Ting, S. H., & Chai, A. S. (2013). Textual and language features of students’ written discussion...

21
Difficulty with discussion texts: Structure or language features? Abstract The study examined whether it was more difficult for students to use the macro-organisational structure or the language features of discussion after genre-based instruction. Data were collected from 100 students enrolled in an English for Academic Purposes (EAP) course at a Malaysian university. They wrote an advantages- disadvantages essay which was analysed using Feez’s (1998) framework. The results showed that about half of the students were unable to state the issue clearly at the beginning and assess the issue to conclude the discussion. For the arguments, the topic sentences and supporting details were satisfactory. The arguments were signposted with sequential, additive and adversative connectors, and causal connectors were important for developing the arguments. Modality was conveyed mainly through “can” and “we” but the potential of conditional clauses to move arguments forward was not realised. The findings indicate that structure deserves more attention than language features in EAP courses. Keywords: English for Academic Purposes, discussion, argumentative, exposition Introduction Research has shown that English as a second language (ESL) learners have some problems writing effective expository texts, which encompass both the argumentative texts requiring a stance and discussion texts requiring a balanced discussion of the issue at hand (Derewianka, 2003). Of the two types of expository texts, more is understood of the ESL learners’ problems with argumentative texts because of more studies focusing on this than on discussion texts. For example, Lee (2005) found that the East- Asian students were not as successful as Australian-born native speakers of English in showing audience awareness through the argument structure and interpersonal components of language 1

Transcript of Ting, S. H., & Chai, A. S. (2013). Textual and language features of students’ written discussion...

Difficulty with discussion texts: Structure or language features?

Abstract

The study examined whether it was more difficult for students to use the macro-organisational structure or the language features of discussion after genre-based instruction. Data were collected from 100 students enrolled in an English for Academic Purposes (EAP) course at a Malaysian university. They wrote an advantages-disadvantages essay which was analysed using Feez’s (1998) framework. The results showed that about half of the students were unable to state the issue clearly at the beginning and assess the issue to conclude the discussion. For the arguments, the topic sentences and supporting details were satisfactory. Thearguments were signposted with sequential, additive and adversative connectors, and causal connectors were important for developing the arguments. Modality was conveyed mainly through “can” and “we” but the potential of conditional clauses to move arguments forward was not realised. The findings indicate that structure deserves more attention than language features in EAP courses.

Keywords: English for Academic Purposes, discussion, argumentative, exposition

Introduction

Research has shown that English as a second language (ESL) learners have some problems writing effective expository texts, which encompass both the argumentative texts requiring a stance and discussion texts requiring a balanced discussion of the issueat hand (Derewianka, 2003). Of the two types of expository texts,more is understood of the ESL learners’ problems with argumentative texts because of more studies focusing on this thanon discussion texts. For example, Lee (2005) found that the East-Asian students were not as successful as Australian-born native speakers of English in showing audience awareness through the argument structure and interpersonal components of language

1

features. Another study pinpointed the problem with the argument structure. From their analysis of the writing of English majors in Argentina and students learning English to undertake post-secondary education or to re-enter a profession in Australia, Jenkins and Pico (2006) found that the difficulty lies in the introductions and conclusions for the weaker writers. Similar findings on the problems with the statement of stance and conclusion were found by Author (year). Malaysian learners with lower English proficiency were either not able to state and restate the stance clearly at the beginning and end of argument texts or they omitted these crucial stages (see also Author, year).

In addition to the structure, studies have also shown that the difficulty in writing effective argumentative essays extend to the use of language features characteristic of expository texts such as connectors and modal verbs. For example, the Malaysian learners in Author (year) study relied on two modals verbs (“can”, “will”) and their English proficiency did not make much difference in either the frequency of use or the choice of modal verbs. However, the denotative meanings of modal verbs posed greater problems for the less proficient learners. Even university academics who are non-native speakers of English have some difficulty handling modality in their research articles – a type of argument text – for hedging and boosting purposes (Guinda, 2003). These studies indicate that while connectors and modal verbs are used in argument texts, the skill in using them to make convincing arguments is often lacking for ESL learners.

The research on argument texts has offered insights into ESL learners’ problem for teaching purposes. However, less is known about the writing of discussion texts by ESL learners. Although argument and discussion texts are both expository writing, the purpose differs in that the argument text uses points to support a stance whereas the discussion text offers a balanced consideration of opposing views on an issue without any necessaryendorsement of a position (Jenkins & Pico, 2006; see also Derewianka, 1991; Feez, 1998). If argument and discussion texts

2

are coalesced and treated as one, this would compromise the effectiveness of the texts in achieving the communicative purpose. While the language features for these two types of expository texts are similar, the organisational structure would differ, particularly in the introduction and conclusion. In view of the lack of research on discussion texts, studies are needed to identify problem areas for more effective teaching.

The study examined whether it was more difficult for students to use the macro-organisational structure or the language features of discussion after they have been taught academic writing using the genre-based approach.

Theoretical framework of study

The genre-based approach in academic writing is based on the functional view of language, and the functional purpose of texts is viewed at the whole text level (Baker, 2010) rather than at the sentence level. The different purposes of texts are reflectedin the variety of genres. Genres are recognised by its schematic structure, language and register which vary with the context of culture and specific contexts of situation (Eggins, 1994). The context of culture focuses on the communicative purpose which shapes the text produced, but the text may vary in textual features depending on the context of situation (field, tenor and mode).

For the text structure, Feez (1998) outlines the following as essential: Statement of Issue, Arguments for and against, and Assessment/Recommendation (p. 99). To deal with issues, the discussion texts draw upon the use of theme and rheme, connectivity, transitivity and modality to achieve the purpose ofthe text (Jenkins & Pico, 2006). The relevant language features are conditional clauses, modal verbs and reader engagement strategies (Feez, 1998). In argument or discussion texts, conjunctions associated with reasoning to express cause-and-effect are more commonly used than sequence connectors (see Derewianka, 1991). The use of modal verbs avoids sweeping

3

generalisations and unqualified claims that would jeopardise the argument (Derewianka, 1991).

The Study

The discussion texts analysed in this study were written by 100 students enrolled in an English for Academic Purposes (EAP) course at a Malaysian university. The students were in their second or third year of their degree and from different disciplines, and they were mostly 21 to 22 years old. These students have spent about 12 years learning English in school before entering university, and their home languages are Malay, Chinese or other indigenous languages spoken in Malaysia.

In the EAP course, the academic genres taught using the genre-based approach were information report, explanation and discussion but there were also modules on reading skills, and citation and referencing. The modules for the academic genres took 12 hours each, spread across three weeks. Vygotsky’s (1978) notion of scaffolding was adopted whereby the instructor built upstudents’ background knowledge on the topic (general knowledge and vocabulary) and genre in the initial field-building stage, followed by a modeling or text deconstruction stage in which the instructor explained the structure and language features of the genre (Appendix 1), and this was followed by joint and finally independent construction of the text.

The discussion texts analysed in this study was written during the final examination, and the topic and the available time was the same. The topic was “advantages and disadvantages of human dependency on computers”. One text was used trial analysis and the other 100 texts were for the actual analysis based on Feez’s (1998) framework of discussion texts. The student discussion texts were analysed for the presence and absence of the following: Statement of Issue, Arguments for and against, and Assessment/Recommendation. If the arguments were repeats of earlier arguments, they were not counted as appropriate argumentsthe second time. Feez (1998) listed the following as language

4

features which are important for achieving the purpose of discussion texts: conjunction, modality, conditional clauses. Forthe connectors, Celce-Murcia and Larsen-Freeman’s (1999) categorisation was used: additive, adversative, causal, sequential. For additive and adversative connectors, only those used in the sentence initial positions were counted but not conjunctions in inter-clausal positions (e.g., “and”, “also”, “but”). These examples show that these conjunctions do not signpost arguments:

“The dependency on computers has both advantages and disadvantages” (Text 47)

“They feel that computer networking, also known as the internet, helps their business grow faster” (Text 12)

“Computers have revolutionized the world, and we have becomedependent on computers – even computers give advantages but we can argue in many ways” (Text 11)

These conjunctions were identified and the frequency of use notedin the preliminary analysis but they were subsequently omitted from the results due to the reason explained above. For the analysis of sequential connectors, care was taken to separate useof “since” for reasoning and for indicating duration (e.g., sincethe creation and revolution of computers). The former was categorised with causal connectors. For the latter, the frequencies were reported in the sequential connector category.

Results and Discussion

In this section, excerpts from student discussion texts are included to illustrate the use and the texts are referred to as Text 1 to Text 100. See Appendix 2 for sample student discussion text, marked for text structure and language features.

Structure of discussion texts

5

In this study, only the presence of stages with appropriate content was considered effective structuring of the discussion text.

The results showed only 47 of 100 students clearly stated the issue for discussion (Table 1). The remaining 53 students either stated a different issue from what was presented in the question (34) or took a stance on the issue (19). If the deviation in content is disregarded, this means that 81 students began their discussion texts with a statement of the issue, showing awarenessof the compulsory nature of this initial stage.

Table 1. Frequency of appropriate stages in discussion texts

Stages of discussion

Frequency

Issue 47Argument 1 –point

98

Elaboration

98

Argument 2 – point

90

Elaboration

92

Argument 3 – point

32

Elaboration

28

Argument 4 – point

29

Elaboration

29

Assessment 40

Next, the arguments were analysed for the relevance of the point and elaboration. More than 90 students constructed the first two arguments appropriately, and these were usually positive points

6

to show human dependency on computers. In contrast, the students found it more difficult to argue that humans have not become dependent on computers in certain aspects of their lives such as in counseling. Less than one-third of the students were able to present the subsequent arguments appropriately (32 for Argument 3and 29 for Argument 4). They strayed into presenting negative effects arising from misuse of computers and the general weaknesses of computers (e.g. virus attack, loss of files and high cost). The problem could be topic-related as students were more familiar with ways how computers have benefitted the world but not the counter-argument. The results show that students havea general understanding of the stage in discussion texts but lackcontent knowledge to write it well.

For the final stage of discussion texts, the assessment of arguments, Table 1 shows that only 40 students wrote appropriate evaluation of the issue under discussion, for example:

In conclusion, after considering the arguments against and arguments for, I suggest that people are not fully dependenton computers. Computers only needed when they have some problems or work to do. But for some reason, people may become dependent on computer because they are too lazy to think and find solutions for their work. So, people must consider the attitude of the people who using computers before make a conclusion or decisions. (Text 64)

This kind of recommendation indicates that this student had a good grasp of the issue discussed. The remaining 60 students wrote conclusions which either deviated from the issue or made a stance that is detached from foregoing arguments (e.g., There areadvantages and disadvantages of human dependency on computers). During the course, students were taught not to take a definite stance like what is done for argument texts as shown in the following example:

Thus, in summary we can see that in terms of production, we depend on computer to increase our production. In addition,

7

we also use computer to surf the internet to get informationand communicate with others. However, not every people in this world can afford to buy a computer. Furthermore, some people do not know how to operate a computer. Therefore after examining all the arguments, the statement ‘computers have revolutionised the world and we have become dependent on computers’, we should learn more about computers so that we can fully maximise the function of computers. (Text 75)

The assessment of issues at the end of the discussion text makes it different from the ending of argument texts which requires a restatement of the stance stated at the beginning of the argument. To sum up, for the students in this study, the most difficult part of the structure are statement of issue in the introduction and the assessment of the arguments in the conclusion – concurring with findings of Jenkins and Pico (2006). The students have a general understanding of how to writetopic sentences to present the main point of the arguments and tosupport these with relevant elaboration but some lacked content knowledge to write persuasive arguments.

Language features of discussion texts

In this section, results are presented for the three language features of discussion texts: conditional clauses, modal verbs and connectors.

Conditional clauses. Out of 100 discussion texts analysed, only 35 students used conditional clauses. “If … then” statements were used by 29 students once, two times by four students, and three times by two students. A common type of conditional sentence is the “if + simple present + will (verb)” to express likelihood andpossibility of the condition to be fulfilled. For instance, “if the computer doesn’t work, the people will be afraid and frustrated” (Text 37). The students did not use past tense to refer to unreal (impossible, hypothetical) conditions, and they also did not use “may” to present hypothetical propositions or “might” to speculate open possibilities. Students did not have to

8

use conditional clauses but this is a technique which opens up possibilities in logical reasoning when facts are not available to back up the arguments. In the context of research articles, if-conditionals are found to have an interpersonal role in the discussion section:

to guide the reader’s interpretation while allowing for a certain degree of independence in reaching the conclusions, to engage the reader by leaving some questions open for further discussion, to negotiate terms and concepts, to wardoff possible criticism, to signal problem areas, to acknowledge other points of view or potential threats to thecogency of argumentation, and to involve the readers by directly soliciting their approval. (Warchal, 2010, p. 140)

Thus, in the context of expository texts (discussion texts included), if-conditionals function as hedges to limit the assertiveness of a claim by making its validity conditional on some other premises and also as emphatics to promote a claim to the status of the obvious once another claim is accepted (Warchal, 2010). However, the potential of if-conditionals in handling claims in arguments was not exploited much when studentswrote the discussion texts.

Modal verbs. The students made ample use of modal verbs in their discussion texts, at an average of 95.3 modal verbs in a text of about 250 words (Table 2). The most frequently used modal verb is“can” (55.8%), followed by “will” (22.8%). In this study, “can” was mainly used with the meaning of ability, for example, “Peoplecan complete their work easily …” (Text 23). This is similar to findings of another study on argument texts, the other type of expository texts which have the same language features as discussion texts. Neff-van Aertselaer and Dafouz-Milne (2008) found that native writers use “can” more to denote definite possibility to present a change from problem to solution whereas non-native writers use “can” in a dynamic sense with variable meanings. The scope of the present study does not allow such a comparison but what is evident is that the students relied on

9

“can” rather than a variety of modal verbs to convey degrees of certainty.

Table 2. Frequency of language features of discussion texts

Modal verbs Frequency PercentageCan 532 55.8Could 33 3.5Shall 1 0.1Should 33 3.5Will 217 22.8Would 11 1.1May 57 6.0Might 30 3.1Must 39 4.1Total 953 100.0

Next, in this study “will” was often used when the present tense is more appropriate. For instance, “without computers, these presenters will not be able to present well” (Text 34). A contextual factor contributing to the frequent use of “will” needs to be brought into focus. In the Malaysian variety of English, the future tense marker is often used in place of the present tense to express a statement of fact (Checketts, 1999; Nair-Venugopal, 2003; Talif & Edwin, 1989). In view of the localised use of English, “will” may not carry meanings of certainty and the high frequency of “will” in comparison to “would” cannot be concluded as a deliberate choice between certainty and possibility.

A closer examination of Table 2 shows that the modal verbs were mainly used for hedging (“can”, “could”, “may”, “might”) rather than boosting (“will”, “should”, “must”). The inclination towardsmodal verbs for mitigation is partly because discussions does notrequire advocacy of action which would be more common in argumentative/persuasive texts. Use of modal verbs for mitigationrather than assertiveness is a characteristic of non-adversarial argumentation according to Helms-Park and Stapleton (2003). The

10

tendency towards mitigation could also be due to cultural influence as Japanese students used downtoning expressions whereas American students use emphatic devices (Kamimura & Oi, 1998) but this is a conjecture which requires further investigation.Table 3. Frequency of connectors used in discussion texts

Category Connectors Frequency

Total

Sequential connectors when 57 195In conclusion 39first(ly) 34secondly 14since 30finally 9In summary 9In a nut shell

3

Causal connectors Because 146 309Thus 48Since (reason)

30

So 35Therefore 26As a result 14Hence 9Consequently 1

Additive connectors Besides (that)

108 217

In addition 46Furthermore 37Moreover 26

Adversative connectors However 76 173On the other hand

54

Although 13Instead 10Even though 8

11

Apart from that

8

In contrast 3Despite 1

Total 894 894

Connectors. The connectors that are used the most are causal connectors (309), and the least used is adversative connectors (173). Altogether 195 sequential and 217 additive connectors wereidentified in the analysis. More causal connectors are expected than sequential connectors because discussion involves reasoning rather than ordering of events as in explanation of phenomenon. An average of two additive and two adversative connectors were used by the students, and this is reasonable because the rubrics required them to write two points for and two points against the issue. While both additive and adversative connectors are needed to signpost advantages and disadvantages of human dependency on computers respectively, the larger number of additive connectors indicates that the positive aspects of human need for computers were elaborated in more detail whereas the arguments for the negative aspects were sketchy. This observation concurs with the results on the structure of discussion text, whereby the studentswere found to write better topic sentences for the pros rather than the cons of the issue. As explained earlier, this is probably due to lack of content knowledge – student familiarity with dependency on computers rather than the converse – rather than lack of ability to organise the discussion text or to use connectors to make the organisation more systematic.

A comparison of the additive and adversative connectors used by students in the discussion texts showed that the range is more extensive for adversative connectors although the frequency is lower than additive connectors. The most frequently used adversative connectors is “however” (76 times), followed by “on the other hand” (54 times). Another six adversative connectors were used to signal contrast in ideas (“although”, “instead”, “even though”, “apart from that”, “in contrast” and “despite”). On the other hand, to signal addition of similar ideas, the

12

students depended on “besides” (108 times). Only three other additive connectors were identified from the analysis, namely, “in addition”, “furthermore” and “moreover”. In their analysis ofargumentative essays written by students in Tucuman and Melbourne, Jenkins and Pico (2006) also found misuse of “moreover” by many Hong Kong students in Australia when “in addition” is more appropriate. Although Jenkins and Pico studied argument texts, the language features are the same as discussion texts as both types are expository texts. The appropriateness of use was not a focus in the quantitative analysis in the present study but this is an aspect worth exploring, particularly in relation to native and non-native speaker conceptions of what is considered appropriate use.

The sequential connectors were mainly used to mark the conclusionof the discussion text and co-occurrence of circumstances. The conclusion was marked using a variety of connectors (“in conclusion”, “finally”, “in summary”, “in a nut shell”), totaling61 (see Table 3). The range is extensive compared to connectors used for signalling their first two arguments (“first(ly)”, and “second(ly)”). These sequence connectors help to structure the discussion text but “when” is used for showing co-occurrence of circumstances. For example, “This can minimise time for work where we do not need to send letter to the post office when we can just e-mail them to the respective persons” (Text 3). In fact, “when” is the most frequently used sequential connector, accounting for 57 out of 195 connectors in this category. When students elaborate the main point to develop their arguments, “when” comes in useful to link clauses.

The causal connectors were used to develop arguments, and the highlight was on the effect rather than the cause. The two causalconnectors that draw attention to the cause is “since” (30 occurrences) and “because” (146 occurrences). In fact, “because”,the most popular causal connector (used 146 times) , usually in the form of “This is because …”. “Because” also introduces the reasons of justification for the proposed idea in across clause complexes in argumentative/persuasive essays (Lee, 2005). The

13

other causal connectors highlight the effect, namely, “thus”, “so”, “therefore”, “as a result”, “hence” and “consequently”, altogether accounting for 133 out of 309 causal connectors identified in the discussion texts. Judging by the range and frequency of causal connectors identified during the analysis, itcan be concluded that the students were versatile in using causalconnectors to develop their arguments when discussing the given issue.

The results on language features of discussion texts showed that students were able to use connectors to structure and develop arguments in their discussion text. Jenkins and Pico (2006) attributed the ease with which students insert connectors in their argumentative writing to the discrete nature of connectors.Other than connectors, the students in this study also inserted many modal verbs, particularly “can” in their discussion text. Itis only conditional clauses which are not used by many students to use hypothetical situations for reasoning. From the aspect of language features, the results suggest that the students do not have much difficulty with the lexicogrammar of discussion texts. Inadequacies in argument are due to the lack of content knowledge(people not relying on computer), rather than a lack of linguistic knowledge.

Conclusion

The study showed that the structure of the discussion text is more difficult to produce than the language features of discussion texts. The student discussion texts meet genre conventions more in language features than in structure. For the structure, about half of the university students in this study were not able to clearly state the issue for discussion in the introduction and to close the discussion with an assessment of the viewpoints on the issue but almost all of them showed the ability to write arguments with main point and elaboration. For the language features, the students frequently used modal verbs and connectors in the discussion texts and the only problem was the lack of variety. The most frequently used connectors were

14

“because”, “besides”, “however” and “when” whereas the two main modal verbs used in the discussion texts were “can” and “will”. The students seemed to handle positive viewpoints on the issue (human dependency on computers) better than negative viewpoints in both the wording of the main point and the use of adversative connectors but this is due to inadequate content knowledge ratherthan linguistic knowledge.

Content is generally understood to be beyond the scope of language instruction but the genre based instruction starts with field building with builds up students’ knowledge of the topic and vocabulary for talking or writing about the topic. The study showed that although the students had been taught academic writing using the genre-based approach and sensitised to the importance of organisational structure and communicative purpose,about half of the group was still not able to state the stance inthe introduction and assess the issues to conclude the discussion. The genre-specific nature of the introduction and conclusion is a break away from the tripartite structure of essays. The general introduction-body-conclusion structure is often taught in school (see Cahill, 2003; Gautreau et al., 1986; Liu, 2005). Having been accustomed to the general introduction and conclusion, it was not easy for the students to meet the structural conventions of discussion texts although learning the language features of discussion textsdid not pose a problem. Yet it is the structure that students need to master because it compromises the effectiveness of the text in achieving the purpose if it is poorly organised and do not meet general expectations of the discourse community. “The reason students’ writing often goes wrong is not because of surface errors such as spelling or the inappropriate choice of vocabulary, but because they are not abiding by the conventions of the genre in some way” (Sidaway, 2006, p. 25). “The global level of argument structure has much more effect on an audience impression of the quality of the argument (Connor, 1987, 1988, 1990, 1991; Connor & Lauer, 1988; Stuart-Smith, 1998)” than grammatical accuracy (as cited in Lee, 2006, p. 8). The non-

15

compliance in text structure has more impact on the perceived quality of academic texts than surface errors in language.

The study has provided evidence-based support on the need to pay greater attention to teaching structure rather than language features when teaching discussion texts in particular and academic writing in general. Nevertheless, we cannot make claims on whether it is easier for students to learn the structure or the language features when taught using the genre-based approach because the research design was not an experimental study. What we assessed was the students’ ability to write discussion texts that meet conventions at the exit point of the EAP course. Our contention is that while students can learn to insert modal verbsand connectors in their academic text, it is not as easy for themto learn how to present and clinch the purpose of their text. Further research needs to be conducted on the ease of learning the structural and linguistic features of academic texts.

References

Baker, T. (2010). Genre matters in academic writing. InternationalJournal of Education and Development, 28. Retrieved September 15, 2010from http://ihjournal.com/genre-matters-in-academic-writingCahill, D. (2003). The myth of the “turn” in contrastiverhetoric. Written Communication, 20(2), 170-194.Celce-Murcia, M., & Larsen-Freeman, D. (1999). Grammar book.

Boston: Heinle & Heinle. Checketts, S. (1999). On the tip of my tongue: Insights into somethings English. Subang Jaya, Malaysia: Pelanduk.Derewianka, B. (1991). Exploring how texts work. Newtown, Australia:Primary English Teaching Association.Derewianka, B. (2003). Trends and issues in genre-basedapproaches. RELC Journal, 34(2), 133. Retrieved August 22, 2011 fromhttp://rel.sagepub.com/cgi/content/abstract/34/2/133.Eggins, S. (1994). An introduction to systemic functional linguistics. London:Pinter.

16

Feez, S. (1998).Text-based syllabus design (Sydney: MacquarieUniversity).Gautreau, A. L., Averett, J. B., Trusock, D., McFerren, S. E., Phipps, C., Lynch. M. … Speer, L. W. (1986). Our readers write: What's a practical idea for teaching organization in writing? The English Journal, 75(5), 61-68. Retrieved October 11, 2012, from http://www.jstor.org/stable/pdfplus/818208.pdf?acceptTC=trueHelms-Park, R., & Stapleton, P. (2003). Questioning theimportance of individualized voice in undergraduate L2argumentative writing: An empirical study with pedagogicalimplications. Journal of Second Language Writing, 12, 245-265.Jenkins, H. H., & Pico, M. L. (2006). SFL and argumentativeessays in ESOL. Proceedings of 33rd International SystemicFunctional Congress, pp. 155-170.Kamimura, T., & Oi, K. (1998). Argumentative strategies in American and Japanese English. World Englishes, 17(3), 307–323.Lee, S. H. (2005). The use of interactional resources inargumentative/persuasive essays (APEs) by tertiary ESL students:A critical perspective. In T. Le & M. Short (Eds.), Proceedings ofthe International Conference on Critical Discourse Analysis: Theory into research (pp.379-403). Launceston, Tasmania, Australia: University ofTasmania.Lee, S. H. (2006). The use of interpersonal resources in argumentative/persuasive essays by East-Asian ESL and Australian tertiary students. Unpublished PhD Thesis, University of Sydney.Liu, L. (2005). Rhetorical education through writing instructionacross cultures: A comparative analysis of select onlineinstructional materials on argumentative writing. Journal of SecondLanguage Writing, 14(1), 1-18.Nair-Venugopal, S. (2003). Malaysian English, normativity, andworkplace interactions. World Englishes, 22(1), 15-19.Neff-van Aertselaer, J.A., & Dafouz-Milne, E. (2008).Argumentation patterns in different languages: An analysis ofmetadiscourse markers in English and Spanish texts. In M. Pütz &J. A. Neff-van Aertselaer (Eds.), Developing contrastive pragmatics:

17

Interlanguage and cross-cultural perspectives (pp. 87-102). Berlin: Moutonde Gruyter.Sidaway, R. (2006). The genre-based approach to teaching writing.English (Spring), 24-27.Talif, R., & Edwin, M. (1989). Error analysis of Form FourEnglish compositions. The English Teacher Vol XVIII. RetrievedOctober 5, 2012, from http://www.melta.org.my/ET/1989/main9.htmlVygotsky, L. (1978). Mind in society: The development of higher psychologicalprocesses. Cambridge, MA: Harvard University Press.Warchal, K. (2010). Moulding interpersonal relations throughconditional clauses: Consensus- building strategies in writtenacademic discourse. Journal of English for Academic Purposes, 9(2), 140-150.Author 1Author 2

18

Appendix 1: Sample of deconstructed discussion text

IntroductionBackground Info

Statement ofIssue

(italicised)

Preview(underlined)

The Pros and Cons of Human Cloning

When Dolly, the first cloned sheep became news,cloning interested the masses. Not only didresearchers investigate deeper into the subject buteven the common people showed great interest inknowing all about how cloning had been done. Most of uswant to know the pros and cons of cloning, its advantages to societyand its potential risks to mankind. While cloning can solve infertility problems and enhance genetic studies, italso causes controversies related to human rightsand religion.

BodyArgument For No.

1

Argument For No.2

Argument Against No. 1

ArgumentAgainst

No. 2

19

The first advantage of human cloning is that itprovides a solution to infertility. Cloning has thepotential of serving as an option for producingchildren especially for couples who cannot have achild. Cloning may make it possible to reproduce acertain trait in human beings. Therefore, cloningenables childless couples to have children with someof their traits.

Furthermore, the second advantage of human cloningtechnologies is its usefulness to researchers ingenetics. They might be able to understand thecomposition of genes and the effects of geneticconstituents on human traits in a better manner.Thus, it simplifies their analysis of genes. Thecloning technology allows researchers to eventuallycombat a wide range of genetic diseases.

However, one disadvantage of human cloning is thatit puts human rights at stake. It remains unclear asto whether cloning may fit into our ethical andmoral principles. Cloning leave man just anotherman-made being. It can devalue human kind and humanbeings might end up being a product as human beingswith better traits would be sought after asmaterials for cloning whereas those with poor traitswould be side-lined.

Conclusion

Assessment ofissue

Language features: Modal verbs (bolded); connectors (boxed); conditional clauses (capitalised and bolded)

Appendix 2: Sample student discussion text

In discussing the impacts of computers, it is necessary to examine a number of arguments for andagainst before making a decision. Since the innovation of computers few decades ago, history of human being has been turned into new era which is era of technology. Gradually, computers become vital in the life of human being.

First of all, computers are utilized in most of the daily matters in human life. Computers are thenecessary tools in the life of human whereby humanuse it for various purposes such as communication,finding information, entertainment and working as well. Such functions bring by the computers make it widely used in the life of human. After all, multifunctions of computers have lead human dependmore on the computers.

The conveniences bring by the computers enhanced the degree of dependent of human on the computers.The conveniences bring by the computers include

20

Issue is stated but not accurate– human dependency on computers is notmentioned.No preview of arguments.

Argument For No.1 – main point and elaboration are satisfactory

Argument For No.2 – main point and elaboration are satisfactory

portable computers, efficiency, accuracy of works and accessibility to the internet. For example, people engaged in the accounting field used computers to calculate and manage their work by easy and fast. In addition, utilized of computers can greatly decreased the percentage of making mistake in work. Therefore, human become more depend on the computers.

On the other hand, human can be more independent and thrived to enhance the function of computers to achieve better living quality. To improve living standards, innovation of computers encouraged more thriving human to continue improveand increase the variety of benefits generated from the computers. Therefore, human will be more independent and put effort to innovate to make sure computers continue contribute to the development of human.

Lastly, human life is still well-rounded even without the computers. There are a lot of things which can be done without using computers. Human can cultivate hobbies, visit relatives and foods. For example, human can involve in outdoor activities such as badminton, football and basketball to spend their time. It is healthy to the life of human rather than sitting in front of the computers.

As a conclusion, computer may help human in different ways but somehow we must control it and not become over dependent on computers.

21

Argument AgainstNo. 1 – does notaddress issue

Argument AgainstNo. 2 – main point and elaboration are satisfactory