Adolescent Chinese Writers: Juggling sociocultural influences and writing demand s Kerri-Lee Krause Dan O'Brien Macquarie University,

School of Education, Australia Introduction Sociocultural theory argues that in order to truly understand the human conditio n we must interpret it in its social, cultural and historical context. This pape r is concerned with the sociolcultural aspects of specifically argumentative wri ting - created by Chinese adolescent writers. In particular, our focus rests on the interdependence of the social and cultural factors within the learning conte xt. We argue that, when teaching and assessing L2 argumentative writing, factors such as students' age and stage of development and the sociocultural nature of the learning environment must be taken into account if we truly seek to understa nd more about the writing process and the teacher's role in it We are reporting on a study conducted in collaboration with three Chinese teache rs of English in three high schools in the Henan Province of China. The teachers collected 114 students' argumentative essays, written in English. The essays we re marked by the three teachers whose classes were involved in the study and tea chers' feedback was analysed thematically. The essays were also analysed for lin guistic features. Our aim was threefold. First, given that audience awareness is a fundamental characteristic of good argumentative writing (Connor, 1990), and given the sociocultural, linguistic and developmental factors that render audien ce awareness a particularly challenging skill for L2 writers, we wanted to inves tigate the level of audience awareness evident in the L2 students' argumentative essays, using Berrill's Audience Awareness Scale (1992). Second, we aimed to ex tend our analysis through close study of the range and number of metadiscourse f eatures in the essays. Finally, we sought teacher feedback on the essays to dete rmine whether or not there was evidence that teachers foster audience awareness through their written feedback on essays. The following literature review draws together three interconnected issues; name ly, the sociocultural context and its role in the pedagogical process, the socio emotional characteristics of adolescent writers and the challenges of composing argumentative text in a second language. Each of these issues in turn presents a range of contextual challenges for adolescent writers. The Sociocultural Context Bruner (1987) argued that the: "Explanation of any human condition is so bound t o context, so complexly interpretive at so many levels, that it cannot be achiev ed by considering isolated segments of life in vitro" (as cited in Luria, 1987, p. xii). Consistent with this position, we commence our literature review on the argumentative writing of Chinese students by examining the context in which the ir writing takes place. Chinese culture is characterised by collectivism and loyalty to the family and s ocial group (Salili, 1996; Stevenson & Lee, 1990; Wilson & Pusey, 1982). Researc h indicates that Chinese students typically associate academic achievement with success in family and social life. These combined with high parental expectation s are important factors influencing students (Salili, 1995). The Chinese Classroom and Pedagogy in a Sociocultural Context

Chinese classrooms typically comprise at least 40 students and the context is sa id to be characterised by strict discipline, authoritarian teachers, an exposito ry teaching style, and much homework (Biggs, 1992; Salili, 1996). The pressure t o succeed in the stringent testing regime - comprising weekly and monthly tests as well as competitive entrance examinations (Hau, 1992; Salili, 1995) - plays a significant role in curriculum and pedagogical principles. The pedagogy within Chinese classrooms tends to emphasise memorizing of lessons from Chinese textbooks and repeated practice of new skills (Liu, 1986). Despite the view of Ballard and Clanchy (1991, p. 23) that this emphasis on rote learnin g and memorisation leads to a "reproductive approach to learning", Kember and Go w (1991) (see also Hird, 1996) contend that the apparent rote-learning approache s adopted by Asian students may be explained more by the "nature of the curricul um and the teaching environment than as an inherent characteristic of the studen ts" (p. 119). The Adolescent Writer within the Sociocultural Context Adolescence is a period in which young people develop the ability to think hypot hetically (Piaget, 1952) and to place themselves in the position of others - to acknowledge multiple viewpoints simultaneously and to understand how these perta in to their own ideas (Lapsley et al., 1986; Selman, 1980). At the same time, ad olescents are struggling to come to terms with their identity and their sense of self (Berk, 2000; Erikson, 1980; Moshman, 1999). The Chinese traditionally collectivist emphasis has implications for the develop ment of self and identity among adolescents, and in turn, for the writer's self. In addition, the Chinese adolescent L2 writer faces the difficulty of creating and defining a "new" self, a new identity appropriate for use in the second lang uage (Rodby, 1990). In argumentation, the skills of thinking for oneself and arg uing a point convincingly are highly valued. These skills may require L2 student s in communitivistic cultures - such as China - to develop "a whole new 'identit y' to approach tasks in the expected way" (Kirby, Woodhouse, & Ma, 1996, p.142; see also Shen, 1989). The "dialectic of identity and difference" (p. 48) in whic h the L2 writer is engaged is intensified in the composition of argumentative te xt since this task, by definition, demands that the writer identify and critical ly interpret a range of different - often opposing - views in terms of his/her o wn position on a specified topic. The writer is expected to convey these views t o an absent addressee whilst maintaining a "certain distance" from the subject m atter (Golder & Coirier, 1994, p. 188). Second Language Argumentative Writing in a Sociocultural Context Writing is, in essence, a social act (Rubin, 1998) New conventions must be learn ed and new literacy skills acquired. In addition to limited vocabulary and unfam iliarity with the rhetorical structures in the second language (Kirby, Woodhouse , & Ma, 1996), the novice L2 writer also has limited understanding of the intend ed audience, their expectations and their demands (Intaraprawat, & Steffenson, 1 995). The novice writer "therefore lacks adequately differentiated audience cons tructs, lacks inference rules for selecting constructs, and lacks a rich body of cues . . . from which to draw social inferences" (Rubin, 1998, p. 57). Rubin, Piche, Michlin and Johnson (1984, p. 229) identify issues pertaining to t he relationship between sociocognitive development and argumentative writing. Th ese include: acknowledging that the audience may have an opinion different to that of the wri ter; the capacity of the writer to construct his/her own arguments, as well as those

of the audience; and the degree to which the writer includes and represents the audience's viewpoints , values and associations. Hays' (1988) analysis of adolescents' writing indicated that in the majority of essays analysed, these novice writers were able to construct their own arguments and acknowledge the existence of opposing arguments; yet there was little evide nce in their writing that they understood why their audience might question thei r points (as cited in Berrill, 1992, p. 83) The Audience Awareness Scale and Metadiscourse Features The Audience Awareness Scale (Berrill, 1992, p. 85) provides a vehicle for analy sing the extent to which writers demonstrate an understanding of the needs of th e audience for which they are writing. Berrill identifies a number of levels of audience awareness outlined in Table 1 below. These range from minimal awareness of audience needs - the egocentric level, to an appreciation and accommodation of alternative viewpoints and a concomitant integration of these views into the argument. Table 1 Audience Awareness Scale (Source: Berrill, 1992, p. 85) Level of audience awareness Characteristic(s) of the writing Egocentric Writer assumes that audience shares same attitudes and knowledge base. Cursory awareness of alternative view Writer anticipates opposing position but does not deal with it. Acknowledges alternative point of view Writer recognises alternative point of view, but does not accept its validity. Appreciates alternative point of view Writer recognises validity of alternative point of view and qualifies initial po sition accordingly. Accommodates alternative point of view Writer explicitly acknowledges validity of alternative view and incorporates thi s view in argument structure. A complementary means of determining the extent of students' awareness of audien ce or reader needs involves analysing the linguistic features of their written t exts. Metadiscourse features are those linguistic markers which, while not inher ently necessary to the topic, show that the writer is aware of the importance of communicating the semantic content of their thoughts to the audience in order t

o be understood clearly. Metadiscourse comprises words and phrases that help "re aders organize, classify, interpret, evaluate, and react to such material" (Vand e Kopple, 1985, p. 83). The task of constructing a successful piece of argumenta tive writing requires that the composer of the text harness cognitive, linguisti c and rhetorical skills while simultaneously maintaining an acute awareness of a udience. Such a task is challenging for any writer, but is particularly so for t he L2 writer coming to terms with the intricacies of the English language and th e associated sociocultural conventions. We now turn our attention to the thrust of our investigation, the identification and subsequent enhancement of audience awareness within the L2 writing environm ent. Our central question therefore is: when teaching and assessing writing, how can teachers simultaneously take account of the sociocultural context of the ar gumentative writing task and the unique cognitive and socioemotional characteris tics of L2 learners, while providing practical assistance to raise their student s' awareness of the audience for whom they are writing? The Study: Investigating Chinese adolescents' writing in a sociocultural context The study reported here is ongoing and involves a number of phases. Initially, t he focus issue is writing assessment, with a particular emphasis on the sociolin guistic components of the writing and assessment process. We based our investiga tion on a sample of argumentative essays written in English by a group of studen ts in China. We had three research questions. What does Berrill's Audience Awareness Scale (1992) reveal about the level o f audience awareness in Chinese L2 students' argumentative essays? What is the relationship between Chinese L2 students' audience awareness and their use of metadiscourse features in argumentative essays? Is there evidence that teachers encourage and enhance audience awareness thr ough their written feedback on essays? Sample Our study was based on 114 essays collected from students in their senior years of secondary school in the Chinese provincial capital of Zheng Zhou, in Henan Pr ovince. Students from three schools took part in the study. These schools were s elected because they each catered to students who were considered relatively com petent English as a second language users. Procedures The writing tasks were administered by class teachers under the guidance of a Ch inese teacher educator acting as a research assistant. Students were asked to wr ite an essay expressing their views on the topic: 'Is it better to do your highe r education in China or overseas?' Discussion of the topic took place in class. In a subsequent class, students wrote the essay. In all cases the essays were co mpleted during a class period of approximately 50 minutes and no word limit was imposed. We also asked three teachers of high school English in China to mark an d provide written feedback on a random selection of 50 of the essays Analysis Two trained raters made a global judgement about students' essays on the basis o f Berrill's (1992) Audience Awareness Scale, categorising each piece of written text according to the five levels of the scale. The two raters then analysed the essays using the metadiscourse taxonomy outlined in Table 2. Table 2 Metadiscourse Features and their Purpose

(Source: adapted from Intaraprawat & Steffensen, 1995 and Vande Kopple, 1985, 19 97) Textual Features Connectives: show the organisation of text Elaboration/ explanation: rephrase to ensure clarity of expression Illocutionary markers: identify the purpose of the sentence or paragraph to foll ow Validity markers: encode writer's certainty about the truth of the content Interpersonal Features Narrators: provide the source of ideas and facts Attitude markers: convey writer's feelings about content Commentaries: address and engage reader more directly

The taxonomy comprises seven categories specifically adapted by Intaraprawat and Steffensen (1995) for the purposes of analysing argumentative writing. These ca tegories are: connectives, elaboration/explanation, illocutionary markers, valid ity markers, narrators, attitude markers and commentaries. One of the researcher s read the essays and categorised the metadiscourse features listed above. The s econd researcher then reanalysed each essay. The interrater reliability for iden tifying metadiscourse features was 0.85. In addition, we analysed the three Chin ese teachers' written feedback on their students' essays. The written feedback w as content analysed using the framework of textual and interpersonal linguistic features (based on the metadiscourse taxonomy of Intaraprawat & Steffenson, 1995 ). To our knowledge this means of analysing teachers' written feedback on argume ntative essays has not been undertaken before. Results and Discussion The Results and accompanying discussion and consideration of implications are pr esented in three parts, according to the three research questions identified ear lier. What does Berrill's Audience Awareness Scale (1992) reveal about the level of au dience awareness in Chinese L2 students' argumentative essays? Table 3 contains results of the categorisation of students' essays according to Berrill's (1992) Audience Awareness Scale. Results indicate that the majority (6 0%) of essays were identified as predominantly egocentric. Essays in this catego ry reflected an assumption that readers shared the writer's attitudes and had th e same knowledge base - summarised in the following closing sentence "This is my idea and I won't change it in my life, for our country and people." Table 3 Categorisation of Essays According to Berrill's (1992) Audience Awarenes s Scale Level of audience awareness Number of Essays

(N=114) 1. Egocentric 69 (60%) 2. Cursory awareness of alternative view 22 (19%) 3. Acknowledges alternative point of view 17(15%) 4. Appreciates alternative point of view 4 (4%) 5. Accommodates alternative point of view 2 (2%) While the majority of essays failed to acknowledge alternative viewpoints, appro ximately one-fifth of them anticipated opposing positions, though they failed to deal with them directly. For example one student wrote: "I saw a lot of chinese go to overseas to go on their higher education, and some of them did not decide d(sic) to come back, I was angry with them. An example of a writer's ability to recognise an alternative viewpoint is illustrated in the following excerpt: "Now , many people realized the foreign university are all better than the university of our country. The writer does not accept the validity of this statement and d evotes no time to explaining why the argument is "not correct". A small number of students demonstrated their ability to recognise the validity of opposing arguments and an even smaller minority were able to sustain this rec ognition throughout the essay. One student began with the statement, "There are always two sides to everything...". S/he proceeded to outline the merits and dis advantages of each side of the debate, concluding with a personal opinion: "Fina lly, if I have to draw a conclusion about this question, I think it is better to doing (sic) my higher education overseas". The student presented opposing views and was also able to accommodate these into the argument, ultimately expressing a personal view. Audience awareness and adolescent egocentricity The egocentricity of adolescence may impede the novice writer's capacity to anal yse and present these varying viewpoints objectively. Even expert adult writers reveal strong levels of egocentrism as they attempt to reconcile their own ident ity and viewpoint with that of an audience who holds an alternative opinion or r ange of views (Berkenkotter, 1981 as cited by Berrill, 1992). The role of culture in the writing process Our analysis would not be complete without some recognition of the pivotal role played by cultural values and beliefs in the process of developing the writer's self. The literature highlights the significance of the communitivistic concern

of the Chinese for the well-being of family, community and society (Hsu, 1985; S alili, 1995). The topic evoked a strong sense of patriotism and concern for the greater good of the community. Some examples to illustrate this concern are incl uded below: If you remember your own homeland you should return to your own homeland after g raduation. You can use those (sic) knowledge to build our homeland . . . There a re (sic) better wisdom in China. At the same time, it was evident that some students were attempting to express t heir views about what would be best for them as individuals. For example: Overseas, if that country is a developed country the study conditions is (sic) g ood and I can study better. Being alone in overseas can build my character. Thus, the apparently egocentric nature of these essays must be interpreted in th e light of the sociocultural context in which they were written. Review and implications In summary, our analyses of audience awareness using Berrill's (1992) scale reve aled that the majority of essays reflected an egocentric orientation suggestive of a low level of audience awareness. We found that Berrill's instrument provide d a useful starting point for identifying the level of audience awareness appare nt in students' essays at a macro level. We believe this tool could be used for the purposes of both teaching and assessing L2 argumentative writing. Students' should be made aware of what constitutes audience awareness and how to demonstra te it in their writing. What is the relationship between Chinese L2 students' audience awareness and the ir use of metadiscourse features in argumentative essays? We used the metadiscourse taxonomy of Intaraprawat and Steffenson (1995) as a me ans of identifying ways in which the student writers demonstrated their awarenes s of the needs of their audience or reader. Table 4 shows the proportion and mea n occurrence per essay of the metadiscourse features identified.

Table 4 Proportion and Mean Number of Metadiscourse Features Used Per Essay Textual Features Mean (%) Connectives 9.92 (41%) Elaboration/ Explanation 4.94 (20%) Illocutionary Markers 0.33 (1%)

Validity Markers 4.46 (18%) Interpersonal Features Narrators 0.52 (2%) Attitude Markers 2.79 (12%) Commentaries 1.54 (6%) Total 24.50 (100%) Frequency counts indicate that textual features (80%) dominated students' use of metadiscourse features. The feature used most often was the connective. This co mes as no surprise, since connectives form the basis of the second language lear ner's repertoire of linguistic skills. These findings are consistent with those of Intaraprawat and Steffensen (1995) who found that between 30% (in the case of good essays) and 47% (in the case of poor essays) of all metadiscourse expressi ons used by ESL university students were connectives. This was clearly the most frequently used feature in the essays analysed for this study. Students also used elaboration and/or explanation of ideas in their essays. Howe ver, the difference between the frequency of these features and the mean number of connectives used is significant. The use of elaboration and explanation was t ypically heralded by the phrases: "for example", "such as", "that is to say", an d "so that". Many of the phrases or expressions used to introduce elaborations or explanation s form part of standard essay writing texts for students of English in Chinese s chools. For example, in the commonly prescribed secondary school English text St art English Writing, "signal words" or phrases such as "for one thing", "for exa mple" and "another example is" are provided in the form of lists from which stud ents are told to choose phrases to be inserted in cloze passages. It is expected , then, that students will not only commit such phrases to memory, but will inco rporate them into their own writing - as is apparent in the scripts analysed for this study. Validity markers - comprising emphatics and hedges - were also relatively well r epresented compared to the other metalinguistic features used. In the argumentat ive essays analysed, emphatics were the most frequently used validity markers. T his may be partly attributed to the topic given: "Which is better, to do your ed ucation in China or overseas?" While approximately one third of students in the sample suggested that it might be better to do their further study overseas, the

re was unanimous agreement that they would only study overseas in order to be ab le to return to their "mother country". Emphatics thus serve the purpose of conv eying the strong sense of patriotism which characterised the essays analysed. "Our chairman, Mao Zedong even said: 'The world is ours, it is also yours' ... Y es the world is ours, and to build our homeland we must work. It is obvious that to build our motherland needs knowledge. Every student must certainly work hard , then build our motherland when we grows up." In this way, one perceives evidence of cultural values influencing the nature of metalinguistic features used. Interpersonal features were much less apparent than were textual features in the se students' essays, comprising only 20% of the mean number of metadiscourse fea tures used. Such features as attitude markers and commentaries allow writers to express their opinions regarding the content of the text and to address the read er directly. These forms of audience awareness did not feature prominently in th e Chinese students' essays. Some students engaged reader attention through the u se of inclusive phrases such as "let us ..." and "Let's think...", and through r hetorical questions. The repeated use of the pronouns "we" and "you" was a featu re of some essays. Students' failure to employ commentaries more effectively may reflect an assumption that readers already share their views thus obviating the need to elicit a specific reader response. Attitude markers and commentaries "c onvey the writer's intentionality and function to increase the acceptability of the text" (Intaraprawat & Steffensen, 1995, p. 259). Unless students are provide d with guided practice on how to render the argument acceptable to the reader th rough the use of metalinguistic features, they may remain unaware of the importa nce of their role in convincing the reader through the use of rhetorical devices . In those essays that used narrators, some writers quoted parents and older relat ives, others cited Chinese proverbs and ancient Chinese wisdom. For example: "Long, long ago, there lived a famous man named Xuu Zi, he said:...". However the most common forms of reference to sources of information were couche d in vague terms such as: "most people think", "everyone knows", and "it is said that...". This reliance on family and community values once again illustrates t he prevalence of value of the greater self (Salili, 1995) and its impact on stud ents' formation of written arguments. Review and implications Table 4 demonstrates that students in the sample did make use of metadiscourse f eatures, yet these were limited in number and scope. Students used considerably more textual markers than interpersonal markers. These results are consistent wi th the tendency toward a dominance of textual markers evident in the poor essays produced by ESL university students in Intaraprawat and Steffensen's (1995) stu dy. The findings of this study suggest a failure on the part of the Chinese adol escents sampled to engage with their audience on a personal level. They did not typically convey their personality and feelings on the subject as effectively as they might had they used a wider range and number of interpersonal metadiscours e features in their writing. Direct instruction of metalinguistic features has been advocated (see Cheng & St effensen, 1996; Steffensen & Cheng, 1996), but there is little point in providin g students with endless lists of metalinguistic features to memorise and use. St udent writers must understand the rationale for using such markers. L2 writers, in particular, will benefit from being made increasingly aware of the needs of t heir readers/audience, but this process must be accompanied by provision of ling

uistic strategies for meeting those needs. Teachers may provide these strategies through the use of modelling, through class exercises which allow students to i dentify a range of metadiscourse markers in samples of writing, and through care ful written feedback directly addressing this linguistic aspect of students' arg umentative writing. Is there evidence that teachers encourage and enhance audience awareness through their written feedback on essays? Consistent with the subdivision of metadiscourse markers into textual and interp ersonal categories (Intaraprawat & Steffensen, 1995; Vande Kopple, 1985, 1997), we analysed three Chinese teachers' written feedback according to their comments on textual and interpersonal features of students' writing. There are many ways of grouping the themes that emerged in the teachers' written comments, however in the light of our focus on audience awareness, we deemed it most appropriate t o use these two themes as the framework for the analysis. We acknowledge that wr itten feedback is not the dominant form of feedback for Chinese students owing t o the large class sizes, the limited teacher time available, and the primary con cern of the education system with quantitative feedback. Nevertheless, we asked three Chinese English teachers to read a random selection of 50 essays (which we re devoid of any form of identification) and to provide a score and a written co mment at the end of each essay. Our interest in this paper is with the written c omments only. The results of our analyses are presented in Table 5, with a selection of verbat im comments from the markers included. The findings indicate that teachers' feed back on textual features predominated. Identifying comments on interpersonal fea tures of students' writing proved to be more of a challenge. Yet, despite this c hallenge and the relatively small number of comments in this category, we decide d to include these feedback comments as we believe they raise several pertinent issues. In harmony with VandeKopple (1985, 1997) we defined interpersonal feedba ck as that feedback which acknowledges the role of the writer and his/her abilit y to convey personality to an audience/reader through writing. While the comments on interpersonal features were in the minority, we have inclu ded a range of them in Table 5 to illustrate the types of features to which mark ers responded. Markers typically commented on the quality of the arguments and t he clarity with which the ideas were conveyed. One marker overtly raised the iss ue of the reader/audience, stating that "we can't really see your view". Referen ce to the writer's personality was also made by one marker who commented on the "friendly attitude" of the essay. We would argue that this marker's valuing of a friendly attitude in writing is influenced by the Confucian philosophy that peo ple are "relational being[s], socially situated and defined within an interactiv e context (Bond, 1986, p. 215) and that fostering "smooth interpersonal relation ships" (Salili, 1995, p. 75) tends to be more highly valued than challenging the status quo through the processes of questioning and critical thought that tend to be valued by Western markers of argumentative writing (Ballard & Clanchy, 199 1). Nevertheless, we did find two teacher comments that reflected a valuing of the s pirit of enquiry, open-mindedness and reflective thinking that is arguably at th e core of the Confucian tradition (Biggs, 1991, 1996; Salili, 1995; Wing On, 199 6): You have your own opinion and have presented and proved it well. The approach is new and interesting. While these comments were certainly in the minority and made by only one of the Chinese markers, they nevertheless exhibit values that challenge the generalised view of Asian learners as adopting a "reproductive approach to learning" (Balla

rd & Clanchy, 1991, p. 23) with "no pressure on students to evaluate . . . no re quirement to argue, to resolve ambiguities or dilemmas, to reach clear-cut concl usins" (p. 32). We are encouraged by the limited though significant evidence tha t some teachers do encourage students to question, to present well-substantiated arguments and counter-arguments in an effort to facilitate deep approaches to l earning (Biggs, 1996) and challenge the stereotypical view of "student-as-tape-r ecorder" (p. 47). Table 5 Overview of markers' written feedback in terms of textual and interperso nal features in students' writing Textual features Verbatim written comments Language Rather a lot of Chinglish (Chinese English). Knowledge of language poor. Quite a few grammar mistakes. The language is fluent. Vocabulary You have a large vocabulary. Some words are not used in the correct way. Sentence structure Many awkward sentences. Many good sentences. Some mistakes in sentence construction. Essay structure and coherence This is not really an essay. The arrangement of the essay is sensible. The organisation of the essay is muddled. The materials aren't well organised. Make sure you link the beginning and the en d. Your essay presents three ideas which are different from each other therefore th ere is no clear point. Interpersonal features

Verbatim written comments Expression of ideas and quality of argument Too many indirect ideas. No persuasive arguments. Some unreasonable arguments in the essay. We can't really see your view on the subject. It is full of facts that have noth ing to do with the topic. Some facts don't seem convincing enough. The reasons you have given are convincing but not strong enough. The attitude is friendly. You have your own opinion and have presented and proved it well. The approach is new and interesting. Handwriting Poor handwriting. Review and implications While the markers' comments are limited in scope they nevertheless provide a use ful viewing platform to more comprehensively reveal the types of features in wri tten argument that Chinese markers consider worthy of comment. As demonstrated i n Table 5, feedback on textual and linguistic features tend to dominate. This is not surprising, particularly in the case of L2 learners for whom the convention s of the English language are often particularly challenging. It may be argued t hat all these comments reflect an emphasis on the part of markers on raising stu dents' awareness of the audience for whom they are writing. We found few explici t references to and suggestions regarding how students might enhance the appeal and comprehensibility of the text for the reader/audience. There is scope for te achers to not only teach metadiscourse features explicitly, but also to ensure t hat their feedback on student essays reflects the value of this emphasis on audi ence awareness. Limitations of the Study We recognise several limitations in this study. These include our limited backgr ound information on the students in the sample and our reliance on a single essa y writing task for analysis. Moreover, we have limited the present discussion to analysis of essays based on frequency counts of metalinguistic features. Crowhu rst (1987) cautions against reliance on this measure as a means of determining h ow usage of specified features is related to writing quality. She comments that the extent of utilisation of a particular linguistic device (in her study, cohes ive ties were analysed) does not necessarily equate with writing quality. In ana lysing essays one must pay close attention to the context in which the devices a re used and the level of complexity and maturity with which they are used. This caution is a valid one and it will guide further analysis of the quality of the essays under investigation. An additional limitation is, of course, imposed by t

he fact that we are interpreting Chinese culture and pedagogy through western ey es. We recognise that this limits our capacity for analysis and interpretation o f sociocultural influences on students' writing. Conclusion This study has significant implications for the teaching of Asian/Chinese adoles cents. Our analysis of Chinese students' argumentative essays (written in Englis h) provides evidence of the sociocultural forces at play in their writing. In ma ny instances, students' patriotism and lack of awareness of reader needs - as de monstrated in their use of language and claims - impeded their capacity to const ruct coherent, objective and effectively substantiated written arguments. These and related findings provide justification for a sociocultural consideration of the pedagogical and assessment issues related to the writing processes of second language learners. By examining a single argumentative writing task from a metalinguistic perspecti ve we have provided some evidence for the ways in which adolescent Chinese write rs are influenced by the sociocultural context in which they live and learn. Fir st, we have noted the power of the sociocultural context and the interdependence of Confucian tradition and societal values, and the learner and her/his approac h to the task. We have also identified the interrelationship between the writer' s stage of cognitive and socioemotional development and her/his capacity to cons truct arguments. Within this set of interrelationships, too, we discern the powe r of interconnecting pathways as the adolescent's sense of self and identity as a writer is influenced by the Chinese values of collectivism, filial piety and t he dual self (Salili, 1995). Finally, we have identified some of the difficultie s L2 writers encounter when trying to compose written text within a sociocultura l context. There is a tacit understanding that they will forge new pathways of c ommunication with their reader/audience, yet the challenges of keeping these pat hways open are great. Students have difficulty identifying and addressing the needs of their audience. Our study has shown that the majority of adolescent writers in our sample were deemed to be egocentric, according to the Audience Awareness Scale (Berrill, 199 2). Moreover, their use of metalinguistic features - those features that specifi cally assist the reader to follow the writer's conceptual pathway - is limited i n range and frequency. Our study indicates that those who assist students to dev elop these communication pathways do not overtly draw students' attention to the importance of audience awareness in their written feedback. References

Ballard, B., & Clanchy, J. (1991). Assessment by misconception: cultural influen ces and intellectual traditions. In L. Hamp-Lyons (Ed.), Assessing second langua ge writing in academic contexts. Norwood, NJ: Ablex. Bartholomae, D. (1985). Inventing the university. In M. Rose (Ed.), When a write r can't write (pp. 134-165). New York: Guilford. Barton, E. L. (1995). Contrastive and non-contrastive connectives: Metadiscourse functions in argumentation. Written Communication, 12 (2), 219-239. Berk, L. (2000). Child development (5th ed.). Boston: Allyn and Bacon. Berrill, D. P. (1990). What exposition has to do with argument: Argumentative wr iting of sixteen-year-olds. English in Education, 24 (1), 77-92.

Berrill, D.P. (1992). Issues of audience: egocentrism revisited. In R. Andrews ( Ed.). Rebirth of rhetoric: Essays in language, culture and education (pp. 81-101 ). London: Routledge. Biggs, J. (1991). Approaches to learning in secondary and tertiary students in H ong Kong: Some comparative studies. Education Research Journal, 6, 27 - 39. Biggs, J. (1992). Why and how do Hong Kong students learn? Using the learning an d study process questionnaires. Hong Kong: Faculty of Education, The University of Hong Kong. Biggs, J. (1996). Western misconceptions of the Confucian-heritage learning cult ure. In D. Watkins & J. Biggs (Eds.), The Chinese learner: Cultural, psychologic al and contextual influences (pp. 48 - 68). Hong Kong: CERC and Melbourne: ACER. Bizzell, P. (1986). What happens when basic writers come to college? College Com position and Communication, 13, 294-301. Bond, M. H. (1986). The social psychology of Chinese people. In M. H. Bond (Ed.) , The psychology of the Chinese people (pp. 213 - 264). Hong Kong: Oxford Univer sity Press. Brostoff, A. (1981). Coherence: 'next to' is not 'connected to'. College Composi tion and Communication, 32, 278-293. Chatwin, B. (1987). The songlines. London: Jonathan Cape. Cheng, X., & Steffensen, M. (1996). Metadiscourse: A technique for improving stu dent writing. Research in the Teaching of English, 30 (2), 149-181. Connor, U. (1990). 'Linguistic/ Rhetorical measures for international persuasive student writing', Research in the Teaching of English, 24, 67-87. Crowhurst, M. (1987). 'Cohesion in argument and narration at three grade levels' , Research in the Teaching of English, 21, 185-199. Crowhurst, M. (1990). Teaching and learning the writing of persuasive/argumentat ive discourse. Canadian Journal of Education, 15 (4), 348-359. Erikson, E. H. (1980). Identity and the life cycle (2nd ed.). New York: Norton. Golder, C., & Coirier, P. (1994). Argumentative text writing: Developmental tren ds. Discourse Processes, 18, 187 - 210. Hau, K. T. (1992). Achievement orientation and causal attribution of Chinese stu dents in Hong Kong. Unpublished doctoral dissertation, University of Hong Kong. Hird, B. (1996). EAP: Teaching Chinese learners to be impolite. English in Austr alia, 115, 29-37. Ho, D. Y. F. (1981). Traditional patterns of socialization in Chinese society. A cta Psychologica Taiwanica, 23, 81 - 95. Hsu, F. L. K. (1985). The self in cross-cultural context. In A. J. Marsetta, G. De Vos, & F. L. K. Hsu (Eds.), Culture and the self (pp. 24 - 55). London: Tavst ock. Intaraprawat, P., & Steffenson, M. S. (1995). The use of metadiscourse in good a nd poor ESL essays. Journal of Second Language Writing, 4 (3), 253-273.

Kember, D., & Gow, L. (1991). A challenge to the anecdotal stereotype of the Asi an student. Studies in Higher Education, 16 (2), 117-128. Kirby, J. R., Woodhouse, R. A., & Ma, Y. (1996). Studying in a second language: The experiences of Chinese students. In D. Watkins & J. Biggs (Eds.), The Chines e learner: Cultural, psychological and contextual influences (pp. 141 - 158). Ho ng Kong: CERC and Melbourne: ACER. Krause, K., & O'Brien, D. (1999). A sociolinguistic study of the argumentative w riting of Chinese students. Education Journal, 27(2), 43 - 64. Lapsley, D. K., Milstead, M., Quintana, S., Flannery, D., & Buss, R. (1986). Ado lescent egocentrism and formal operations: Tests of a theoretical assumption. De velopmental Psychology, 22, 800 - 807. Lau, S. (1986). The value orientation and education process of Chinese universit y students in Hong Kong. Educational Journal, 14(2), 7 - 13. Liu, L. M. (1986). Chinese cognition. In M. H. Bond (Ed.), The psychology of Chi nese people (pp. 73- 102). Hong Kong: Oxford University Press. Luria, A. R. (1987). The mind of a mnemonimist. Cambridge, MA: Harvard Universit y Press. Marton, F., Dall'Alba, G., & Lai Kun, T. (1996). Memorizing and understanding: T he keys to the paradox? In D. Watkins & J. Biggs (Eds.), The Chinese learner: Cu ltural, psychological and contextual influences (pp. 69 - 84). Hong Kong: CERC a nd Melbourne: ACER. McLellan, H. (1997). Cyberspace songlines: Building trails to learning communiti es. Retrieved October 20, 2000 from the World Wide Web: http://tech-head.com/son gline.htm Moshman, D. (1999). Adolescent psychological development: Rationality, morality, and identity. Mahwah, NJ: Erlbaum. Piaget, J. (1952). The origins of intelligence in children. New York: Internatio nal University Press. Rodby, J. (1990). The ESL writer and the kaleidoscopic self. The writing instruc tor, 10 (1), 42-50. Rubin, D. L. (1998). Writing for readers: The primacy of audience in composing. In N. Nelson & R. C. Calfee (Eds.), The reading-writing connection, Part II (pp. 53-73). Chicago: The National Society for the Study of Education. Rubin, D. L., Piche, G. L., Michlin, M. L., & Johnson, F. L. (1984). Social cogn itive ability as a predictor of the quality of fourth-graders' written narrative s'. In R. Beach, & L. Bridwell (Eds.), New directions in composition research. N ew York: Guilford Press. Salili, F. (1995). Explaining Chinese students' motivation and achievement: A so cio-cultural analysis. Advances in Motivation and Achievement, 9, 73-118. Salili, F. (1996). Accepting personal responsibility for learning. In D. Watkins & J. Biggs (Eds.), The Chinese learner: Cultural, psychological and contextual influences (pp. 85 - 106). Hong Kong: CERC and Melbourne: ACER. Salili, F., & Ching, S. L. (1992). Motivation and Achievement. Unpublished resea rch report, The University of Hong Kong.

Salili, F., & Ho, W. (1992). Achievement and motivation. Unpublished research re port, The University of Hong Kong. Selman, R. L. (1980). The growth of interpersonal understanding. New York: Acade mic Press. Shen, F. (1989). The classroom and the wider culture: Identity as a key to learn ing English composition. College Composition and Communication, 40, 459 - 465. Steffensen, M. S., & Cheng, X. (1996). Metadiscourse and text pragmatics: How st udents write after learning about metadiscourse. In L. F. Bouton (Ed.), Pragmati cs and language learning. (Monograph Series Vol. 7, pp. 153-170). Stevenson, H. W., & Lee, S. (1990). Context of achievement. Monographs of the So ciety for Research in Child Development, Serial no. 221, Vol. 55, Nos. 1-2. Stigler, J. W., Smith, S., & Mao, L. W. (1985). The self-perception of competenc e by Chinese children. Child Development, 56, 1259 - 1270. Tang, C., & Biggs, J. (1996). How Hong Kong students cope with assessment. In D. Watkins & J. Biggs (Eds.), The Chinese learner: Cultural, psychological and con textual influences (pp. 159 - 182). Hong Kong: CERC and Melbourne: ACER. Tu, W. M. (1979). Humanity and self-cultivation: Essays in Confucian thought. Be rkeley, Calif.: Asian Humanities Press. Vande Kopple, W. (1985). Some exploratory discourse on metadiscourse. College Co mposition and Communication, 36, 82-93. Vande Kopple, W. (1997, March). Refining and applying views of metadiscourse. Pa per presented at the Annual Meeting of the Conference on College Composition and Communcation (48th, Phoenix, AZ). Wilson, R. W., & Pusey, A. W. (1982). Achievement motivation and small business relationship patterns in Chinese society. In S. L. Greenblatt, R. W. Wilson, & A . A. Wilson (Eds.), Social interaction in Chinese society (pp. 195 - 208). Praeg er. Wing On, L. (1996). The cultural context for Chinese learners: Conceptions of le arning in the Confucian tradition. In D. Watkins & J. Biggs (Eds.), The Chinese learner: Cultural, psychological and contextual influences (pp. 25 - 41). Hong K ong: CERC and Melbourne: ACER. Yang, K. S. (1986). Chinese personality and its change. In M. H. Bond (Ed.), The psychology of the Chinese people. Hong Kong: Oxford University Press. Yiri (1997). Retrieved October 20, 2000 from the World Wide Web: http://www.nano u.com.au/songlines/index2.html