Analyses TEXT

8/13/2019 Analyses TEXT

1/14

INTRODU TIONAnalyses of second language L2) text largely examine L2writing, while L2 spokenproduction is usually investigated in the field of conversational analysis see Markee,chap. 20, this volume). Since the emergence of applied linguistics as a discipline in the1950s and 1960s, three large domains of research have focused on various propertiesof L2 written text: structuring of the information flow in discourse, syntactic, lexical,and rhetorical features employed in L2 text, and to a smaller extent, L2 grammar, andlexical errors .In general terms, the analysis of grammatical and lexical errors in L2 written text isderived from the contrastive error) analysis that predominated in L learning researchbetween the 1950s and 1970s. Error analysis was based on an assumption that many ifnot most) L2 errors are an outcome of L1 to L2 trans fer of syntactic and lexicalregularities and language properties.Discourse analysis accounts for global features of text and the organization of ideasin writing. ontrastiverhetoric, as a subdomain of applied linguistics Kaplan, 1966),gave rise to and continues to promote an examination of discourse features in the L2writing of non-native speakers of English NNSs) see also chap. 33). It is important tokeep in mind, however, that relatively little research on L2 writing had been carriedout prior to the 1980s, when the numbers of NNSs in US, colleges and universitiesbegan to climb dramatically.Historically, important advancements in contrastive rhetoric were reflected in agrowing body of knowledge about the order and ideational structure of discourseand discourse moves, also called Discourse Blocs Kaplan, 1983), in L2 writing. Theinnovations in contrastive rhetoric studies coincided with rapid advances in text linguistics. Speaking broadly, text analysis had the goal of identifying global discoursefeatures that can be marked by means of syntactic and lexical elements, such as

6


2/14

HINKEL

verb tenses or sentence transitions Coulthard, 1985; de BHalliday Hasan, 1976, 1989; van Dijk, 1985).In a distinct domain of applied linguistics, investigationdiscourse structuringin various rhetorical traditions and aalso been extended to studies of comparative uses of teprose.Analyzing written discourse paradigms and text attofmany studies that worked with L1 writing of native speexample, Australia, Canada, the United States, United Kiand those in the English L2 writing of sp eakers of mabody of research has thus far compared discourse and teL2 writing of speakers of such languages as in alphabetiCzech, Dutch, Farsi, Finnish, French, Hebrew, Hindi GerKorean, Malay, Russian, Spanish, Swedish, Tha i, and Vivarieties of Engli sh.Most studies of various features of L2 discourse andby immediate and long-term research and curriculumas pedagogically driven needs of particular groups ofvarious locations, specific interests of individual researchdata, and/or attempts to apply the findings of predominatext linguistics to L2 text e.g., Al-Khatib, 2001; Carlson1987; Hinkel, 1994; Johnson, 1992; Taylor Chen, 1991).of syntactic, lexical, rhetorical, or dis coursal features ofHowever, in sum total, much has been learned about featwriters in different contexts and for divergent academic,purposes.This chapter provides an overview of the methods widiscourse and text, and the findings of research on macroAdditionally, computerized databases of L2 writing andconfound analyses of L2 written corpora are briefly discu

QU LIT TIVE N QU NTIT TIVEL TEXT N LYSIS

Analyzing L2 writing is a time consuming and laborioustext have varied dramatically in the size of their data saMost analyses have consisted of small-scale studies thatduced by one to a half dozen writers e.g., Arndt 1993; Cand onl y a few had access to or resources to work on) larby dozens or hundreds of L2 writers. On the other hand,cific excerpts of discourse or narrowly defined textual feato research papers or hedges and modal verbs, include10 to 25 NNSs e.g ., Basham Kwachka, 1991; Swales, 19To a great extent, the theoretical frameworks and reseayses of L2 discourse and text rely on those developed andof applied linguistics as text linguistics, di scourse analystive psychology. Although much research on L1 Englishcarried out in such disciplines as rhetoric and compositiof rhe toric has had a minimal impact on analyses of L2derpinnings of the Anglo-American and other we stern rthe classical Aristotelian and Greco-Roman foundationaorganization, and language and stylistics) not easily apptext. On the other hand, quantitative and qualitative rese


3/14

Thecontrastive rhetoric hypothesis Kaplan, 1966) is largely concernedwith discoursestructuring and logical organization of information in various rhetorical traditions.Thecrosscultural analysis of discourse postulates that systematic differences exist inhow topiccontinuity is established and coherence is developed in rhetorical traditionsand writing in different languages and cultures. In the case of l iterate and educatedspeakers of various languages, Ll-specific ways of organizing discourse and information may be transferred to the discourse organization in L2 writing. Thus, theoverarching goal of the contrastive rhetoric hypothesiswas to assist teachers and academicallybound ESL students who needed to learn to write in L2 by identifying thedifferences among patterns in written discourse.Global features of L2 written discourse, such as discourse moves, organization,structuring, as well as attendant issues of clarity, explicitness, fluidity, and contents ofwriting, represent broader and more abstract constructs than those commonly examined in analyses of text Grabe Kaplan, 1996; Hinkel, 1997, 1999; Indrasuta 1988;Johns, 997; Kaplan, 2000). For example, a number of studies has been devoted tostylistic properties of various types of L2writing, such as textual indirectness in academicessays or narrative personalization.In the 1980s and 1990s, research on the L2 writing of university students in anumber of English-speaking countries established that discourse construction andrhetoricalparadigms differ in consistent and important ways in the Ll writing of NSsand L2written prose. Additionally, investigations devoted to discourse constructionacross languages and cultures have been able to determine that, for instance, markedsimilaritiesexist in the rhetorical development of text written in some European languages,such as Czech and German or Asian languages, such as Chinese, Japanese,and Korean. For example, in German and Czech academic writing, as well as in theL proseofGerman learners of English and scholars, digressions and deviations fromthemain topic are considered to be acceptable, as are repetitions, recapitulations, andrestatements, abstract argumentation, and broad generalizations e.g., Clyne, 1987;Cmejrkova, 1996).On the other hand classical rhetoric and discourse construction in Chinese,Japanese, and Koreanwriting, as well as in the L2prose of speakers of these languages,


4/14

8 HINKEL

have been shown to include predictable organizational paraat the end, indirect argumentation, allusions, and referencesas evidence (Hinds, 1983, 1987, 1990; Matalene, 1985; Park,(1999, p. 294)points out that classical Chinese rhetoric and ston the academic writing of L2 students in U S, colleges anto the author, in L2 writing instruction, Anglo-American ddevelopment, rhetorical and linguistic norms, as well as thin which these norms are embedded should be explicitly tessential in English academic writing.The division of discourse organization paradigms into wreader-responsible or writer-responsible text was origi(1987, 1990), based on his research on Japanese, Korean, annoticed that in the written discourse in these languages, theis not necessarily presented to the reader at the beginningmain implicit throughout the text. In such texts, the responsiwriter s main idea is left to the reader, who then needs totion and the central argument from context.On the other haprose, explicitness, clarity, and lexical precision are considerit is the writer s job to construct discourse and text in whichdirectly stated at the outset, and the writer s discourse and aclear and easy to follow. At present, in L2 writing instructioor writer-responsible texts have become commonplace. Todfind an ESLwriting textbook that does not mention that init is the writer who is responsible for making the text transreader.

Other studies examined discourse construction and rhetodiverse languages as Arabic and Spanish. Specifically, Ostlewriting of Arabic speakers included a significantly higher rdinate constructions, as well as greater numbers of discousupport elements than were found in the writing of NSs.explains that in argumentation and rhetorical persuasion, corelies on parallelism, repetition, and broad generalizations,orate vocabulary. Thus, according to Sa adeddin, when wspeakers may simply transfer from L1 the usage of coordintions prevalent in interactive rhetorical style and persu sioArabic writing, Hatim (1991) similarly found that considinteraction with audience play an important role in howpersuasion, and rhetorical moves are constructed.In regard to the L2 writing of Spanish speakers, researchstrated that they write longer essays and longer, more comtences than NSs do (e.g., Carlson, 1988;Montane-Harmon, 1speakers use significantly higher rates of coordinate clauseswords, and broad generalizations when compared to thosesimilar age and educational levels. In fact, Reid s (1992) stuSpanish, and Arabic speakers demonstrated that the prosehibits coordination patterns similar to those that Ostler (1writing of Arabic-speaking university students.

According to some researchers, however, the divergencecourse structuring can also be attributed to L2 writers dand inexperience rather than the transfer of L1 rhetorical1985). Additionally, the studies of published articles writtespeakers (Taylor Chen, 1991), and essays written by Koreversity (Choi, 1988) demonstrated that discourse structuring


5/14

velopment in L1 asic an stu ent writing, L2 writers: Organize and structure discourse moves differently. Takea logically and conceptually different approach to rhetorical argumentationpersuasion, and exposition/narration Over- or underestimate the amount of readers ' background knowledge and theneed for textual clarity, explicitness, and specificity. Differently orient the reader, and introduce and develop topics. Employ different strategies for extracting/ citing information from sources, aswell as paraphrasing, quoting, and including source material in their writing. Develop text cohesion differently, with weak lexical semantic ties and themeconnections, and a preponderance of overt conjunctive markers.Silva concludes his overview of research by saying that L2writing is strategically,rhetorically, and linguistically different in important ways from Ll writing (p 669In light of these fundamental differences, Silva points out that the learning needs ofL2 writers are distinct from those of L1 writers, whether basic or skilled, and thatteachers who work with L2 writers require special and focused training to deal withcultural, rhetorical, and linguistic differences of their students.To this end, research into how L2 discourse and text are constructed, as well as

contrastive analyses of discourse, have proven to be very useful in the teaching ofL2writing and creating more appropriate curricula (e.g., Leki, 1992; Reid, 1993). Inparticular, an important outcome of research into L2written discourse is the increasedknowledge about discourse and text in writing traditions other than Anglo-American,includingsuchwrittengenres as news reports, academicpublications, studentwriting,e-mailmessages, and business correspondence. N LYSES OF MICRO FE TURES OF L TEXT

In addition to the analysis of discourse in L2 writing, a large number of studies havebeen devoted to the comparative analyses of lexical, syntactic, and rhetorical features


6/14

6 HINKEL

of L1 and L2 texts, usually produced in the contexts ofMuch research, for example, investigated the uses of disccoherence devices,modal verbs, hedges, and modifiers inJohns, 1990; Field Oi, 1992; Flowerdew, 2000;Hinkel, 1Johnson, 1992; Khalil, 1989;Mauranen, 1996; Reid, 1992;A typical study of L2 text features may undertake to dexplicit cohesive devices are used in L1 and L2 academresearchers may compare the frequencies and contexts offurthermore however and thus coordinating conjunction

and /or summary markers (e.g., in short and in sum (e.2001b; Johns, 1984; Khalil , 1989; Schleppegrell, 1996). Siof modal verbs, usage measurements can be computedpossibility and ability modals (e.g., can may might couldmodals (e.g., must should need .A study methodology can entail counting the numbverbs by type in each essay, followed by obtaining aoccurrences of a particular conjunction in all essays in a sessay. In quantitative analyses, for instance, descriptivfeatures can be obtained for NS and NNS texts. Then thesanalyzed statistically to determine whether they are usdifferently in the two samples .For example, a short excerpt from an L2 text on the tevents and competitions is presented here. In this 94wordthe ability modal can six times and the obligationmodalnotice that the text probably relies on the usage of can toWe can see that spor ts make people show their love of theispecia l team songs in every game. This is not just about ican see passion, happy or sad around audiences and playenergy and living. In the sports world, everyone is equal, evcondition to do what he or she should do the best. Also, we cais equal. So, we should be proud of our country, and we can

In this case, a computation of the frequency rate of can iand should is 2.1 2/94). Of course, no generalizationthese two modals or their frequency rates can be made bHowever, in larger and representative samples of L1 andes criptive measurement values by means of appropriaresearchers to gauge whether in an L2writing sampleof,overall usage of specific syntactic and lexical features apL1 writing sample. In L2 writing instruction, such comlead to fine-tuning course curricula, added attention toindividualized assistance for L2learners.In analyses of L2 text, the degrees of fluency, for exampof measuring relative text lengths, as we ll as lengths of sin combinations with supplemental measures of accuracomplexity (e.g ., Carlson, 1988; Hinkel, 2003a; Park, 1large-scale assessments that involve hundreds or thousanindexes have also been driven by research into specifictext (e.g ., Basham Kwachka, 1991, Hamp-Lyons, 1991a;Reid , 1993;Weigle, 2002).

Analyses of L2 text have further delved into variouassociated with various L2writing tasks.These include fo


7/14

an ngu s c ea ures o ex pro uce y a u s n aca em c, pro ess ona , anguagelearning, or literacy contexts.

IN INGS OF L TEXT N LYSESAswasmentioned earlier, Silva s (1993) synthesis of research on L2writing processesand discourse also addresses the research findings that deal with morphosyntacticand lexical features of L2 text. Specifically, according to Silva's summary, L2 writersemploy simpler sentences with more authoritative warnings, admonitions, personalreferences and narratives, and repeti tions of ideas and vocabulary. In regard to itslinguistic features, NNS prose contains fewer syntactically complex constructions,such as subordinate clauses, descriptive adjective phrases, hedges, modifiers of mosttypes, compound noun phrases, and possessives, but more coordinators , sentencetransitions, and pronouns . To summarize, compared to L1 writers, L2 writers havea restricted syntactic and stylistic repertoire, as well as severely limited range of accessible lexis that can be used in writing. In all, Silva points out that L2writers textswereless fluent . . . , less accurate . . . , and less effective (p. 668) than those of NSs, andin terms of lower level linguistic concerns, L2writers texts were stylistically distinctand simpler in structure.Based on earlier research, Silva (1993, p. 669) calls for a reconceptualization ofhow L2writing and text production are taught in Ll.S. colleges and universities: Theprevalent assumption that L1 and L2 writing are, for all intents and purposes, thesame is unexamined and has remained largely unvalidated, and themany findings ofresearch make this assumption untenable. Silva emphasizes that L1 compositiontheories, which are, incidentally, largely monolingual, monocultural, ethnocentric,and fixated on the writing of NES [native English speaker] undergraduates in NorthAmerican colleges and universities are inapplicable to teaching L2 writing, do L2writers a great disservice, and are, quite possibly, counterproductive.Almost a decade later, a large scale empirical analysis of 68 lexical, syntactic, andrhetorical features of L2 text was carried out by Hinkel (2002), who examined 1,457(around 435,000words) NS and NNS placement essays written in several universities


8/14

HINKEL

across the United States. The L2 text corpus included textslanguages: Arabic, Chinese, Indonesian, Japanese, Korean,students 0,215) were advanced and trained L2 writers,were holders of U.S. academic degrees. Hinkel reports thand composition training, L2 writers text continues to diof novice (first-year) NS students in regard to most featurThe results of her analysis indicate that even advanced anseverely limited lexical and syntactic repertoire that enabltexts restricted to the most common language features thconversational discourse. Hinkel concluded that L2 texts dL1 basic academic writing because NSs of English alread(native) language proficiency that a majority ofNNSs requicases as adults. Like Silva 993 , Hinkel calls for changeteaching L2writing that are based on the pedagogy intendeto NSs.

COMPUTERTEXT B SES N N LWRITTEN CORPOR

Beginning in the late 1980s, due to the advancements inpermitted analyses of large amounts of printed or typeddramatically altered the methodologies for analyzing syntional features employed in various written genre in EngliRayson, Wilson, 2001; Renouf Sinclair, 1991; Sinclairrapid changes in text linguistics and the new knowledge oof large text corpora have also influenced the methods awriting.As an outcome, the studies of the syntactic, lexical, aL2 text have similarly changed their methodological apprand types of L2 written corpora. To illustrate, the new gthemselves to computerized investigations, NS and NNexamined in regard to their grammar structures, syntactand complexity, and common politeness formulae (ChanBueno Perez, 2000; Li, 2000).

Other research projects investigate L2 academic writing50,000 to 500,000words and include texts produced by te(Flowerdew, 2001; Granger, 1998;Hyland Milton, 1997).the size and scope of analyses evolved from exclusive reliaings of essays by specially trained readers to computationsage rateswith which particular syntactic,lexical, and discoin learner prose (Frase et al., 1999; Hamp-Lyons, 1991b; ReIt is important to note, however, that despite the increten corpora, the investigations of discoursal, syntactic, aremained focused on the attributes of text similar to thosestudies carried out prior to the technological and methodolinguistics (e.g., Choi, 1988; Field Oi, 1992;Hinkel, 2001Ostler, 1987). In particular,language errors, coherence andcoordinating conjunctions and demonstrativepronouns), cverbs, personal pronouns, prepositions, adjectives, hedgesamong the mainstays of research on computerized text1998,2000; Granger, 1998,2002; Granger Tribble, 1998;2000;Hyland Milton, 1997;Lorenz, 1998;Reid, 1992;Rin


9/14

frequencies of their occurrence.Meunier 98 points out, however, that the typical computations of typetokenratio of words in a text used to measure the amount of lexical variation do not seemto reflecta relative quality of L2 texts because lexically varied prose is not necessarilyofgood quality. For example, if some NNS writers have a good vocabulary range butpoordiscourse organization or grammar skills, their text would not be of high quality,even when it is lexically rich . To be specific, Lorenz 998 found that in the case ofadvanced German speakers with substantial and developed L2 English vocabularyrepertoires, it was not so much a lack of accessible lexis that made their L2 writing

appear non-native, but the ways in which particular lexical items were us ed seealso Lewis,1993, 1997, and Nattinger DeCarrico, 1992, for additional discussion onidiomaticity, collocations, and formulaic lexical phrases in written text .Similar to the findings of studies carried out earlier based on smaller samples ofvarious genres in L2 writing, the results of the computerized analyses of L2 writinghave also clearly demonstrated that L2 written discourse and text are crucially andsignificantly distinct from those produced by L1 writers in English.ISSUES N COMPLEXITIES IN L TEXTRESE RCH

Technological advances in computerized text analyses of English-language corporahave allowed researchers to shed light on the real-life uses of syntactic, lexical, and collocational features in native speaker language production. In the past several decades,some of the large native English corpora have worked with spoken and written textbases as large as 30 million words Biber et al., 1999; Leech, Rayson, Wilson, 2001;Stubbs,2001 see also Collins COBUILD corpus-based learner dictionaries developedbyresearchers at the University of Birmingham .One of the key differences between analyses of corpora of L1 English-languagecorpora and L2 text analyses lies in the fact that, for instance, the corpora of writtenEnglish include large amounts of published i.e., polished and edited texts. On theotherhand, L2 texts are either handwritten or typed by NNSs, whose language usage


10/14

6 HINKEL

(e.g., spelling, word formation, or phrasing) is distinct ftype of native speaker prose.

One of the stumbling blocks in investigations of L2 harent software for converting handwriting into typed textovercome software problems, L2 texts are keyed after the1998; Shaw Liu, 1998). However, to allow computer prooccurrences of, for example, particular words or construcmade whether to correct spelling, grammar, or lexis to mthe original text intact.

Although studies of L2 text have mainly been drivefindings and publications in computerized analyses ofto date a consistent methodology for L2 corpus researchresearchers, such as Ferris (1993), claimed that computer anpossible in the near future. Computerized analyses of typescale assessments (e.g., the Test of Written English) repor(up to 21 ) of misidentified L2 textual features (Frase etDevelopers of L2 written corpora also reported additioties associated with building and analyzing text bases (Gr

When L2 text is keyed or scanned, a number of erro(e.g., omissions, additions, and misrepresentations),texts against the originals is an extremely laborious,tive process. Computerized analyses of L2 corpora do not perabstract discourse features (e.g., sufficient/explicitgrammar features (e.g., referential nonreferential prosentence or T-unit counts, or measurements of lexiccarried out manually. In analyses of L2 errors, error tagging is a manual anand the classification of errors is often subjective. Word counts of the same texts can diverge significaferent types of software are used (e.g., hyphenated wcounted as either one or two words). In such cases,mentally dependent on baseline word counts, can be

The following example is excerpted from a L2 universorder to succeed to illustrate the difficulties of computeriL2 text.

I think that the adecatte combination between this aditudes isucced's door. In one way, take a risk can be important to ricexample, shakira , who is a very important pop singer, takfamily recordered her first Long play. They spend all the monand now they are rich and she is a important artist. bu t in andall them fortune for a bat business so many people preffer wIn conclusion. In all the success cases we will discoves that thrisk and take care is the key.

In many cases, even if the writer s spelling is corrected, coand lexical features, as in this example, would be very diftext may need to be sounded out at least to understandIn light of the fact that reliable automatic analysesthose of published or transcribed L1 English language co


11/14

and text have to do w ith shor tfalls of w riters language proficiencies and restrictedlinguistic repertoire that significantly undermine L2 writers ability to produce highquality texts. Based on the re sul ts of t he ir stu di es, many researchers of L2 learningand devel opment have emph asized th at even school-age children or highly educatedadult L2learners require years of la nguage training to at ta in the levels of proficiencynecessary to create effective written prose. KNOWLEDGMENTS

My sincere thanks to Robert B. Kaplan and Rodney Hill for th eir insi ghtful commentsand sugg estions on an earlier draft of thi s chapter. Through the years, their un failinghelp and support have become indispensable.NOTES

1. For similar findings of large-scale em pirical investigations, see also, Carlson 1988), Flow erdew 2000,2001), Hink el 2001a,200 1b, 2002,2003a, 2003b), Joh ns 1997), Reid 1992), Ringbom 1998),Schleppegrell 1996 2002),and Shaw and Liu 1998), to mention just a few.2. Other L2corpora, e.g., Lon gman sLear nerCor p us and Hon g Ko n g U nive rsity of Science and Technol ogyLearner Corpus, are pro prietary and accessible onl y to the researchers affiliated w i th th ese institutions.

REFEREN ESAl-Khatib, M. 2001). The p ragmatics of lett er-writing. WorldEnglishes 20 2 , 170-200.Arndt, V. 1993). Respo nse to wri ting : Usin g fee db ac k to inform the wr iting p rocess. In M. Brock 1. Walters Eds.), Teaching composition around thePacificRim:Politics andpedagogy pp. 90-116 ). Clevedon,UK:Multilin gual Matters.Basham, C Kwachka , 1991). Reading the w orld d ifferently: A crosscultural ap p ro ach to wr itingassessment. In 1. Hamp-Lyons Ed.), Assessingsecond language writing in academic contexts (pp , 37-4 9).Norwood, NJ: Ablex.Biber, D. 1988). Variation a ross speechandwriting. Cambridge, UK: Cambridge Un iversity Press.Biber, D., lohan sson . S; Leech, G., Co nra d S ., Fine gan , E. 1999). Longman grammarofspoken and written

English. Harlow, Essex: Pearson.


12/14

INK L

Bouton, 1. (1995). A crosscultural analys is of the structure and content of lettSecond Language Acquisition, 17(2), 211- 244.Cai, G. (1999). Texts in contex ts: Und ers tanding Chinese students' English co1. Ode ll (Eds .), Evaluating writing: The role of teachers knowledge about tex279-299). Urbana, lL: NCTE .Carlson, S. (1988). Cultural di fferences wr iting and reaso ning skills. In A.languages and cultures: Issues in contrastive rhetoric (pp. 109-137). Newbury PChang , Y-Y Hsu , Y P. (1998). Requests on e-ma il: A crosscultural compa121-1 51.Choi, (1988). Text structure of Korean speakers ' argumentative essays in En129-142.Clyne, M. (1987). Cultural d ifferences in the organization of academic texts .211 247.Cmejrkova, S. (1996). Academic wr iting in Czech and English. In E.Ventola A.writing:Interculturaland textual issues (pp , 137-15 2).Amsterdam: John BenjConnor,U., Johns,A.M. (Eds.), (1990).Coherence in writing:ResearchandpedagogiVA:TESOL.

Coulthard M. (1985). An introduction todiscourseanalysis (Znd ed.). London: Lode Beaugrande, R. Dressler, W. (1972). Introduction to text linguistics.London:Edelsky, C. (1986). Writing in a bilingualprogram: Habial l vez. Norwood, NJ:Ferris, D. (1993).The design of an automatic analysis program for L2 text researcJournal of Second Language Writing, 2(2), 119- 129.Field, Y, Oi, Y 1. M. (1992). A comparison of internal conjunctive cohesion iof Cantonese speakers and native speakers. RELCJournal, 23(1), 15- 28.Flowerdew, 1. (1998). Integrating expert and in terlanguage computer corpDiscoveries for teachers and students . English forSpecificPurposes, 17(4)329Flowerdew.L. (2000). Investigat ing errors in a learner corpus . In 1. Burnard .in the teaching and language corporaconference (pp. 145 154). Frankfurt: PeterFlowerdew, 1. (2001). The exploitation of small learner corpo ra EAP materiA.Henry, R. Roseberry (Eds .), SmallcorpusstudiesandELT (pp. 363-380). AFrase, 1. , Faletti, L Ginther, A., Grant, 1. (1999). Computer analysis of the T(Resea rch report 64). Princeton, NJ: Educational Testing Service.Gonza lez-Bueno , M., Perez, 1. C. (2000). Electronic mail in foreign languagema tical and lexical accuracy, and quantity of language. Foreign Language AnGrabe, W. Kapl an, R. B. (1996). Theory andpracticeofwriting.Lond on: LongmGrabe,W. (1987).Contrastive rhet oric and text-type research. In U.M.Conno r across languages: Analysis of L2 text (pp. 115-1 37). Reading, MA: Addison-WGranger, S. (1998). LeamerEnglishon computer.London : Longman.Granger, S. (2002). A bird's-eye view of comp u ter learn e r corpus research.S. Perch-Tyson . (Eds.), Computer learner corpora, second language acquisitioning. (pp, 3-36). Amsterdam: John Benjamins.Granger, S., Rayson, (1998). Automatic lexical profiling of learner texts. IEnglishon computer(pp. 119-131). London : Longman.GrangerS ., Tribble, C. (1998). Learner corpus da ta in the foreign languaginst ruction and data driven learning. In Granger, S. (Ed .), Learner English

London: Longman .Green, C Christopher, E., Lam, J. (2000). The incidence and defects on cohein terlanguage texts: A corpus-based enquiry. English for SpecificPurposes, 19Hakuta, K., Butle r, Y, Witt, D. (2000). How long does it take English learners toJu ly 3, 2003 from w w w n f o r ~ o wHalliday, M. A. K. (1994). The construction of knowledge and va lue in the grawith reference to Charles Da rwin 's The Origin ofSpecies . In M. Coultha rd (analysis, (pp. 136 156). New York: Routledge Kegan Paul.Ha lliday, M. A. K., Hasan, R. (1976). Cohesion in English. London: Longman.Halliday, M. A. K., Hasan, R. (1989). Spoken and written language. Oxford, UKHamp-Lyons.L (1991a). Issues and di rections in ass essing second language wrIn 1. Hamp-Lyons (Ed.), Assessing second language writing in academic contexNJ: Ablex.Hamp-Lyons.L . (1991b).Reconstructing academic wri ting proficiency. In 1. Hsecond language writing in academic contexts (pp . 127-153). Norwoo d, NJ: AblHamp-Lyons, 1. , Kroll, B. (1996). Issues in ESL wr iting assessmen t: An overHat im, B. (1991). The pragmatics of argumentation in Arabic: The rise and f189-1 99.Hinds, J. (1983). Contrastive rhetoric. Text3(2), 183-1 95.Hinds, J. (1987). Reader versus writer responsibility: A new typology. In U. CoWriting across languages: Analysis ofL2 text. (pp. 141-152). Reading, MA: Ad


13/14

, - .Johns, A. 90). Coh erence as a cultura l phenomenon: Employing ethnographic principles in the academicmilieu. In U. Connor A. Johns Eds .), Coherence in writing p p. 211-2 25). Alexandria, VA: TESOL.Johns, A. 1997). Text role and context: Developing academic liieracies. Cam br idge, UK: Cambrid ge UniversityPress.Johnson, P 92). Cohesion and coherence in compositions in Malay and English. RELC Journal 23 2),1-17.Kaplan, R. B 966). Cultural thought patterns in intercultural education. Language Learning 160 ), 1-20.Kaplan, R B. 1983). Contrastive rh etorics : Som e implications for the w riting process . In A. Freedman ,I Pringle, J. Yalde n Eds .), earning to write: First language/second language pp. 139-1 61). London :Longman .Kaplan, R. B. 988). Contrastive rhetoric and second language learning: Notes towards a theory of contrastive rhetoric. In A. Pu rves Ed.), Writing across languages and cultures: Issues in contrastive thetoric pp. 275- 304). Newbury Par k, CA: Sage .Kaplan, R. B. 2000). Contrastive rheto ric a nd di scou r s e analysis: Who w rites wha t to w hom? W hen? Inwhat circumstances? In S. Sarangi M. Coulthard Eds.), Discourse and social life pp, 82-102). Harlow,UK: Longm an.Khalil, A. 989). A stu d y of cohesion a nd cohe rence in Ara b EFL college students writing. System 17 3),359-371.Leech, G., Rayson , P., Wilson, A. 2001). Wordfrequencies in written and spoken English. Lon d on : Long ma n.Leki, I 992). Understanding ESL writers. Portsmouth, N H: Boyton / Cook.Lewis, M. 993). The lexical approach. H ove, UK: LTP.Lewis, M. 1997). Ped agog ical im plica tions of the lexical approach . In J. Coa dy T H uckin Eds.), Secondlanguage vocabularyacquisition:A rationale for pedagogy pp. 255-27 0). Cambridge: Cambridge Universi tyPress.Li, y. 2000). Linguistic characteristics of ESL writing in task-based e-mail activities. System 28, 229-245 .

Lorenz, G. 998). Ove rstatement ina dvanced learners wri ting:Stylistic aspects of adjective in tensification .In S. Granger Ed.), Leamer English on computer pp . 53- 66). London: Longman .Magu ire, M., Grav es, B. 2001). Spe aking personalities in p rima r y sch ool children s writing. TESOLQuarter ly 35 4), 561-593 .Matalene, C. 985). Contrastive rhetoric: An Am erican writing teach er in China. CollegeEnglish 47 789--807.Mauranen, A. 996 . Discou rse competence: Evidence from thematicdevelo p men t in native and non-nativetexts. In E. Ventola A. Mauranen Eds.), Academicwriting:Intercultural and textual issues p p . 195-230).Amsterda m: John Benjami ns.Meunier, F 98). Computer tools for th e an aly sis of learner corp ora. In S. Gra ng er Ed .), earner Englishon computer pp. 19-37). London: Longman.Mohan, B., Lo, W. A. 985). Academicw riting an d Ch inese stu dents: Transfer and developmental factors.TESOL Quarterly 19 3), 515- 534.Montane-Harmon, M. 991). Discourse features of writt en Mexican Spa nish: Curren t resea rch in contrastive rh etoric and its im plications. Hispania 74 3), 417-425.


14/14

6 28 INK LNa ttinger, J. DeC ar rico,]. 992). Lexical phrases and language teaching. OxfPress.Os tler, S. 987 . English in parallels: A comparison of English a n d A ra bi c pr os Eds.), Writing across languages:Analysis ofL2 text pp , 169-1 85). Reading,Park, Y.M. 988). Academic and e thnic b ackground as factors a ffecting w riti nEd .), Writing across language and cultures pp. 261-273). N e wb ur y Par k, CReid , J. 992). A com pu ter text a nalysis of fou r co hesion d e vices in Engli sh dnative writers . Journalof SecondLanguageWriting 1 2)79-107.Reid , J. 9 93). Teaching ESLwriting. Englewood Cliff s, N J: Prentice-Hall.Renouf, A, Sinclair, ]. 991 . Co llocational framewor ks in English. In K. Anglish corpus linguistics pp, 128-143). New York: Lon g man.Rin g b o m, H . 998). Vocabulary freq uencies in ad vanced learn er En glish. In S.on computer pp . 41- 52). London : Lon gman.Saadeddi n, M. A 989). Text d ev elopmen t a nd Ara bic-Eng lish negative inte100 ), 36-51.Scarcella, R , C hunok, L. 989). Differen t pa t hs to w ri ting pr oficiency in a secinvestigation of ESL writers of short-term an d lon g-t erm resi d e nce in th e U Ed.), The dynamic interlanguage: Empirical studies in second language variatiPlenum.Schleppegrell, M . 996). Co njunction in spoke n English a nd ESL w riting . AppliSchleppegrell, M. 2002). C ha lleng es of th e scie nce regi st er for ESL s tu d ents:In M. Schleppeg rell M. Co lombi Eds.), Developing advanced literacy

, 119-1 42). Ma hwah , NJ: Law rence Erlb au m Associates .Scollon, R 991). Eigh t legs an d on e elbow : Stance and struc ture in Chinese EProceedings of the 2nd North American Conference on Adult and AdolescentIntern ational Reading Association .Shaw, Liu, E. T. K. 998). Wh at develops in the d evelo p ment of seconLinguistics 192), 225-254.Silva, T. 993). Tow ard a n understanding o f th e di st inct nature o f L2 writ iimp lica tions. TESOLQuarterly 274), 657-677.Sincl air, J. 991). Corpus concordance colloca tion. Oxf ord , UK: O xford Uni versStu b bs, M . 996). Text and corpus analysis.Oxford, UK: Blackw ell.Stubbs, M. 2001). Words and phrases:Corpus studies oflexicalsemantics.Oxford ,Sw ales, J. 990 . No n-n at ive speaker gra d ua te enginee rin g st u d en ts an d thheren ce and local man agement. In U. Connor A Joh ns Eds .), CohereA lexandria, VA: TESOL.Taylor, G ., C he n, T. 991). Lin g ui st ic, cultura l, an d subcultural issu es in coAnglo-Amer ican a nd Chinese scientific text s. AppliedLinguistics 12,319-3Tri bb le, C. 2001). Sma ll corp o ra and teac hing writing . In M . Ghade ssy, A. HSmall corpus studies andELT pp, 381-408). Amsterdam: John Benjamins.Valdes, G., Sanders, 999 . Lat ino ESL students and the developC.Co oper L. O dell Eds.),Evaluatingwriting:Theroleofteachers knowledge pp. 249- 278 ). Urbana, L N ational Co uncil for Teacher s o f Engl ish .va n Dij k, T. Ed .). 985). Handbook of discourseanalysis. 4 vols) London : AcadWeigle, S. 2002). Assessing writing. Cambridge, U K: Ca m bridge Un ive rs ity P

Analyses TEXT

Documents

Transcript of Analyses TEXT