Searching for reading instruction methods to promote the development of EFL writing ability

(1)

Searching for reading instruction methods to promote the development of EFL writing ability

著者（英） YOSHIMURA Fumiko

journal or

publication title

Journal of Institute for Research in English Language and Literature

number 34

page range 43‑65

year 2009‑03‑20

URL http://id.nii.ac.jp/1204/00024288/

(2)

Searching for readin g instruction methods to promote the development of EFL writing ability

Fumiko

Yoshimura

Introduction

The ability to write wellin English is becoming more and more important in the current world with the advancement of technologyand the globalization of research and business It is true even in countries where English is spoken as asecond or foreignlanguage. For exam

-

ple, i n Japan,wherepeoplelearn English as a foreignlanguage,the ability to write in English is gradually becoming recognized as an important ability for the successof individuals and the prosperity of the country. Japanesepeople,in general,have received English instruc

-

tion focused on reading and not so much on writing up to the high schoollevel. Therefore,as a way to promote the development of English writing ability,making useof Japaneselearners'ability to read Englishseems tobethe most effective. In this paper,I wouldlike to search for possibilities of taking advantage of learners'English reading ability to promote their English writing ability.

Reading

-

Writing Connection(RWC)research(e.g.,Hirvela,2004;

Tierney&Shanahan,l991)has provided us with important new under

-

standing of the connections between reading and writing abilities. For example,correlationalstudies have shown almost consistently that better writers tend to read more and that better readers tend to produce more syntactically mature writing than poorer readers(Stotcky,l983). Tiemey and Shanahan's review of research results (1991)shows that

( l ) 43

(3)

Searching for reading instruction methods

reading and writing abiiities correlate between.50 and .70 (25%to50% overlap). Thus,moderate overlap in reading and writing abilities has been demonstrated in correlationalstudies. Reading supports writing and writing supports reading. Through this supportive relationship, reading and writing are drawn together as acts of composing(Hirvela, 2004).

However,our experience as English writing instructors,especia11y in FL situations,suggests that somelearners may develop into a very competent reader without developing their writing ability very much.

They represent asymmetricaldevelopment between English reading and writing abi価es. Their reading ability does notseem to affect their writing ability and the discrepancy remains the same or becomes greater as they become more ski11edin their reading.

Thus,we have observed mixed phenomena regarding connections between reading and writing abilities. How are reading and writing abilities connected and how do they support each other? This is the question stiIlneeds to be explored. Though we stilllack a comprehen

-

sive understanding of reading

-

^writingconnections,researchers and practitionersseem to agree that reading ability is acquired earIier than writing ability. Carson and Leki(1993)note, reading can be,and in academicsettings nearly always is,the basis for writing (p.l ). According to Ferris and Hedgcock(2005),reading becomes the basis for writingbecausethe information acquired through reading contains print

-

^encoded^messages^asw e l l a s clues about how the message's grammatical,lexical,semantic,pragmatic,and rhetoricalconstituents combine to make the message meaningfu1(p.31). For FLlearners, reading is often the main source of input for the foreignlanguage and the FL writing properties and conventions. Therefore,it is plausible that EFLleamers can draw on their English reading ability and use input from reading practice inlearning about English writing.

In this paper,I wouldlike to review previous research in order to search for reading instruction methods that promote the development

(4)

of EFL writing ability. First,reading and writing abilitiesare compar

-

ed and contrasted tofind the commona価es and differences. Secondly, the nature of writing expertiseand the ways writing expertise develops are examined. Then,the possibility of L2reading practice to promote L2writing ability is discussed. Fina11y, w a y s of readings which are likely to promote the deve1opment of learners'writing ability are concretized in checklist questions and directions.

How reading and writing abiiities are connected Common features between reading and writing abilities

Krashen(l984) claims that it is reading that givesthe writer the 'feel'for the1ook and texture of reader

-

based prose (p.20). Hirvela (2004) agrees with Krashen by saying that reading supportswriting through meaningfulinput . Meaningfulinput can not only be facts but how writers think through the problems they are addressing(Bolch

&Chi,1995),and specific components that constitute writing(Hirvela, 2004). The components that reading and writing share identified by Hirvela are common r hetorilca1 or orgamzationa1 pattems in target language writing(e.g.,1ocation of such staples of writing in English as thesis statements and topic sentences) , 1inguisticfeaturesof writing (e.g.,transitionalwords and phrases, t h e frequency of certain verb tenses in specific kinds of situations) ,and examininglexicalas wellas stylisticcharacteristics of writing(e.g.,the use of informaland formal vocabulary in different circumstances) ( p . 1 l 5 ).

If one considers the constructs reading and writing abilities share, it is plausible that practice in reading can promote the development of writing ability by giving practice in the underlying constructs and cognitive processes.

In addition,reading can support writing by changing conceptualiza

-

tion of the acts of reading and writing from mere decoding or encoding into acts of composing (Hirvela,2004). Flower et al.(1990),promi

-

( 3 ) 45

(5)

nent advocates of Constructivism,conceptuaIize reading and writing as processes to compose menta1 representations of texts (p.146).

Through reading,1earnerslearn how to select,connect,and organize information from a source text and prior knowledge to create a new representation of meaning. Writers, o n the other hand,generate a number of representations that may grow in purpose,fu1lness,and coherence as an idea develops. The creation of such representations involves similar cognitive processes, s u c h as selection, organization, and connection of ideas. Thus,by using similar cognitive processes to reading,1earners can conceptualize writing in a more sophisticated way^.

Shared constructs and cognitive processes may explain the positive relationship between reading and writing abilities. It is possible for the components and cognitive processeslearned through reading to support the development of writing ability. However,the possibility is not always realized.

Differences between reading and writing abiiities

T h e 2 5 % t o 5 0%overlap between reading and writing abilities in Tierney and Shanahan's review of research(1991)can be interpreted that there is room for differences between the abilities. Ski1lacquisi

-

tion theory predicts that comprehension ski11s and production skills deve1op separately and empiricalresearch by DeKeyser and Sokalski (1996) supports the prediction. Their research suggests that input practice is better for comprehension ski11s and output practice is better for production skills. Thisline of research explains the asymmetry between reading and writing abilities in some FLlearners who have received a great deaI of reading practice and insufficient amount of writing practice.

Grabe(2001)points out two important differences between reading and writing modalities. First,while reading requires more automati

-

city of subprocesses, w r i t i n g requires more deliberate awareness.

(6)

Second,while a goalfor good reading is to reflect on the meaning rather than the Ianguage,a goalfor good writing requires reflection on the language choicesmade (p.20). These differencesmay aIso explain why practice in reading does not automaticallylead to the development of writing ability. These differences are discussed further in the fol

-

lowingsections.

Deliberate awareness

Even if a reader has deve1oped automaticity of subprocesses and orchestrating related ski1ls and processesin the act of reading,he or she may notbeable to apply the sk加s and processes to writingbecause he or she may nolonger callthem into conscious awareness in writing.

Krashen(1984)contends that-competence in writing develops the same way as competence insecondlanguage develops. Krashen distin

-

guishessecondlanguage acquisition fromsecondlanguagelearning by saying that the former is a subconscious processsimilar to firstlan

-

guage acquisition whiIe thelatter is a conscious pr^{( l}cess. Acquisition is responsible for our ability to uselanguage in both production and comprehension,while consciousleaming serves only as aneditor or monitor. Because Krashen believesthat acquisition is a far more important processthanleaming,he recommends extensive exposure to print through reading to acquire writing ability. Though his explana

-

tion supports the positive relationsbetween reading and writing abil

-

ities to a certain extent,if writing requiresdeliberateness,the very fact that related skills and processeshave been acquired subconsciously in acquisition may make them difficult to be ca11ed into conscious atten

-

tion andbeaccessed in the act of writing.

Reflection onlanguage

Another differencebetween reading and writing which is pointed out by Grabe (2001)is the focus of reflection. While the main goalof reading is to reflect on meaning,one of the important goals of writing

( 5 ) 47

(7)

is to reflect onlanguage to make sure the ideas are expressed appropri

-

ately. Though both reading and writing require attention to both meaning andlanguage,the focus of re?ection may be different. This difference may be one of the reasons why reading ability does not automatically transfer to writing ability. In fact, research by VanPatten(1990)reveals that it is difficult to attend to both meaning andlanguage form in incidentallearning mode. In such a case,aduIts tend to prioritize meaning overlanguage form. Thus,in reading for comprehension,which is the most common reading mode,it is unlikely that sufficient attention is directed to language,though it should be the focus of re

a

^ectionin the act of writing.

Clark and Clark(l977)argue that native

-

speaker comprehension is probabilistic in nature and would not rely on a thorough parsing of the utterance concerned. Native speakers use a variety of means to maximize the chances to reconstruct the intended meaning in compre

-

hension. According to Anderson and Lynch(l988),comprehension depends on three main sources of knowledge:schematic,contextual, and systemic. Schematic knowledge is made up of background knowl

-

edge and proceduralknowledge^. Contextualknowledge is made up of knowledge of situation and knowledge of co

-

text. Systemic knowl

-

edge consists of syntactic, semantic, and morpho1ogicalknowledge.

These knowledge sources are drawn on,interactively,to achieve com

-

prehension. Swain(1985) contends that becauseof a wide range of knowledge sources utilized,the underlying interlanguage system may beuntouched in comprehension. In L2reading,learners maybelikely to rely on other knowledge sources rather than systemic(1anguage) knowledge because of thelimitation of systemic knowledge. Thus, research onlanguage processing andlanguage acquisition has suggest

-

ed that comprehension alone may not be sufficient for interlanguage change and L2deve1opment,which maylater become available for production.

In summary,re?ection onlanguage or shifting sufficient attention

(8)

t o l anguage form is something whichlacks in a common mode of reading,i.e.reading for comprehension.

Reading for comprehension versus reading for revision

What welearn from reading practice differs based on our purpose of reading and our involvement into a specific reading task. Though Grabe(200l)found two main differencesbetween reading and writing modalities, i .e.deliberatenessand focus of reflection,what he meant by reading was reading for comprehension . Some other ways of read

-

ing may have the characteristics important forlearning about writing.

One such reading is reading for revision . Hayes(l996)compar

-

ed reading for revision with reading for comprehension and found that when people read to revise,they treat text quite differently from reading to comprehend. In reading for revision,people attend to text problems such as bad diction,wordiness,and poor organization. In this process,texts are c1osely examined to identify problems or to make sure meanings are effectively conveyed bylanguage. As Kennedy and Smith(2006)suggest,even if we know grammar conventions thorough

-

ly,we willmake mistakes in initialdrafts becauseour primary focus is on getting thought down on paper. In revising drafts,writers pay close attention tolanguage problems and try tofix them.

Revision also provides opportunities to consider the effectiveness of expressions and aIternative word choice. This processmay push their interlanguage development by mapping form and meaning.

Considering the effectiveness of expressions requires deliberateness and reflection o n l anguage,both of which are important in the act of writing. Possibility of reading for revision to contribute to the deve1

-

opment of writing ability has a theoreticalsupport from Krashen's theory of learning(1984),which claims that consciouslyleamedability is usefulinediting and monitoring our output. This kind of reading is especially suitable for L2learners who have leamed not acquired thesecondlanguage and L21iteracy.

( 7 ) 49

(9)

Another important difference of reading for revision is that readers are also writers who experience the text production process. They are the ones who know the intentions and ideas to express. They are the ones who willbeevaluated bythe produced writing. They are the ones who should improve the text they have produced. They try to assume the responsibility as writers intheir own reading process. Because of this responsibility,readers tend to be involvedin reading textsmore deeply.

To sum up,reading for revision has important features which cannot be found in reading for comprehension:i.e. deliberateness, attention towardlanguage,and the experience of the text production process. If these features are incorporated in reading,the possibility of transfer of reading ability to writing ability may increase. The problem in implementingthis reading,however,is thatleamers do not assume text problems or effectiveness of language expressions in their usualreading becausein most casestexts are written by experts and alreadyeditedand proofread for publication. Learners tend to treat reading texts as autonomous. This attitude toward reading texts can be changed by manipulating goals of reading. The corollary is that instruction can make a difference. An important implication from Hayes'research(l996)isthat reading behavior and the resultant learning changes depending on the purposeof reading and attention a11ocation.

Writing expertise

The concept of writing ability is in itself difficult to defu:●e because of the integration of various skills,knowledge,and processes and inclusion of linguistic,cognitive,and socialdimensions. Conceptual

-

ization of writing expertiseis also difficult because it is an elusive concept. However,it isessentialto consider the concept of writing expertise and what it is made up of so that we can have abetter

(10)

understanding of how reading ability can contribute to thelearning of writing ability. In order to consider writing expertise,an influential writing modelby Bereiter and Scardamalia (1987),empiricalstudies comparing expert and novice writers,and expertise in academic writing are investigated

Bereiter and Scardamalia model (l987)

To distinguish expert and novice writers,Bereiter and Scar

-

damalia (l987) proposea distinctionbetween knowledge telling and knowledge transforming . Knowledge telling preservesthe straight

-

ahead form of orallanguage production and requiresno more planning or goal

-

setting than an ordinary conversation. In contrast,knowledge transformation requires much more effort and skill. In knowledge transformation,the processof writing is one in which thethoughts come into existence through the composing processitself(pp.9

-

^10).

At the initialstage of the knowledge transformation,problems are analyzed and writing goals are set. They are followed by problem

-

solving activities in two domains,calledthe content problem space and the rhetoricalproblem space. Knowledge is dealt with in the content problem space,while in the rhetoricalproblem space writers attempt to tackle with how to achieve the goals of a writing task. The solutions to the content and rhetoricalproblemsbecome the input for the actual text production. According to this model,the differencebetween expert and novice writerslieslargely in the pre

-

^writing^planning^stage.

Empiricalstudiescomparing expert and novice writers

Empiricalstudies revealdetailed description of differencesbetween expert and novice writers. For example,Cumming(l989)conducted research usinglearners in an English

-

^French^bilingualprogram. In the research, leamers were classified into professionally experienced writers,average student writers,and basic writers. They were asked to write a narrative text,an argumentative text and a summary in

( 9 ) 5 l

(11)

English, which is their L2. The analysis revealed a significant difference between more andlessexpert writers.

Whilelessexpert writers exhibitedalack of control,more expert writers approached their composing with a clear notion of what it should entail,knowledge of how it should be organizedas discourse, and concerned for how it mightbebest expressed inlanguage. They formedgood rhetoricalplans toguide their decision

-

^making. ^Because

they were able to reach major decisions aboutthe gist and organization of their compositions with facility and confidence,they subsequently devoted much of their attention to evaluating their intendedexpression at the Ievelof wording and phrasing. In contrast,most of lessexpert writers formulated their gist in progressive but constrained steps, focused exclusively at thelevelof a single phrase

.

sentence,or thought.

Lessexpert writers seldom had definite notions nor did they monitor their production of writing. Lessexpert writers wrote almost allof their thought spontaneously without reflection or modifications.

Expert writers considered both the form and content of expres

-

sions. Their concerns ranged from brief considerations of preposition usage to Iengthysearchesfor words. In contrast,thelessexpert writers showed verylittle concem for the qualities of word choice.

Even if they did,they attendedlocally to surface features of thelan

-

guage they were using,applications of grammar rules,and verifications of spellings. Thusbehaviors of more expert writers are qualitatively different from those of lessexpert writers. In monitoring theirlan

-

guage use while writing,more expert writers tended to search for language which could best communicate their intendedmeaning,con

-

sidering the connotations, appropriateness, a n d purposes of their expressions,as w e l l a s theirlinguistic accuracy,whilelessexpert writers focused almost exclusively on its grammaticalaccuracy and neglected its semantic or pragmatic dimensions.

Other research(e.g.,Flower&Hayes, l 9 8 0 ; H a y e s & F l o w e r , l980;Hirose&Sasaki,l994;Raimes,l985;Zamel,1982,1983)also

(12)

indicates that more expert writers spend more time in planning and revising their work than Iess expert writers,and tend to edit their writing for content and organization rather than making surface changes to the text. Thus,empiricalstudies havesuggested that more expert writers'conceptualization of the act of writing or task schema seems to be more sophisticated than that of lessexpert writers and their attention is directed to various aspects of writing.

Expertise in academic writing

If the ability to write can vary depending on the discourse commu

-

nity,it is important to think what constitutes academicliteracy in order to consider expertise in academic writing. After entering a university, learners'writing is judged not so much by their English writing ability per se,but by their thinking ability as a member of the research community(e.g.,Spack,1988).

Performance of academic writing depends heavily onlearners' performance as readers of academic texts. However,reading aca

-

demic texts is difficult for mostlearners because the content is abstract and difficult,the text organization is different from what they are familiar with,andlearners are expected to assume a different role from what they are used to. Since the content of academic texts is abstract and theoretica1,it is very difficult forlearners to make connections between the text content and their own prior knowledge. Learners need to make an extra effort to be active readers in reading academic texts. The fact that text organization is different from whatlearners are familiar with is another reason for difficulty in comprehending academic texts. Most readers are familiar with choro1ogica1order which narrative texts use. However,academic texts are written with a variety of text organizations depending on the purpose for writing.

More importantly, i n reading academic texts,learners are expected to read them critically . Reading texts critica11y means paying atten

-

tion to what authors are doing as w e l l a s saying (Kennedy&Smith,

( 1 1 ) 53

(13)

2006,p.22). Charney (1993) contends that the more professionally advancedreaders tend to treat texts rhetorica1ly. Since they know the details of text production processvery wellas insiders of the research community,when they read works of other researchers,they try to resist the codification of the contexts of production. Bazerman(l988) notices thatthescientists he observed paid carefulattention to method

-

ologicaldetails. By paying attention to such methodologicaldetails, thescientists tried to deconstruct the seemingly smooth virtualexperi

-

ence the textlaid out for them. Instead,they tried to construct what they considered t o b e a more accurate representation of the actual laboratory procedure. In this way,researchers read texts critically and try to reconstruct the contexts in which texts were produced.

Thus,leamers are expected to assume the persona of a researcher in an academic community and evaluate texts for their credibility when they read academic texts. Considering such expectation from members of a target discoursecommunity is important in developing expertise in academic writing.

Thus,writing expertisecan vary from one discoursecommunity to another. To be an expert writer in afield,leamers need to consider expectations and conventions of thefield.

Development of writing expertise

To acquire writing ability,one should practice writing. According to DeKeyser and Sokalski(1996),practice is skill

-

^specific. ^In^their

experiment,learners who had receivedcomprehension practice im

-

provedmore in comprehension,andleamers who had received produc

-

tion practice improved more in production. DeKeyser and Sokalski (l996) argue that these resultslend support to a modelof sk加acquisi

-

tion that predictsthat declarative knowledge changes into procedural knowledge,which then is automatizedwithin the same components.

However,writing expertisedoesnot develop by means of writing

(14)

practice alone. Writing ability can develop by having an appropriate representation of writing that expert writers in a target discourse community have and by receiving input from various sources.

One of thesources isspokenlanguage. Weissberg(2006)contends that there is generalagreement among appliedlinguists and composi

-

tion theorists that childrenlearn to read and write based ontheir prior knowledge of spokenlanguage and on their experiences with social interaction. In the case of L1literacy development,a naturalprogres

-

sion from speaking to writing canbeassumed. In the caseof L2 writing,however,the picture is more complicated. According to Weissberg's observation,somelearners use both spoken and written moda価es to drive the acquisition of L2writing skills forward. Some leamers tend to give preference to either spoken or written modality over the other as the primary engine for their acquisition of new linguistic information,which may or may not be transferred to the weaker one.

Another source is through exposure to written language.

Research to date generally supports the important role of extensive and self

-

initiatedexposure to print in order to acquire writing ability.

Flower and Hayes (1980) contend that goodwriters demonstrate significant amounts of tacit knowledge concerning conventionaland formalfeatures of written textlearned through the act of reading.

Input acquiredunconsciously through extensive reading supports learners'ability to write. Thus,important input seems to come from extensive exposure to writtenlanguage in both L 1 (e.g.,Belanger,l987;

Stotsky,1983) and L2(e.g.,Krashen,l984,1993).

Regarding L2writing,another important source that provides input for writing is transfer from L1literacy. Cummins (l984)made an important claim about the notion of an underlying common proficiency acrosslanguages. Cummins explained the distinction of language sk加s in terms of basic interpersonalcommunication skills

(BICS) and cognitive/academiclanguage proficiency(CALP). BICS

( 1 3 ) 55

(15)

is defined as the manifestation of language proficiency in everyday communicative contexts ,whereas CALP is defned as the manipula

-

tion of language in decontextualized academic situations (pp.136

-

l37). According to Cummins,there is an underlying cognitive/aca

-

demic proficiency that is common acrosslanguageswhich allows the transfer of literacy

-

related skills acrosslanguages. Thus,learning to beliterate in a secondlanguage may be affected byliteracy capabilities in thefirstlanguage. However,the mediation of L2proficiency makes the picture of literacy transfer complicated.

Carson et al.(1990)explored relations acrossLland L2Ianguages and reading and writing modalitiesusing Chineseand Japanese ESL leamers. The results indicate that theliteracy sk加s are related across languages,but that the pattem of relationship varies for the two language groups and for the two modalities. The table below illus・

trates the correlations bylanguage groups for L l a n d L2reading and writing tasks.

The results suggest that reading ability transfers from L l t o L2 more easily than doeswriting ability,if it is transfer that occurred in their experiment and if transfer occurred in the direction Carson et a1.

(l990) assumed. Another important relationship canbedrawn from the results. Reading and writing abilities are related in both L l a n d L2 significantly or moderately. Thus,the results also suggest possibIe transfer acrossreading and writing modalities within the samelan

-

Table1. Correlations bylanguage groups for L l a n d L2reading and writing tasks

Chinese (n=48) Japanese( n=57) Llreading x L2reading

Llwriting x L2writing Llreading x Llwriting L2reading x L2writing

r=.3的' '

r=

-

^.^0l9

r = .27l'

r=.494' '

r=.509' ' r=.230'

r=.493' '

r=.27l' 'p<.05,''P< . 0 l Carson et al. (l990,p.256)

(16)

guage.

Possibiiity of L2reading practice to promote L2writing ability Though Krashen(1984)emphasizes a far more important role acquisition plays thanlearning in deve1oping writing ability,he also admits the usefulness of consciouslylearned ability in editing and monitoring our output under certain conditions. Some of the condi

-

tions he gave as examples are when the performer knows the rule well and when the performer is consciously concerned with accuracy.

Though most L21earners may be disadvantaged because of alesser amount of exposure to spoken or writtenlanguage,some may be advantaged because they havelearned the secondlanguage and L2 1iteracy explicitly and deliberately. In writing,monitoring and editing skills are as important aslanguage generation skills,especially at the advancedleve1,where carefulattention towardlanguage and writing conventions are necessary. Even though L2writers may not be able to generatelanguage asfluently as Llwriters,they may have stronger monitoring and editing ski11s.

Weissberg(2006) observes that some L21earners develop and refine their writing skills without a strong basis in the spokenlanguage.

Weissberg assumes that suchlearners may rely on their expertise in L1 writing to support their L2writing. They may havelearned sophisti

-

cated and effective editing skillthrough L11iteracy training. For these learners,writingmay beeasier thanspeaking. In fact,Harklau(2002) suggests the possibility of learning a secondlanguage through written language. Harklau's assumption is thatlearners'L1literacy first facilitates their acquisition of L2literacy, w h i c h in turn pushes their secondlanguage acquisition process. It is possible for suchlearners to 1earn newlexicaland syntactic information through L2reading prac

-

tice rather than speaking andlistening practice. In addition,Harklau (2002)points out the insufficiency of acquiring basic morpho

-

^syntax

( 1 5 ) 57

(17)

and phonology through spokenlanguage and^・the importance of acquir

-

ing a wide range of sophisticatedlanguage use through writtenlan

-

guage in order to be considered proficient in most domains. For some L2writers and for writersbeyond a certainlevelof leaming,conscious awarenessof rules and processes deliberatelylearned in L2reading practice can assist L2writing especially in monitoring andediting stages. In this way,theoretically it is possible for L2reading to serve for better L2writingperformance. To increase the possibilities,spe

-

cialways of reading have been proposed by some researchers. They are rhetoricalreading,writerly reading,mining and rhetoricalreading strategies. The following are the brief explanations of them.

Rhetoricalreading .

In rhetoricalreading,leamers are first taught about the main rhetoricalorganizations of the texts in the targetlanguage and then askedto use that knowledge in their own reading of those texts.

Research by Carre11and Conner(1991) suggests that explicit training in rhetoricalstructures for ESL reading facilitates ESL writing.

Considering differences in text organization acrosscultures as demon

-

strated by research in contrastive rhetoric(e.g.,Carrell,l984,l985, l992;Kaplan,1966,1987) and providingexplicit training in rhetorica1 organization in L2reading islikely to promote L2writingperformance.

Writerly reading

The reader of writerly reading takeson thepersona of the writer of the text being read. Smith(l983)wrote, To readlike a writer we engage with the author in what the author is writing. We anticipate what the author wi1lsay,so that the author is in effect writing on our behalf,not showing how something is done but doing it with us ( p . 563). Through this process,readers gain greatersensitivity to and understanding of what happens during writing. This way of reading givesleamers opportunitiesto experience text production procedure

(18)

and thereforelearners are expected to be involved in text processing more deeply.

Mining

Mining is an analogy of reading process as miners exp1oring their sources to gain valuable input for writing. Greene(1993),an advocate of mining,explains it as follows, Whereas teachers often encourage a criticalreading of individualtexts as an end in itself,mining is part of an ongoing effort tolearn specific rhetoricalandlinguistics conventions. The strategies students observe in reading can become part of their own repertoire for writing on different occasions (p.36). In mining, therefore,learners are expected not only to passively decode the text meaning, b u t to actively engage in the text to dig up valuable input for their own writing. By providing the kinds of informationlearners should1ook for in advance and offering guidance while reading, teachers can ensure thatlearners encounter the eIements of L2writing they need exposure to in order to bring those elements into their own writing repertoires.

Rhetoricalreading strategies

This approach helpslearners analyze andlearn from the rhetorica1 situation in which a text has been written. Learners are asked to1ook c1osely at the situation in which the writer produced his or her text and then to study the strategies adopted by the writer.

The idea is explained by Haas and F I o w e r ( l 9 8 8 ) a s fo1lows:

R heloncalslrategies take a step beyond the text itself. They are concemed with constructing a rhetoricalsituation for the text,trying to account for an author's purpose,context,and effect on the audience.

In rhetoricalreading strategies,readers usecues in the text,and their own knowledge of discourse situations, t o recreate or infer the rhetori

-

calsituation of the text they are reading(p.176).

( 1 7 ) 59

(19)

This idea is re

a

ectedin text

-

^basedsy1labus design (e.g.,Fees,1998) and curriculum design that advocates of genre approach are trying to promote (e.g.,Hyland,2004;Paltridge,200l).

These ways of reading have characteristics which are missing in reading for comprehension,i.e. deliberateness,attention towardlan

-

guage,and the experience of the text production process. And these characteristics canbefound in reading for revision(Hays,1996). Therefore,they are promising ways of reading to support the develop

-

ment of writing ability.

Conclusion

The purposeof this research was tosearch for reading instruction methods which arelikely to promote the development of EFLleamers' writing ability. Previous research has suggested that reading and writing are connected(e.g.,Tiemey&Shanahan,199l),reading is the basis for writing(e.g.,Carson&Leki,l993;Ferris&Hedgcock,2005) and some ways of reading arelikely to promote the development of writing ability (e.g.,Carrell&Conner,l991;Greene,1993;Haas&

Flower,l988;Smith,l983). These ways of reading,however,are too theoreticaland abstract;consequently,it is difficult to seehow they work and to discusstheir effects. Somehow theseideas need to be concretized. The fo1lowing checklist was created by synthesizing and concretizing the proposed ways of reading.

CheckIist to read English academic texts

Before you read:Survey the title,headings,and the beginning and ends of the paragraphs and ask yourself the following questions.

l. What do you think the text is about?

2. Callup your prior knowledge and feelings about the topic.

l^S

'

reading:Read the text paragraph by paragraph in order tofind the

(20)

overalistructure.

3. Where is the thesis statement in the introduction?

4. What doeseach paragraph tellyou? Check your understanding after each paragraph.

5. Find claims and data in each paragraph.

6. How has the author organized his or her ideas? How are the paragraphs related to each other?

7. What is the conclusion?

2nd reading:Read the text carefully in order to understand the details.

8. Read the text phrase by phrase. Check your understanding after each phrase.

9. Check for the predicate verb and the subject in eachsentence.

l0. Check your understanding after eachsentence.

3「 d reading:Take your time and check thelanguage form. Consult a dictionary if necessary.

l1. Findthe key words to understand the text.

l2. Find words which show the development of the text and relation

-

ships between sentences.

l3. Check for the word choice.

l4. Check for the word form and grammaticalstructure.

l5. Check for the verb tense and verb voice.

16. Check how the words are combined with each other(collocation). e.g.verb and preposition,verb and noun,adjective and noun,^, preposition and noun,etc.

After reading:Identify the rhetoricalcontext of the text and your own reading.

l7. For whom do you think the author is writing the text?

l8. For what purposedo you think the author is writing the text?

l9. Do you have any opinions about the ideas presentedin the text?

How are your opinions similar to or different fromthe author's?

20. If you wrote a paper aboutthe issue presented in the reading material,what would you write about?

( 1 9 ) 61

(21)

The effectsof the checklist were tested empirically(Yoshimura,In Press). Refer to the article for the effects. Questions and directions needtobefurther changed,rephrased,or supplementedto increase the possibility forlearners tolearn about English writing from them.

Acknowledgement

This research is supported in part by Grants

-

ⁱⁿ

-

^Aidfor Scientific Research from the Japan Society for the Promotion of Science(No.

l952050l5001).

References

Anderson,A.,& Lynch,T. (l988). L i:stemng. Oxford:0xford University Press.

Bazerman,C.(1988). Shapingωnlttenknololedge

. -

^th^egenre and actit

,

^{i t y o}f the experilmentalarticlei nsc1'ence. Madison:University of Wisconsin Press

.

hlanger, J. (l987). Theory and research into reading and writingconnections:

A criticalreview. Readin9

-

^Canada

-

^Lec

;

^turesy^f(ch^,o!^{5 , 1}oⁿev

o ; -

^l8.^e^、

?reiter

.

C.,&Scardamalia, M . (l987). T hepsychologyof u

,

rilttencomposition

.

Hillsdale,NJ:Erlbaum.

Bolch, J.,&Chi,L. (1995). A comparison of the useof citations in Chineseand English discourse. ln D.Belcher&G.Braine(Eds.),Acade

,

^m^eωnting:'n a second language

.

・

n

ssay s o n researchand

,

pedagogy (pp

.

23l

-

²⁷⁴⁾. Nor

-

wood,NJ:Ablex.

Carrell,P.L.,&Conner,U.(199l). Reading and writing descriptive texts.

Moden,;Languag eJo,ltm a 1,7 3, l 2 1

-

^134.

Carrell,P.L.(l984). The elfects of rhetoricalorganization o n E S L readers.

TES〇LQuarterly, 1 8

.

⁴⁴¹

-

^469.

Carrell,P.L. (1985). FaciIitating EFL reading by teaching text structure.

? 〇L Q m

n

erly, 19,727

-

⁷⁵²

^.

Carrell,P.L.(l992). Awarenessof textstructure:Effectsonrecall. Language Learning,42, l

-

^17.

Carson, J . E.,Carrell,P.L.,Silberstein,S.,Kroll,B., & K u e h n , P . A. (l990). Reading

-

writing relationships in first andsecondlanguage. TES〇L Quarter ly,2 4(2),245

-

^266.

(22)

Carson, J . , & L e k i , I . (1993). Introduction. In J.G.Carson&I.Leki (Eds.)

.

Readingi n t hecompo,stlti

.

onclassroom

.

' Second languagepensPectit

'

es (pp.l

-

7). Boston:Heinle&Heinle.

Chamey

.

D. (1993). A study in rhetoricaI reading:How evolutionists read The Spandrels of San Marco. In J.Selzer (Ed.), Undenstanding sctentificprose(pp.203

-

^23l)

^.

Madison:University of Wisconsin Press.

Clark,H.H.,&CIark,E.(l977). Psychology a n d l a n guage. New York:Har

-

court,Brace, Jovanovich.

Cumming,A.(1989). Writing expertiseand second language proficiency.

?ng?eLearning,3 9 ( 1 ) , 8 l

-

^l41.

Cummins, J. (l984). B

a

ingl talism a n d slpealaleducatio

,

^;. San Diego,CA:Col

-

lege Hill.

hKeyser,R.M.,&Sokalski,K. J.(l996). The differentialrole of comprehen

-

sion and production practice. Lang^MgeLearning,461(4),613

-

^642.

F:eez,S. (1998). Text

-

^basedsy

^u

^a^busdesign

^.

Sydney NSW:Macquarie Univer

-

sity.

ll;:erris,D.R.,&Hedgcock, J.S.(2005). TeachingESL composition.・Purpose,,

process

.

^and^pmctice. Mahwah,NJ:Lawrence Erlbaum Associates.

Flower,L.S.,&Hayes, J.R.(1980). The dynamics of composing:Making plans and juggling constraints. In L.W.Gregg&E.R.Steinberg(Eds.) ,^, Cogm'a^'oe p m a 加 ωn^'a^'ng(pp.31

-

⁵⁰⁾^.^Hill的^ale,NJ:Lawrence Erl

-

baum Associates

Flower,L.,Stein,V.,Ackerman, J.,Kantz,M. J.,McCormick,K., & P e c k , W.C.

(1990). Kaadfag

-

^fo一ωnh.^' ^{的 f o n}^'mg a cogm'tt'ue and soa'a1 P m^解. New York:0xford University Press.

Greene,S. (1993). Exploring the relationship between authonship and reading.

In A M.Penro能& M M . S i t k o ( 田s) , 用的n'ng o

'

^容^eiυgt^加^'^nあ.' Cagm'n'ue research加thec( )11egeωnting classroom (pp.33

-

^{5 l ) .} New York:0xford University Press.

Grabe,W. (200l). Reading

-

^writingrelations:Theoreticalperspectivesand instructionalpractices. In D.Belcher&A.Hirvela (Eds.),Linkingl i:ller

-

a a' 因.' 月e^用pediu容・ on i 2 Mmdfng一ωn'

a

'ag amneeffons(pp.l5

-

^47). ^Ann

Arbor M I : T h e University of Michigan Press.

Haas,C.,&Flower,L.(l988). Rhetoricalreading strategies and the construc

-

tion of meaning.

a

ollegeComposi

n

on andCo

,

nmunication,3 9,l67

-

^l83.

Harklau,L.(2002). The role of writing in classroomsecond Ianguage acquisi

-

tion. Jounta 1 of SecondLangm geWriting,1 1,329

-

³⁵⁰^.

Hayes

.

J . R . , & F l o w e r

.

^L.S. ^(l980)^.Identifying the organization of writing processes. In L.W.Gregg&E.R.Steinberg (Eds.),Cogmtit

,

^e^p^rocess^m

u

,

n lting (pp.3 l

-

⁵⁰⁾^.Hillsdale,NJ:Lawrence ErlbaumAssociates.

( 2 l ) 63

(23)

Hayes, J.R. (l996). A new framework for understanding cognition and affect in writing. In C.M.Levy&S.Ransdel1(Eds.),Thes a'enceof u

,

^nting^.^'

Theon'es, methods, md h;idua1 di1

a

ⁱ^e^rences^, ^{a n d} ^a^pp^lications ^(pp.l

-

²⁷⁾.

Mahwah,NJ:Lawrence Erlbaum Associates.

Hirose

.

^K^.

.

& S a s a k i , M . ( l 9 9 4 ) . Explanatory variablesfor Japanesestudents''

expositorywriting in English:An exp1oratory study. Jliour n a 1 of seco

'

^{t d}

a

ang町e ω n h'ng,3(3),203

-

^229.

Hirvela,A. (2004). (;llonnectingreading&ωntingm second langta getorll加g m struction. Ann Arbor M I : T h e University of Michigan Press. Hyland,K.(2004). G enreandsecond languageu

,

ⁿ^lti^ng. Ann Arbor,MI:The

University of Michigan Press.

Kaplan,R.B.(1966). Culturalthought pattems in interculturaleducation.

?

,

^t^{g u}^電^e^Learn^t^{n g}^,¹⁶^{, 1}

-

²⁰

^.

Kaplan,R.B.(1987). Culturalthough pattemsrevisited. In U.Conner&R.B.

Kaplan(Eds.),Wriltingacrosslanguages.・Analysisof L 2texts

.

Reading,^, MA:Addison

-

Wesley Publishing Company.

Kennedy,M.L., & S m i t h , H . M . (2006). Reading andulntingm theacademic com mlm ity (3rd ed.). New Jersey:PearsonEducation.

Krashen,S. ( 1 9 8 4 ) . Writing.・ Research,theory,ja n d applilcations. Oxford

.

^UK:

Pergamon.

Krashen, S.D.(1993). T hepoωerof reading. Englewood,C0:LibrariesUn

-

limited.

Paltridge, B. ( 2 0 0 1 ). G enre a n dthe language learningctassroom. Ann Arbor, M I : T h e University of Michigan Press.

Raimes,A. (1985). What unskilled ESL studentsdo as they w r i t e : A classroom study of composing. TES〇LQuarterty,1 9(2),229

-

²⁵⁸^.

Smith,F.(1983). Readinglike a writer. LanguageArts,60,558

-

567.

Spack

.

^R.(1988). InitiatingESL students into the academic discoursecommu

-

nity:How farshould w e go? TES〇LQt,tarterly, 1 8,649

-

670.

Stotsky,S. (l983). Research on reading/writing relationships:A synthesis and sugg1esteddirection. LanguageArts.6〇,627

-

^642.

Swain

.

M. (1985). Communicative competence:some roles of comprehensible input and comprehensible output in its development. In S.Gass & C.

Madden(Eds.) , b tput msecondlanguage acqut's^ttion. Rowley,Mass.:

NewburyHouse.

Tiemey,R.J., & S h a n a h a n , T.(l991). Research on the reading

-

writing rela

-

tionship:Interactions,transactions, and outcomes. In R.Barr, M.L.

K a m i l , P . B . M o s e n t h a l , & P.D.Pearson(Eds. ) , Handbo o k of reading research(Vol.2,pp.246

-

^280). New Y o r k : L ongman.

VanPatten,B. (l990). Attending to formand content in the input:An experi

-

(24)

Searching for reading instructionmethods

ment in consciousness. S lu di e s i n Sec o n d Ia nguageAcqut'sition, 1 2,287- 301.

Weisberg, R . ( 2 0 0 6 ) . Comeclfag spea々mg & ω

,,

fing l' seco

'

¹^{d i a}

'

¹^g¹^ageω

''

^'f

-

i ngi nstrtllction. Ann Arbor,M I : T h e University of Michigan Press.

Y o s h i m u r a , F . ( I n Press). Effects of connecting reading and writing and a checklist to guide the reading processon EFLlearners'leaming about English writing. To appear ProcediaSocialScienc、es.

Zamel, V.(l982). Writing:The processof discovering meaning. TESOL QMarter ly, 1 6 ( 2 ) , l 9 5

-

²⁰⁹^.

Zamel,V.(1983). The composing processesof advanced ESL students:Six casestudies. TES〇LQ^M^arterly,17,165

-

l87.

( 2 3 ) 65