mohan, B. (1986). Language and content. Reading MA: Addison-Wesley.Mohan, B. & Lo, W. (1985). Academic writing and Chinese students: Transfer anddevelopmental factors. TESOL Quarterly, 19(3), 515-534.Newmark, P. (1981). Approaches to translation. Oxford: Pergamon.Nida, E. (1976). A framework for the analysis and evaluation of theories of translation. InR. W. Brislin (Ed.), Translation: Applications and research (pp. 47-91). NY: GardnerPress.127Odell, L. & Cooper, C. R. (1980). Procedures for evaluating writing: Assumptions andneeded research. College English, 42 (1), 35-43.Odell, L., Goswami, D., & Herrington, A. (1983). The discourse-based interview: Aprocedure for exploring the tacit knowledge of writers in nonacademic settings. In P.Mosenthal, L. Tamol, & S. Walmsley (Eds.), Research on writing: Principles andmethods (pp. 221-236). NY: Longman.Odlin, T. (1989). Language transfer. Cambridge: Cambridge University Press.Olson, D. (1977). From utterance to text: The bias of language in speech and writing.Harvard Educational Review, 47(3), 257-281.Ono, M. (1990). From sentence to discourse: Integrated explanationsfor certain linguisticphenomena in Japanese. Unpublished doctoral dissertation, University of BritishColumbia.Pen, S. (1979). The composing processes of unskilled college writers. Research in theTeaching ofEnglish, 13 (4), 317-336.Piaget, J. (1967). Six psychological studies (A. Tenzer, Trans.) New York: Random House.(Original work published 1964).Pianko, S. (1979). A description of the composing processes of college freshman writers.Research in the Teaching ofEnglish, 13(1), 5-22.Polanyi, M. (1958). Personal knowledge: Toward a post-critical philosophy. Chicago:University of Chicago Press.Raimes, A. (1983). Techniques in teaching writing. NY: Oxford University Press.Raimes, A. (1985). What unskilled ESL students do as they write: A classroom study ofcomposing. TESOL Quarterly, 19(2), 229-258.Raimes, A. (1987). Language proficiency, writing ability, and composing strategies: A studyof ESL college student writers. Language Learning, 37(3), 439-468.Ringbom, H. (1987). The role of the first language in foreign language learning.Philadelphia: Multilingual Matters.Rose, M. (1980). Rigid rules, inflexible plans, and the stifling of language: A cognitivistanalysis of writer’s block. College Composition and Communication, 31 (4), 389-401.Ryle, G. (1949). The concept of mind. London: Hutchinson.Sa’Adeddin, M. A. (1989). Text development and Arabic-English negative interference.Applied Linguistics, 10 (1), 36-51.Samuel, M. (1990). Language of testing effects. Unpublished master’s thesis. University ofAlberta, Edmonton.128Santos, T. (1988). Professors’ reactions to the academic writing of nonnative-speakingstudents. TESOL Quarterly, 22(1), 69-90.Schachter, J. (1974). An error in error analysis. Language Learning, 24(2), 205-214.Schmidt, R. (1983). Interaction, acculturation, and the acquisition of communicativecompetence: A case study of an adult. In N. Wolfson & E. Judd (Eds.),Sociolinguistics and language acquisition (pp. 137-174). Rowley, MA: NewburyHouse.Schmidt, R. (1990). The role of consciousness in second language learning. AppliedLinguistics, 11, 129-158.Schmidt, R. (1993). Awareness and second language acquisition. Annual Review ofAppliedLinguistics, 13, 206-226.Schmidt, R. (1994). Deconstructing consciousness in search of useful definitions for appliedlinguistics. AILA Review, 11, 11-26.Seleskovitch, D. (1976). Interpretation: A psychological approach to translation. In R. W.Brislin (Ed.), Translation: Applications and research (pp. 92-116). NY: GardnerPress.Seizer, J. (1983). The composing processes of an engineer. College Composition andCommunication, 34 (2), 178-187.Shaughnessy, M. (1977). Errors and expectations. NY: Oxford University Press.Silva, T. (1992). Li vs L2 writing: ESL graduate students’ perceptions. TESL CanadaJournal, 10(1), 27-47.Snell-Hornby, M. (1985). Translation as a means of integrating language teaching andlinguistics. In C. Titford and A. E. Hieke (EcIs.), Translation inforeign languageteaching and testing (pp. 21-28). Germany: Gunter Narr Verlag Ttibingen.Sommers, N. (1980). Revision strategies of student writers and experienced adult writers.College Composition and Communication, 31 (4), 378-388.Sorhus, H. (1975). Translation and second language learning: Should the mother tonguereally be discouraged? Ottawa: Research Division, Directorate of Studies, StaffDevelopment Branch, Public Service Commission.Stern, H. H. (1983). Fundamental concepts of language teaching. Oxford: OxfordUniversity Press.Stern, H. H. (1985). The time factor and compact course development. TESL CanadaJournal, 3 (1), 13-27.Swain, M. (1985). Communicative competence: Some roles of comprehensible input andcomprehensible output in its development. In S. Gass & C. Madden (Eds.), Input insecond language acquisition (pp. 235-253). Cambridge: Newbury House.Swain, M. (1988). Manipulating and complementing content teaching to maximize secondlanguage learning. TESL Canada Journal, 6(1), 68-83.129Tarone, E., Cohen, A. D., & Dumas, G. (1983). A closer look at some interlanguageterminology: A framework for communication strategies. In C. Faerch & G. Kasper(Eds.) Strategies in interlanguage communication (pp. 4-14). London: Longman.Tarvin, W. & Al-Arishi, A. Y. (1991). Rethinking communicative language teaching:Reflection and the EFL classroom. TESOL Quarterly, 25(1), 9-27.Taylor, B. (1981). Content and written form: A two-way street. TESOL Quarterly, 15(1), 5-13.Terrell, T. D. (1977). A natural approach to second language acquisition and learning.Modern Language Journal, 61,325-337.Titford, C. (1985). Translation - a post-communicative activity for advanced learners. In C.Titford & A. E. Hieke (Eds.), Translation inforeign language teaching and testing(pp. 73-85). Germany: Gunter Narr Verlag Tubingen.Titford, C. & Hieke, A. E. (Eds.) (1985). Translation in foreign language teaching andtesting. Germany: Gunter Narr Verlag Tubingen.Troyka, L. & Nudelman, J. (1975). Taking action: Writing, reading, speaking, and listeningthrough simulation-games. NJ: Prentice-Hall.Tudor, I. (1987a). Using translation in ESP. ELT Journal, 41 (4), 268-273.Tudor, I. (1987b). A framework for the translational analysis of texts. The Linguist, 26 (2),80-82.Uzawa, K. & Cumming, A. (1989). Writing strategies in Japanese as a foreign language:Lowering or keeping up the standards. Canadian Modern Language Review, 46(1),178-194.van Dijk, T. A. (1972). Some aspects of text grammars. The Hague: Mouton.Vann, R. J., Meyer, D. E., & Lorenz, F. 0. (1984). Error gravity: A study of facultyopinion of ESL errors. TESOL Quarterly, 18(3), 427-440.Váradi, T. (1983). Strategies of target language learner communication: Message adjustment.In C. Faerch & G. Kasper (Eds.) Strategies in interlanguage communication (pp. 79-99). London: Longman.Vygotsky, L. (1986). Thought and language (A. Kazulin, Trans.). Cambridge, MA: MITPress. (Original work published 1934).Weinberg, J. (1990, June). Pennies from He Vinh. TESOL Newsletter, P. 5.Wenden, A. & Rubin, J. (Eds.) (1987). Learner strategies in language learning. NJ:Prentice-Hall.Weymouth, A. G. (1984). A learner-centred approach to translation at the post “A” levelstage. The Linguist, 23 (3), 167-170.White, E. (1985). Teaching and assessing writing. San Francisco: Jossey-Bass.130Whorf, B. L. (1956). Language, thought, and reality. Selected writings edited and with anintroduction by John B. Carroll. Cambridge, MA: MIT Press.Wilss, W. (1976). Perspectives and limitations of a didactic framework for the teaching oftranslation. In R. Brislin (Ed.), Translation: Applications and research (pp. 117-137).New York: Gardner Press.Witte, S. P. (1985). Revising, composing theory, and research design. In S. W. Freedman(Ed.), The acquisition ofwritten language: Response and revision (pp. 250-284). NJ:Ablex.Yalden, J. (1975). Language teaching and language learning in the universities. CanadianModern Language Review, 31 (4), 334-339.Zamel, V. (1976). Teaching composition in the ESL classroom: What we can learn fromresearch in the teaching of English. TESOL Quarterly, 10(1), 67-76.Zamel, V. (1982). Writing: The process of discovering meaning. TESOL Quarterly, 16(2),195-209.Zamel, V. (1983). The composing processes of advanced ESL students: Six case studies.TESOL Quarterly, 17(2), 165-187.131APPENDIX A: Memo to the InstructorsTo: E., J., K., T, and C.From: Kozue Uzawa (685-6957)Re: Research project on writing and translating processes of second language learnersDate: September 14, 1992Your name has been given to me by M. L. because you are teaching students whomight be able to become volunteers for a research project for my Ph.D. dissertation (UBC). Iam looking for 24 volunteers (12 fourth- and third-year and 12 second-year students who aretaking/not taking translation courses). Would you please spare 10 minutes of your class timefor me so that I could talk about the project to the students in your class? I will gathervolunteers’ names and phone numbers. Time and place for the data collection will bearranged later over the phone. If the place is not CIC, transportation will be provided. Fordetails about my project, please ask M. L. (She has a copy of my research proposal.) Thevolunteers will get feedback on the research in the middle of January, 1993. The summary ofthe results will be available to the faculty by the end of January, 1993, if everything goessmoothly.The schedule for visiting your class is tentatively arranged as follows, following M.L.’s guidance. If inconvenient, please notify.Oct5(Mon) E.(Rm218) 10:10J. (Rm 203) 12:30E.(P4) 2:10Oct 6 (Tue) K. (P1) 8:30T.(Rm 205) 10:10L. (P4) 10:30Your supportive comments and encouragement for the students’ participation is reallyindispensable. I am looking forward to seeing you soon. Thank you.132APPENDIX B: “Volunteers Needed for Educational Project”If you are:* a second or third/fourth year student at CIC,* a native speaker of Japanese,could you participate in a research project for my dissertation?The purpose of the project is to explore writing and translating processes.I am looking for students who are taking the translation course (six studentsin each year), and those who have never taken translation courses (sixstudents in each year).What you will be asked to do is write a composition in Japanese and English,and translate a Japanese passage into English. You can use dictionaries. Youwill be asked to talk aloud what you are thinking while writing, and this willbe tape-recorded. Also, you will be interviewed in Japanese after each task.Confidentiality is assured. The time required is about 3 - 3.5 hours at yourconvenience. A reward for your efforts will be offered.If you are interested in participating, please write down your name andtelephone number on the sheet circulating or call me at 685-6957 (KozueUzawa) for details. I am a Ph.D. student in Education at UBC. Thank you!p.s. You can withdraw or cancel participation at any time, no explanationnecessary.n-UC)III‘-1rTn°0 S LI)U-I0*ISn-r°*rLII71*4:-%CLIII.)*Iflç5p?4r*nfSC0SLILU0rjgPff—ILISH4m(—9+....‘.-V(r4i’LlLLL\c4‘4-fIIOrT[I11C’i*C HfHf)F)4‘49crrr5134APPENDIX C: Statement of ConsentI consent to participate in this study under the following conditions:1. Title of Project: Translating and Writing Processes of Second Language Learners.2. Investigator: Kozue Uzawa (685-6957).3. Purpose of the Project: To explore language learners’ translating and writingprocesses.4. Procedures: Talking aloud while writing, followed by interviews.5. Confidentiality: Confidentiality is assured, and each participant’s real name will not bedisclosed to anybody.6. Total Amount of Time Required: About 3 - 3.5 hours.7. Participants Award: $20 Eaton’s Gift Certificate.8. Questions: Any questions concerning the procedures, etc. will be answered.9. Right to Refuse to Participate: The participant has the right to participate, or towithdraw at anytime without prejudice from the college toward anyone’s current or futureeducational plans.Signed by: Date:I acknowledge receipt of a copy of the consent form:Signed by: Date:135T{#øe.1. M: ZèU12.Jf: R (685—6957)3.4. 4, tflLt (*ci’)o1M,ko)1. Wø >5. O): U *ØZ6. Ifl: *3fl )7. j: L4fIS8. VL: o, ø1ieLk0136APPENDIX D: Questionnaire on Backgrounda. Your name is_____________________________b. Your age is___________c. Your educational background:high-school graduate______ others_________________d. Are you male? or female?_____________e. How long have you been in Canada?____________f. Have you lived in English-speaking countries before coming to Canada?g. Do you have some experiences in translating letters, reports, essays, etc.for your friends?h. Do you like to write reports, essays, poems, etc. in Japanese?i. Do you have any experience in publishing your writings?N>s0e0 ]llfnS3-- 44J(05rEE°O4-i’tIILII*?v9,0>?41\5-7j40->{-r4;;0-:3‘°14S >-.Q43c-frr-000138APPENDIX F: Model Translation<Training New Employees>April is the month for joining companies. It’s a time when many new employees comeinto being. To help these employees turn into good workers, companies start off by trainingthem.A variety of training methods are used. These methods differ according to thecompany. Some companies request a lecture from a famous person. Others teach the properuse of polite language. Sometimes, employees will go to a camp; this helps them adjust toworking in a group. Some employees do Zen mediation at a temple. Zen meditation helpsthem build a calm, relaxed frame of mind.Recently, a certain publishing company put together a textbook on employeeeducation. The textbook, which is targeted to a generation of young people raised on comicbooks, consists of comics from cover to cover. In the book, a senior employee tells a newworker-- in easy-to-understand terms -- the ins and outs of the adult world. Eat breakfastbefore coming to work, he says. Read the newspaper. Leave the house about ten minutesahead of time; pick a train that isn’t too crowded. The comic is truly considerate.To some people, such consideration is unnecessary. Yet other people feel that it is.Having suddenly embarked on the real world after many years as students, the young peopleare torn with anxiety. Companies have to make many efforts to reduce this anxiety -- or sothese people believe. The training of new employees has become an enormous issue facingtoday’s companies. (from The Nihongo Journal, April, 1990, p. 28)139APPENDIX G: Transcribed and Coded Think-Aloud ProtocolsM = metacognitive levelD = discourse levelL = linguistic levelP = personal comments< > = English translations by the researcher= phrases and sentences created by the student; words and phrases in the assigned text;words and phrases in the dictionary just read by the studentII = end of the protocolJapanese Task (Madoka- 2T)M 1. iroiro agetekuto, kitanaku natchau kedo, toire toka mo sooda shi, toire no benki gatakai toka. . ..ato doa no shita no hoo ga aiteiru no ga iyada toka....<I have many examples, it may be gross, but the toilet thing is one example, the toilet seatis too high for me, and the underneath of the toilet door is open, I don’t like this....>M 2. ato ichiban taioo dekinakatta no ga chippu dato omou’n dakedo.... ichiban wa chippuda to omou kedo, ma, ikura agetara iino ka toka...<another thing that I could not adjust well to was tipping, I guess... the most difficultadjustment was tipping, I think, well, how much should I give, and so on....>M 3. tatoeba taxi toka dattara ageru kedo, makku dattara agenai toka, nanka sono henga. .. .dakara tamatama haitta tokoroga agete iino ka... tatoeba nihonjin no resutorantoka dattara hitsuyoo nain’ja naikanaa to omoitsutsu agetari shitemasu kedo....<for example, we tip the taxi driver, but we don’t at Mcdonald’s, it’s not clear tome so, I’m not sure whether I should tip at a restaurant, for example, at a Japaneserestaurant, I usually tip, thinking it may not be necessary to tip...>P4. kore ichi-peeji kaku’n desu yo ne?<I write one page, right?>P5. demo moto moto ji ga dekkai’n desu yo<I usually write Japanese characters very big, you know>M 6. uum, yappari nihon ni nai shuukan dakara, te yuttara, saisho ni nihon ni naishuukan’tte kiite sugu chippu toka, saisho no uchiwa zenzen agete nakattashi, minna gaageteru no o mite aa ageru mono nanda to omotte. ..demo atashi no baai kanari ookuageteru to omou’n desu kedo son shiteru na’tte omou’n desu kedo140<umm, anyway, it’s a custom which we don’t have in Japan, or, when I thought about acustom which does not exist in Japan, tipping immediately came to mind. When I firstcame to Canada, I did not tip at all, but my friends were tipping, so I realized that Ishould tip... but I think I am tipping too much... .sometimes I feel I am tipping toogenerously >P7. dai wa kono mama kaite ii’n desu yo ne?<I can write down the topic as it is, right?>P8.ichiban,ichiban<first, first>L 9. chippu o watasu no watasu’tte sanzui ni do’tte yuu ji desu yo ne<the Chinese character of “watasu”, I’m not sure how to write it>P 10. dokkara kakidashite iinoka wakaranai<I don’t know where to start>M/P 11. resutoran toka taxi no hoka niwa watashita koto nai desu yo ne....<we don’t need to tip in other places besides restaurants and taxies, do we?...>P 12. sakubun o kaite iru toki’tte nanka kangaenaide kaichau taipu no yoo na ki ga suru<when I write a composition, I’m the kind of person who writes without thinking>L 13. ageru no kanji ga wakaranai (checks the dictionary)<I don’t know the Chinese character, “ageru”>L 14. ageru’tte...mono o ageru toki’tte hiragana nandesuka nee<do we use hiragana for “ageru”? umm>DIE 15. chigau (erases a whole sentence)<no, not this way> (erases a whole sentence)P 16. koko made wa nanka nami ni notte kaite kita kedo tsumazuichatta naa’tte kanji<so far, I have been writing very smoothly, but I am stuck now, sort of...>M 17. koko kara. . .kanari oomeni chippu o watashiteta’tte koto kaita kedo.<from here I wrote that I used to tip very generously, but...>LIP 18. keigo de nakute ii’n desu yo ne?<I don’t need to use honorifics for writing this, do I?>141M 19. kaita kedo, sokokara doo nattaka’tte yuujibun no koto o kaku kaa....<I’ve written what I wanted to say. Now, I will write about myself, about what happenedto me....>M 20. soretomo, soo yuu nihonjin ga ooi’tte koto kaku kaa...tte nayanderu’n desu....<or, I write that there are many such Japanese people..., I am wondering...>DIM 21. nanka,jibun no nagare ni shiyoo<oh, well, I will write about myself>M 22. jibun de chippu no gaku o kimett&tte yuuno kana.<I decided the amount of the tip each time myself, sort of...>P 23. nanka sakubun’tte kanji ja nai<somehow, this is not like a composition>L 24. ummm, “jibun de control shita”<ummm, “1 controlled myself”>L 25. “jibun de tsukai wakeru yoo ni natta” de ii kana<“I started to differentiate”, is it OK, I wonder>P 26. nihongo ni tekisanai kiga suru<I am not good at Japanese, I’m afraid>U 27. chotto saisho kara yomikaeshite miyoo kana<I will read back from the beginning>P 28. koe agete yomimasuka?<shall I read aloud?>L 29. “watashitari shimasen deshita” datte, “watashita koto ga arimasen deshita” (corrects thesentence)<I would not have tipped”, oh, no, “I have never tipped”> (corrects the sentence)P 30. nanka nihongo’tte muzukashii<Japanese is kind of difficult>L/D 31. chigau naa (erases a whole sentence)<no, not right> (erases a whole sentence)142P 32. nanka settokuryoku ni kakeru ki ga suru<it lacks persuasive power, I think>P 33. moo conclusion ni itchaoo kana’tte omotteru<I am thinking about writing the conclusion already>P 34. moo ikkai yomikaeshite miyoo<I will read back again>L 35. kyuuryoo’tte donnaji dakke....(checks the dictionary)<what’s the Chinese character for “kyuuryoo” (salary)?> (checks the dictionary)L 36. kasegu, kasegu’tte ji Wa... (checks the dictionary)<“kasegu” (to earn), how to write “kasegu”?> (checks the dictionary)PIL 37. sugoi kanji ga yowai<I am poor at Chinese characters>L 38. “gaijin”tte shitsurei na iikata nanda’kke?<is the expression “gaijin” (foreigner) rude?>L/D 39. bun ga tsunagatte inakatta (corrects the sentence)<this sentence was not connected well> (corrects the sentence)L 40. keikoo, keikoo no kanji ga wakaranai (checks the dictionary)<“keikoo” (tendency), I don’t know the Chinese characters for “keikoo”> (checks thedictionary)D/P 41. chotto saigo m ochi ga haitchatta mitai ni natchatta<I tried to make the ending funny>P 42. ura ni kaite ii’n desu ka?<may I write on the reverse side?>M 43. “shuukan no chigai de ichiban kuroo shita koto” desu yo ne?<“the most difficult adjustment caused by differences in customs”, isn’t it?>M 44. nanka chigau naa (comparing the topic and the composition)<I feel it’s not right....> (comparing the topic assigned and the composition written)P 45. nanka hisashiburi ni sakubun kaitara nanka tanoshikute<1 enjoyed writing a composition, it’s been a long time....>P 46. nanka pen ga susunjatta kedo, yappari sakubun heta danaa’tte kanjil/<I wrote without thinking much, anyway, I think I am poor at composition>!!143144Translation Task (Madoka - 2T)P 1. (reading the title) yarn ki ga denai, aa, yaru ki ga denai, ha, ha, ha, ha, nigate na monowa yaru ki ga denai’n desu yo<(reading the title) I am discouraged, oh, I’m discouraged, (laughing), this title is notinteresting, I don’t know much about it>L 2. urn, mazu, “shin nyuu sham” ga wakaranai (checks the dictionary)<urn, first, I don’t know how to say “shin nyuu sham” in English> (checks thedictionary)P 3. watashi ga dame na tokoro wa chotto, korekan&tte omottemo, honto ni jisho hiku’ndesu yo, nanka, jibun gajishin ga naito, dakara yokei jikan ga kakatchau’n desu yo,dakara kore, kanari jikan kakatchau’nja naikana<my weak point is that I don’t have a confidence in vocabulary, you know, I have tocheck the dictionary frequently, so it takes extra time, you know, I think it will take along time to finish this>P4. hon’yaku’tte yuuto, koo, jibun ga atama ga warui no o mitome nakutcha naranai kara.<translation is, well, I have to admit that I’m not smart enough to translate....>L 5. kono baai no “kyooiku” tte<“kyooiku” in this case means....>D 6. mazu zentai yonda hoo ga ii desu ne (starts reading the text)<I’d better read the whole text first> (starts reading the text)P 7. muzukashi soo....<it looks difficult....>L 8. “education” ja naishi, “kyooyoo”, “shitsuke”<no, not “education”, “kyooyoo” (education, culture), “shitsuke” (training, drill,discipline)>L 9. “educate”<“educate”>L 10. “training” da<it’s “training” !>P 11. nanka, daimei de konna dattara doonaru’n da roo<it took such a long time just to translate the title, well, I’m not sure whether I can gothrough the whole thing>145L 12. “sham” (checks the dictionary)<“sham” (employee)> (checks the dictionary)L 13. “nyuusha” nante eigo aruno kanaa (checks the dictionary) a, atta....<I wonder whether there is an English word for “nyuusha” (checks the dictionary) oh,here it is!>P 14. wakan’nai...<I don’t know....>L 15. “toki’ tte yuu no cut shite iino kanaa...<can I cut “toki” (“when”) in this sentence, I wonder....>L 16. “it’s a time” nante iimasu? ma, iika.<can I say “it’s a time”? oh, well, I think it’s OK>L 17. “because” tsukatte....<I use “because”....>L 18. ima “kooen” shirabetara, nanda “speech” ka’tte kanji<I just checked “kooen” in the dictionary, then I found “speech”, oh, well, I knew thisword>L 19. nihongo de “kono hitotachi ga yoi sham ni naru yooni kaisha wa kyooiku o hajimeru”tte natteru kedo, watashi wa “company start training for them to be a good staff’ ttekatachi de kakinaoshite....<in Japanese, it says “kono hitotachi ga yoi sham ni aru yooni kaisha wa kyooiku ohajimeru” (lit, companies start training these people so that they will become goodworkers.), but I would restructure this sentence, “company start training for them to bea good staff”>L/D 20. soshite don’nani chigau ka’tte kaite arukara “for example” to shite sore o ireta’ndesu<and it lists various examples, so I inserted “for example”>P/E 21. jisho ni nottetari suru to lucky nandesu kedo<it’s lucky if it’s listed in the dictionary>L 22. “kooen o tanomu” “tanomu” (checks the dictionary) “ask” wa “tazuneru” desu yo ne,iino kana, “ask” de ii no kana<“kooen o tanomu” (“request a lecture” ), “tanomu” (request, ask) (checks thedictionary) “ask” means “inquire”, right? Is it OK to use “ask” here, I wonder146L 23. shugo ga nai kara.... “a company”, “a company” ja hen ka naa?<there is no sentence subject, so....”a company”, “a company”, is it odd?>L 24. “ask forti ka, zenchishi ga mata wakan’naku natchau....<oh, “ask FOR”, I’m confused, I don’t know prepositions....>L 25. “to make a speech” de iika<“to make a speech”, I think it’s OK>P 26. hon’yaku shitete ichiban iyani naru koto wa nihongo o eigo ni shite baa to kaitemitesoshite baa to yonde mite jibun ga rikai dekinai koto, ha, ha, ha, ha, sore ga ichiban iyadesu ne<the most discouraging thing in translation, you know, in translating a Japanese textinto English, when I read back the translation, I cannot understand it myself,(laughing), that’s the most discouraging thing>P 27. chotto shita machigai nara wakarushi, jibun ga kaita bun dakara, demo hito ni yondemorattara hito ga wakaranakattari, jibun de rikai dekitemo sensei ni naoshite morattaramakkaka dattari, yappari chigaunoka’tte kanji de.<I think I am beginning to see some grammatical mistakes in my own writing, and Ican understand what is written, but other people often cannot understand my writing,and my teacher often corrects my English so much with red ink, so anyway, myEnglish is not good, I’m afraid...>P 28. kore no baai wa, jisho o hikanakereba wakaranai tango ga ippai aru kedo, hon tokadattara, nihongo dokuji no iimawashi toka, sooyuu no o kangaeteru no ga ichibanmayou kara, zettai toohi shimasune, hon’yaku no shukudai deta toki Wa, zenjitsu ninaru made zettai yaranai kara bunshoo ga kongaragatchau’n da keredo<in this case, there are many difficult words which I have to check in the dictionary,and in the case of translating a book, culture-specific Japanese expressions are verytroublesome. I definitely avoid translation, you know, when I have a translationhomework, I never do it until it is day before the due date, so sentences often gettangled, you know>L 29. “kotoba zukai o oshjeru” “kotoba zukai” “kotoba zukai” de nottetara ii nee (checks thedictionary) “kotoba zukai” nai ja nai<“kotoba zukai o oshieru” (“to teach the proper use of polite language”), “kotobazukai” (the proper use of polite language), if it is listed in the dictionary as “kotobazukai”, it’s lucky, (checks the dictionary), no, it’s not...>L 30. “gasshuku” (checks the dictionary)<“gasshuku” (“to go to a camp”)> (checks the dictionary)L 31. “nareru” (checks the dictionary)147<“nareru” (“to get used to”)> (checks the dictionary)L 32. “dantai seikatsu ni”, “dantai” tte iikata ga aru? (checks the dictionary) aa, “group” de iino ka, “group life”<“dantai seikatsu” (“group work”), is there an expression such as “dandai” in English?(checks the dictionary), ah, “group” ! “group life”>P 33. kono tekitoo sa ga nobinai hiketsu desu ne<I’m a lazy translator, and this may be the reason why I won’t improve, you know>P 34. moo jisho hiku dake de tsukare tchatte.. .kangaezu ni moo pappa tango kaitchaukara...kore ga eigo nobinai hiketsu da to omotte. ..kitto motto chanto yaru ko wa nanteeigo no benkyoo ni naru nante omoundaroo kedo<just checking dictionaries makes me tired....I just copy down words from thedictionary without thinking I think this may be why my English does notimprove some smart guys may think that checking dictionaries is very helpful forlearning English....>P 35. fudan eigo o yonderu tokiwa nihongo ni yakushi nagara yonderu tokimo arukedodaitai wakaranai tango wa tobashite yonde, sonomama rikai shitchaukara, aete konnafuu ni kami ni kaite mini to umaku ikanai’n desuyo ne, dakara imi wa wakatteru’ndakedo bunshoo ni dekinai toka....<when I am reading English, sometimes I may be translating into Japanese, but Iusually skip difficult words and try to understand directly, you know, so when I writethis down for translation, I cannot express myself well, you know, I cannot composethe sentence although I know itL 36. “zazen” (checks the dictionary) “meditation in Zen Buddhism temple” nihongo de ittarakantan nanoni....<“zazen” (checks the dictionary) “meditation in Zen Buddhism temple” (reading thedictionary), the Japanese expression is much simpler....>PfL 37. “zazen o suru koto mo am” toka yuuto, atashi no baai, “suru” to yuuto sugu “do”toka kimetsukete shimau’n dakedo, “koto mo am” toka yuu komakai tokoro ga yokuyakusenai’n desu, demo “do” ni shitchaoo...<when we say “zazen o suru koto mo am” (“some employees do Zen meditation at atemple”), I immediately translate “suru” as “do”, I cannot translate such a subtleexpression as “koto mo aru” (“sometimes”), you know, but anyway, I translate it asjust “do”....>P 38. jibun wa hon’yakuka ja nai kara wakaryaa iiya’tte kanji<because I am not a translator, I will be satisfied if the translation is understandable, Idon’t want anything more>L 39. “tera” (checks the dictionary) “tera” tte “temple” kana, “temple” de ii no ka<“tera” (checks the dictionary) “tera” is “temple”, I think, that’s right, it’s “temple”>148P/E 40. “zazen ni yotte shizukana...”, nihongo’tte shugo ga cut sareteru to choo atama nikimasu ne<“zazen ni yotte shizukana... “(“by zazen, quiet...”), when the sentence subject isomitted like in this Japanese sentence, I get angry, you know>L 41. kono baai dattara, watashi no baai Wa, hito o shugo ni suru toka, nani ka shugo nisuru toka kangaete, “it” ni shitchau toka...<in this case, I will make the person or something as the sentence subject, I may use“it” or....>L 42. “zazen ni yotte” tte yuuno cut shitchatte ii kanaa...<can I omit this phrase, “zazen ni yotte”, I wonder....>D 43. “comma” de tsunagechaoo...<I’m gonna connect these sentences with comma...>L 44. “tsukuru tame de aru”, “sore ni yotte tsukurareru” ,“hito ga sore ni yotte shizukanaochitsuita kokoro o tsukuru koto ga dekiru”..a, kongaragatte kita...<“in order to make”, “to be made by them”, “people can make (build) a quiet andcomposed mind through them”, ah, I’m getting confused...>L 45. “shizukana” (checks the dictionary)<“shizukana” (quiet)> (checks the dictionary)p 46. nanka chigau kanji...<I feel it’s somewhat different....>L 47. “saikin” (checks the dictionary)<“saikin” (recently)> (checks the dictionary)L 48. “aru shuppansha” tte. . ..(checks the dictionary)<“aru shuppansha” (a certain publishing company)>, how do you say... (checks thedictionary)D 49. “kyookasho o tsukutta ga”, kono nihongo ga okashii, “tsukutta ga” tte kitakara atonihiteikei ga kuruka to omottara....<“kyookasho o tsukutta ga”, this Japanese sentence does not make sense, after the “ga”,I expected a negative sentence >L 50. “kokoroe”? (checks the dictionary) “chishiki”, kaiten sasete kangaeru, “chishiki” de11....149<“kokoroe” (“information, knowledge, hints”)? (checks the dictionary), “chishiki”(knowledge), let’s be flexible, let’s use “chishiki” (knowledge)>L 51. “shakaijin”, son’na eigo arunoka naa (checks the dictionary) “member of society”<“shakaijin”, is there such an expression in English, I wonder... (checks thedictionary), “member of society”>L 52. “mangajidai” “comic” ‘jidai” wa...(checks the dictionary)<“mangajidai”, “comic”, how do you say “jidai”....(checks the dictionary)>P 53. nanka chotto chigau....<this isn’t right....>L 54. kono “wakariyasuku” tte...”easy for me to understand” “it’s easy for young people tounderstand”, nanka bumpoo ga chigau ki ga suru kedo<this phrase, “wakariyasuku”....”easy for me to understand”, “it’s easy for youngpeople to understand”, I’m not sure about the grammal>L 55. “shukkin” (checks the dictionary) “go to work”, sono mamma<“shukkin” (checks the dictionary), “go to work”, what a simple expression>P 56. yatto hanbun da<I’ve just finished the half>L 57. “10-pun gurai hayaku iku”<(“10-pun gurai hayaku iku” (“to go (leave) about ten minutes ahead of time”)>(restructuring the original Japanese sentence)L 58. “suiteru densha” dakara “man’in ja nai densha”<“suiteru densha” (lit. “a train that is empty”) means “man’in ja nai densha” (“a train thatis not crowded”)>L 59. “jitsuni” “very”<“jitsuni “, “very”>P 60. aa, ato juppun de ichi jikan....<ah, it’ll be one hour in ten minutes....>P 61. kore wa kesshite wakaru eigoja nai<this is probably an incomprehensible English>L 62. “kyuuni” tte “suddenly” ja okashii shi... (checks the dictionary) “totsuzen”,“suddenly” de ii no kanaa150<“kyuuni” is “suddenly”? it does not sound right....(checks the dictionary), “totsuzen”,“suddenly”, it may be airight....>D 63. “to yuu iken de aru” tte yuu nowa “sono hitsuyoo ga aru” tte hito no....ue ni iretchaoo<“to yuu iken de aru” (“these people believe so”), this phrase belongs to the abovesentence, “sono hitsuyoo ga am to yuu hito” (some people think that it is necessary),let’s insert it here>L 64. “doryoku”, dete konai, (checks the dictionary) “effort”<“doryoku”, what was it... (checks the dictionary) “effort”>P 65. owarii!<finished!!>D 66. yomikaeshite miyoo (reads back the translated text once)<let’s read back>P 67. ma, iika! ok ni shiyoo! II<sounds airight, it’s OK>!!151English Task (Madoka- 2T)M 1. umm, shakai ni tsuite wa nani mo shiranai<umm, I don’t know anything about the society>M 2. shakai’tte yuutemo shinbun toka yomanaishi, zenzen wakaranai<I seldom read newspapers, so I don’t know about the society at all>M 3. shakai’tte yuu’nja nai kedo, hito’tte yuuka....nihon wa nihonjin shika inai kedo,kanada, bunka ni natchau kamo shirenai kedo, kono kuni wa iron’na jinshu ga iru kara,sooyuuno’tte, nihon ni ita told gaijin mite bikkuri shiteta kedo, aru teido henken no mede mimasu yo ne, tabun sooyuu no nakunatta shi, hito no kangaekata no chigai’tte nattekuru kedo, shakai Wa....<it’s not the society, but people...., in Japan, there are only Japanese, but in Canada, itmay be a cultural comparison, there are many ethnic groups in this country, you see,when I was in Japan, I used to be frightened by the sight of foreigners; we, Japanesepeople have a prejudice against foreign people, you know, but I’ve overcome this kindof prejudice now, so, it will be a comparison between Canadian and Japanese ways ofthinking, and not their societies....>P 4. umm, sooda naa, shakai muzukashii umm<umm, well, society it’s tough umm>P 5. essay-choo de kaku’n desu ne?<it’s essay writing, right?>L 6. a, mae no translation de “shakai” no spelling machigaetchatta, “e” o “a” ni shitchatta<in the previous translation task, I made a mistake in spelling, I spelled “a” instead of “e”in “society”>L 7. “before” tte “b-e-f-o-r”, “e” ga tsukimasu yo ne?<“before” is “b-e-f-o-r”, it needs “e”, right?>L 8. “foreigner” no spell ga wakaranai (checks the dictionary)<I’m not sure about the spelling of “foreigner’> (checks the dictionary)D 9. dakara nanda’tte tokoro made mochikomu tameni...<people might say “so what?”, so I have to write about....>D 10. ma toriaezu, sassa kaita koto wa, “kanada wa maruchi-culture datte koto to nihon washimaguni dakara, nihon no hito wa nihonjin da” tte koto kaite...<so far, I’ve written this way: “Canada is a multicultural country and Japan is an islandcountry, so only Japanese people live there”....>152U 11. kono tsuzuki doo nagarete ikoo kana’tte omotte....<how to continue from here....>U 12. tsunagaranai....<it does not connect well....>M 13. demo soo yuu henken no me de gaijin o miru’tte koto Wa, nihon no shakai ga,nihonjin shika inai kara.., saikin wa gaijin mo fuete kitakedo...<we have a prejudice against foreigners because Japanese society consists of onlyJapanese people,...well, recently, many foreigners are living in Japan, but...>M 14. toriaezu, henken no koto kaite...<anyway, first, I’ll write about prejudice....>P 15. umm, yappa, shakai’tte yuu no muzukashii<umm, it’s still difficult to write about the society>M 16. nihonjin te yappa amari tanin ni hontoo no koto misenai te toko arukedo, kanajin waooppirogge’tte kanji desuyo ne, nihonjin ni kurabete, dakara kanadajin wa sugoku“friendly” dana tte kanjita<Japanese people are reserved, but Canadian people aren’t, you know, I think,compared with Japanese people, Canadian people are very friendly>M 17. kore’tte bunka no sei tte koto mo arushi, ichigai ni nihonjin wa friendly janai,kanadajin wa friendly tte ienai shi...umm<this might be due to the culture, I cannot simply say that Japanese people are notfriendly, and that Canadians are friendly, umm>M 18. kore wa seikatsu shuukan mo kakawatte iru to omoushi, kanada no kodomo’ttejibunga happyoo shiyoo to shite sugu te o ageru kedo nihonjin wa sugoku shizuka dashi,gakkoo no koto kara mo sooyuu seikatsu shuukan no chigai’tte ki ga sum shi<this is related to everyday customs, you know, Canadian children are very aggressivein the classroom, expressing their opinons, but Japanese pupils are very quiet, youknow, differences in everyday customs, including the school life, affect thesociety....>P 19. demo sore o eigo ni shite iku to owaranaku naru kara....<however, I cannot say it in English, it takes a long time....>D 20. nanka tsunagaranai naa....<it sounds incoherent, somehow....>M 21. nihon, nihon no shakai toka seikatsu shuukan toka...153<Japan is.urn, Japanese society and customs, and....>M 22. kono dai wa “kanada to nihon no shakai no ichi ban juyoo na chigai wa nani ka” to yuunode kekkyoku saigo ni sono chigai o doo omouk&tte koto o kakeba ii’n desu yo ne?<this topic, “what is the most important difference between Canadian and Japanesesociety”, I should state my own opinion about the differences, right?>M 23. kantan ni yuuto, kanada no hoo ga yokute nihon no hoo ga warui tte koto ni naru nokanaa....umm<to make it simple, it may boil down to say, Canada is better than Japan, Iwonder. ..umm>P 24. doo kaite ikoo kana....<how to continue...>P 25. muzukashii...<it’s tough...>P/D 26. moo conclusion ni itchaoo kana, sugoi hayai....<let’s go to the conclusion now, it’s so fast....>U 27. umm, yomikaeshite rniyoo kana (reads back the text created so far)<umm, let’s read back>P 28. moo ii kana, nanka motomaranakatta kedo!!<1 think it’s OK now, it is not coherent, but...>!!154APPENDIX H: ESL Composition ProfileESL COMPOSITION PROFILESTUDENT DATE TOPICSCORE LEVEL CRITERIA COMMENTS30-27 EXCELLENT TO VERY GOOD: knowledgeable • substantive • thoroughdevelopment of thesis • relevant to assigned topic26-22 GOOD TO AVERAGE: some knowledge of subject • adequate range .limited development of thesis • mostly relevant to topic, but lacks detailZ 21-17 FAIR TO POOR: limited knowledge of subject • little substance • made-O quate development of topic16-13 VERY POOR: does not show knowledge of subject • non-substantive, notpertinent • OR not enough to evaluatez 20-18 EXCELLENT TO VERY GOOD: flupnt expression • ideasclearly stated?O supported • succinct • well-organized • logical sequencing • cohesive17-14 GOOD TO AVERAGE: somewhat choppy • loosely organized but mainN ideas stand out . limited support • logical but incomplete sequencingZ 13-10 FAIR TO POOR: non-fluent • ideas confused or disconnected • lackslogical sequencing and development9-7 VERY POOR: does not communicate. no organization • OR not enoughto evaluate20-18 EXCELLENT TO VERY GOOD: sophisticated range • effective word/idiomchoice and usage • word form mastery • appropriate register17-14 GOOD TO AVERAGE: adequate range • occasional errors of word/idiomD form, choice, usage but meaning not obscured13-10 FAIR TO POOR: limited range • frequent errors of word/idiom form,choice, usage • meaning confused or obscured> 9-7 VERY POOR: essentially translation • little knowledge of English vocabulary, idioms, word form • OR not enough to evaluate25-22 EXCELLENT TO VERY GOOD: effective complex constructions • few Nerrors of agreement, tense, number, word order/function, articles, pronouns, prepositions21-18 GOOD TO AVERAGE: effective but simple constructions • minor problems in complex constructions • several errors of agreement, tense,number, word order/function, articles, pronouns, prepositions but meanjog seldom obscured17-11 FAIR TO POOR: major problems in simple/complex constructions •frequent errors of negation, agreement, tense, number, word order/funcz tion, articles, pronouns, prepositions and/or fragments, run-ons, deletions- -. meaning confused or obscured10-5 VERY POOR: virtually no mastery of sentence construction rules • dominated by errors • does not communicate • OR not enough to evaluate5 EXCELLENT TO VERY GOOD: demonstrates mastery of conventions.few errors of spelling, punctuation, capitalization, paragraphing4 GOOD TO AVERAGE: occasional errors of spelling, punctuation, capitaliZ zation, paragraphing but meaning not obscured3 FAIR TO POOR: frequent errors of spelling, punctuation, capitalization,paragraphing • poor handwriting • meaning confused or obscured2 VERY POOR: no mastery of conventions • dominated by errors of spelling, punctuation, capitalization, paragraphing • handwriting illegible •OR not enough to evaluateTOTAL SCORE READER COMMENTS155APPENDIX I: Ratings of Writing SamplesContent(max=8, min=2)FourthSecondTrained in Translation Not-Trained in TranslationLi Writing L2 Writing Li Writing L2 Writing4.2 4.0 7.0 5.57.0 5.0 5.5 4.58.0 7.0 5.8 5.5L.3.5 6.5 6.07.0 6.0 6.0 5.06.7 - 6.55.0 2.7 3.54.5 6.0 6.0 5.07.0 6.0 4.5 6.05.0 4.0 4.8 5.03.8 5.5156OrganizationFourthSecondTrained in Translation(max8, min=2)Not-Trained in TranslationLi Writing L2 Writing Li Writing L2 Writing3.8 5.0 5.8 4.56.5 6.0 5.8 5.06.9 7.0 6.5 4.0453.3 6.5 5.0 4.56.9 6.0 6.4 4.56.0 6.54.5 5.0 2.5 3.03.3 6.0 5.5 6.06.3 6.0 4.0 5.05.0 4.5 4.2 5.03.0 4.5157Language UseFourthSecondTrained in Translation(max=8, min=2)Not-Trained in TranslationTrans. Li L2 Trans. Li L25.8 5.0 5.0 6.7 6.7 5.55.5 6.6 5.0 4.8 5.8 4.05.7 7.3 6.0 6.0 5.0:3.06.5 4.8 6.0 6.4 35 55. 6.05 4.0 5.5 4.5 5.0 3.56.0 6.0 6.0 5.5 5.8 4.54.0 6.0 5.05.0 4.0 4.5 5.5 2.0 3.54.0 : 5.3 3.3 4.06.5 6.0 5.5 7.0 5.0 5.55.8 5.0 4.5 5.5 4.0 4.05.6 3.0 4.0158APPENDIX J: Participants’ Writing SamplesToshie (4T) - TranslationEducating New EmployeesApril is a month for entering companies. This is a time for new employees.Companies start several kinds of training for these new employees to make them goodworkers.There are several ways of educating new employees. It depends on the companies.Some companies ask the famous presenters to have a speech. Some companies teach how touse polite words. Some companies lodge all new employees together to be able to coexist.Also some companies make them sit in religious contemplation in the temple. By doing this,they can calm down and be quiet.Recently, one publish company has published a textbook which is about educatingnew employees. This textbook is cartoon and the cartoon teaches them to have breakfastbefore you go to the office, read the newspaper, leave home 10 minutes earlier and catch thetrain which is not too crowded. This cartoon textbook helps recent young people becausethey always read cartoons.Some people say that’s too much to do for new employees. On the other hand, somepeople say it is needed for them. Young people are worried when they work in a society afterthey graduate from school. There is an opinion that to reduce their anxiety companies shouldmake an effort. To educate new employees is a current issue for the companies.159Toshie (4T) - L2 WritingFamilies are the most important for married businessmen for both Canadian andJapanese society. However, their ways to express love for their families are totally different.In Canada, almost all businessmen put their family picture in their office, therefore,other people can notice how much he loves his family. On the other hand, Japanesebusinessmen don’t show how much they love their family. They love their family for surebut their ways of expressing love for family is just work so hard.In Canada, usually family gets together when they have dinner. Businessmen gethome by then and dinner is time for communication for a family. In Japan, businessmen godrinking after they finish working. Sometimes they don’t see their children for a few days.Even though these businessmen say that “I am working so hard for my family.”In this way, businessmen’s ways of expressing love for their families are extremelydifferent between Canada and Japan, however, they love their families.160Toshie (4T) - Li WritingL*køL 1-YAco* I77LtOQttcèj1.O è4lS Jèt:ht-cLk0 ‘tizc. /ck1fF*tWi, ø} èLøøAt L_t<tI., ttèJ.-,t’’kjL, EJ LYG)t Lk0Ak--c—ø5‘)i1, LQjt’) tO 11*—1’-eL<L-<t e’1-0. tt161Shiro (4N) - TranslationRecruit EducationApril is the month when job start. At that time there are a lot of new employees.Companies start to educate new employees to be good businessmen.There are many ways to educate them. These ways depend upon companies. Forexample, making presentation of famous person, teaching communication skills, livingtogether 2-3 days to be accustomed to group life, and doing “Zen” in the temple, for makingcaimful mind.Recently a publishing co-operation has made a text book how to educate employees,because youngers who are living in “comic” generations can easy to understand who politeworkers are. This text book is very kind comic book. It said like “Eat breakfast before goingto work.”, “Read newspapers” or “Leaving home 10 mm. early, and get on non-crowedtrain.”Someone said it is necessary for them to be kind so much. However someone agreethat being kind. Youngers who get into business suddenly from the long students life timefeel anxieties too much. Therefore companies should make an effort to reduce their anxieties.Now recruit education is a big problem in many companies.162Shiro (4N) - L2 WritingI think the most important difference between Canadian and Japanese society is thedifference of multicultural society and homogeneous society. For example, a variety ofpeople such as Chinese, Indian, Japanese and so on gather and make on society. HoweverJapanese people make society just Japanese. In Canada these group keep their own identitiesand make one society. These communities are very friendly each other, but if some otherforeign people get into Japanese society, there is a sort of prejudices for these foreigners.Canadian keep their society with many tribes friendly, but Japanese can’t do this way. Japanused to be a homogeneous country for a long time. Therefore they can’t assimilate otherpeople. Canada was a immigrant country, so they know how to manage each community andhow to be friendly.I think this difference is the most important between Japanese and Canadian society.163Shiro (4N) - Li WritingC. —L<-kJ. øL1. H*4,’tèto• /Lk0è. è<giLk, k&L ALè- eU’ è. )ø’PU1.ètJt-ti<. øè*l:A-,r’ < kttL’t*/IJ. Lfrtt1j*/11tLk0 YALØ tIE’ èØtLOY.cL/,t Lt H*LLk0 44i’k èLk0 4’t1-o e-)S *±Lk-)k164Momoko (2T) - TranslationEducation for incoming employeesApril is the month to income. On this time, we can see a lot of incoming employees.Then the company start to educate them to be a good employee.They have many ways to educate to incoming employees. Each company has differentstyle to educate. To ask some famous people to speak. To teach way of choice of words. Tobe familiar with group life, stay together for somedays. Also sometime people go temple todo Zazen. Because they believe doing Zazen follow the quiet and calm mind.Recently, a publishing company published a textbook for incoming employeeseducation, and it’s easy to get some knowledge or manner from this book as a adult in thesociety for young people who are growing up in the society which is sworm with comicbook. It tells them that they should have breakfast before they go to company. They’d betterto read the newspaper. Also it tells them that they have to leave their home earlier than normalpeople about 10 minutes, and catch a not busy train. It’s a very kind comic.Some people say that it’s too kind and unnecessary. However, somebody say it’suseful. When suddenly youth enter the society feel very uncomfortable, because they spendlong time as a student, the company need to try something for the youth to remove theiranxiety, so some people say the book is necessary. Education for incoming employees isrecognized a big issue in the today’s company.165Momoko (2T) - L2 WritingAs we know, Canada has multiculturism and as against Japan is an island country. Ithink some differences between Canada and Japan come from this point.According my experiences, Canadian has own identity definitely more than Japanese.When I did home stay in Nelson, I thought that. In other words, Canadian always say theiropinion clearly to communicate, even a child. A main cause is probably their culture.However Japanese often say something ambiguously. Canadian sometimes don’tunderstand this feeling.As a conclusion, the difference between Canada and Japan is caused by difference ofour culture.166Momoko (2T) - Li WritingC I C è 5o C< 1-’’ e H*Q)1E LL. fL±øe. ø#’-. L: <-, tf-ck’L0J[L tH*ø—-e0 4k èkL Uri1i’.fr:1- )-*ti,’e’ø0 u1e’è.4 5L7) ‘y74ijL-,k.U) è e-cej0167Naoko (2N) - TranslationIt is April to enter a company. It is time that many new employees appear. Thecompany begins to educate to these people to be good employees.There are many way of education. It is different between a company and a company.We’ll ask a famous man to have a lecture. We will teach new employees how to uselanguage. They will stay same house for one or two night for getting custom to a group life,and they will have religious meditations for making quiet and cool mind.Recently one publication company made textbook of education for new employees,but it teaches rules of a public person plainly for young men of a comic time. We’ll teach newemployees these: Have a breakfast before your attendance. Read a newspaper. Choose andget on a bus which become less crowed before ten minutes earlier, it is a kind comic verywell.Some people said we don’t need to give these kind, other said we need it. Young menwho will appear to society from long school life suddenly feel a big uneasiness. That is aopinion that a company need to have many charenges for lessen the uneasiness. Education ofnew employees is a big problem of modern society.168Naoko (2N) - L2 WritingI had many experiences in Canada, and I felt many difference between Canadian andJapanese society.When I had homestay last summer, I found difference between Canadian and Japanesefamily, because my host mother was very strong for her children. Her children obeyed theirmother. Their mother saying means demanding absolute obedience. There are no these stiffin Japanese family. Japanese children don’t obey their parents, and Japanese parents aren’tangry to their own children now. It is form of modem parents now.When I was a child, my mother got angry to me. Of course, I was scared, but I think itwas important for me to get my mother’s angry. Now, Japanese children need true parentswho are not only soft for them.169Naoko (2N) - Li Writing-Y:*-r. Eltit. tcri-ti- Z *)JIIL- 341. 1jJ-hnk:i0H*z<.*Lt<ii0 otk&th*:. eH*A


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items