A Functional Grammar Approach to Analysing Asian Students’ Writing

Dr Angelia Lu

  Open Access OPEN ACCESS  Peer Reviewed PEER-REVIEWED

A Functional Grammar Approach to Analysing Asian Students’ Writing

Dr Angelia Lu

National Institute of Education- Nanyang Technological University, Singapore


This paper compares aspects of Singaporean and PRC students’ writing at the university level while enrolled in the freshman composition courses. These aspects involve instances of nominalization, theme, and use of modals (or their inappropriate usage) at various parts of the essays submitted by these students. These items are measured and reported in percentages as to gain greater insight to the depth of similarities and differences between the two sets of writing. In addition, the author is concerned that idiosyncractic usage of such anomaly in writing impedes comprehension when reading and is hardly audience-centred. It is noteworthy that the writing of Singaporean and Chinese students do produce various interesting similarities and differences, possibly attributable to their sociolinguistic and educational backgrounds. The implications of the results are analysed and discussed, with possible pedagogical solutions suggested in order to overcome difficulties of teaching separate students with different linguistic abilities.

Cite this article:

  • Lu, Dr Angelia. "A Functional Grammar Approach to Analysing Asian Students’ Writing." American Journal of Educational Research 1.2 (2013): 49-57.
  • Lu, D. A. (2013). A Functional Grammar Approach to Analysing Asian Students’ Writing. American Journal of Educational Research, 1(2), 49-57.
  • Lu, Dr Angelia. "A Functional Grammar Approach to Analysing Asian Students’ Writing." American Journal of Educational Research 1, no. 2 (2013): 49-57.

Import into BibTeX Import into EndNote Import into RefMan Import into RefWorks

1. Introduction

There are many reasons as to why students of ESL (English as a Second Language) students and EFL (English as a Foreign Language) find it difficult to write expository essays. This paper reflects such difficulty in writing by three groups of learners from a pool of matriculated Singaporeans and PRC students. The first group of students are Singaporean students who emerge from an English-speaking background. This means that these students communicate primarily in English since childhood. The second group of students are also Singaporean, but emerge from a Chinese-speaking background. They can be considered to be of ESL background because Mandarin is their predominant language despite having learnt English since childhood. The third group consist of PRC students, who are considered EFL because they interact purely in Mandarin in all social situations except in the classroom condition. Although all three groups of students have had vast amounts of practice in writing expositions at tertiary level, many of them do not attempt expository writing before entering junior college. In fact, secondary schools and junior colleges differ in their focus in the provision of English Language teaching to students. The former provide training in English at a fundamental level, whereby several main genres of writing are presented: narrative, descriptive, expository essays, reports and accounts. Unfortunately, teachers often emphasise narrative and descriptive essays at secondary school level, preferring them to the teaching of expository or argumentative essays and perceiving these genres as being “too difficult” for the students to handle under exam conditions. The Singaporean students are not the only ones who face this problem of coping with expository writing at tertiary level. The author’s experience of teaching PRC tertiary students leads her to the conclusion that these students are often liable to write emotionally charged essays with a narrative generic structure, whenever exposition is required of them. Furthermore, it must be noted that PRC students have little or no experience in writing expository essays, unlike Singaporean students. According to feedback received from PRC students, most of their English assignments included the writing of announcements, letters and advertisements. It is therefore, hardly surprising that numerous PRC students produce “expositions” which are emotionally charged and manifest a high level of interaction with the reader, by personalising their essays with dialogue. The research questions are:

• To investigate the extent to which the lexical density and nominalization of the three groups of essays differs. Lexical Density informs us of the way in which students incorporate words into their writing, and the degree of nominalization indicates to what extent a piece of writing has characteristics of a formal genre.

• To examine the various different types of themes in the students’ essays. The salience of a particular theme reflects an inclination to write sentences which are too long, improperly connected, or have an emotional quality.

• To determine if a relationship exists between the ability to write with a hedging quality (epistemic writing) and the linguistic/educational background from which the students emerge. This study examines the extent to which such a relationship exists and makes a comparison amongst the three groups of students.

2. Literature Review

As in [1] Introduction to Functional Grammar added to meaning mapping, and enhanced what he elaborated on the meaning potential of a clause (Towards a Sociological Semantics) in the 1970s. His widely known text focuses on the functional aspects of grammar and is designed to account for how language is used. The “fundamental components of meaning in language are functional components” as in [1] and in [2]. According to him, all languages are organised around “ideational”/reflective meaning, and “interpersonal”/active meaning. Combined with these is a third metafunctional component known as the “textual”, which “breathes relevance into the other two” in [1]. The rationale for analysing expository discourse based on the ideational, textual and interpersonal metafunctions is thus based on the systemic functional concepts introduced in this paper.

Transitivity is understood as a grammatical system of process types by which we manage and construe the world of experience in [2]. The basic way of representing the patterns of our experience emerge as a result of the three fundamental processes of doing, sensing and being. “In lexicogrammar, the clause is the central processing unit, i.e., it is the clause that meanings of different kinds are mapped into an integrated grammatical structure. These meanings are embedded in clauses and through language, the construing of human experience into such meanings leads to the creation of metafunction, whereby we make sense of our experience by categorising them” in [2]. This is why it is interesting to explore the percentage change of process types from draft to draft, particularly if the sum of meanings of clauses leads to a context in [3]. An example of enriching research carried out was by [4] who found that processes in expositions were less tangible or concrete than those in narratives. Following this strand of research, the author hopes to uncover the ideational aspects concerning clause type changes in drafting.

A model such as the SFG considers not only the “flow” or “grammatical formedness” of a text, but takes into consideration the circumstances under which the essays are written, as well as the interpersonal and experiential aspects of writing and interacting with the reader. It is precisely for this reason that I have selected the SFG framework with its exploration on lexical density and grammatical intricacy, so that many exciting and interesting aspects of writing might be unravelled in the process of analysis. These aspects directly address the question of audience awareness, interaction with the reader and extent of emotional engagement of the essays, issues in writing research which have become more prominent since the nineties.

Hedging is the ability to indicate an unwillingness to make an explicit and complete commitment to the truth of propositions in [5]. In everyday conversation, one hedges by means of modal verbs such as may, or might, as well as word forms such as “perhaps”, “I guess”, and “maybe”. Analysis of formal written corpora have revealed that hedging exists to a large extent in textbooks, abstracts and written discourse. Hedges are often seen as an important means for expressing uncertainty, scepticism and displaying an open mind about one’s own proposition in [5]. In this paper, the extent of hedging is compared across the three groups of students from different English backgrounds, and interesting results are reported with respect to their use of hedges or the lack of them as realised by the presence or absence of mood adjuncts such as “completely believe”.

There are very few comparative papers carried out in Singapore that examine linguistic features of essays to identify differences in linguistic features and text quality. One of such papers is by [6] who did it in the local context and explored linguistic features and text quality between essays written under examination and non-examination conditions. Another one was [7] which compared and discussed interview data pertaining to teachers’ and students’ expectations to academic writing. Much of the works done in Singapore on comparing groups of academic writing is not based on systemic functional grammar; or not comparative in nature between Singaporean and PRC students’ English, and hence this research is rather unique in nature.

3. Participants and Research Design

The subjects who participated in this study were first year Singaporean undergraduates and pre-matriculated PRC students at the National University of Singapore. The Singaporean participants were first-year undergraduates who required extra lessons in English Proficiency, and enrolled in the proficiency course for Science students. The PRC participants, on the other hand, had been matriculated in a Chinese University as first-year undergraduates before coming to Singapore. They were enrolled in mandatory Intensive English Courses offered by the Centre to prepare for entrance to the local university.

The research design involves three contrastive groups of students, namely:

SE (Singaporean students from an English Speaking Background)

SC (Singaporean students from a Chinese-Speaking Background)

PRC (Students from the People’s Republic of China).

These students were formed in these three groups, thus forming a non-equivalent groups design. The reason for grouping the students in the manner was to acknowledge the existence of these three dominant groups of Singaporeans and PRCs which formed the classes taking proficiency lessons. There was a need to distinguish Singaporeans from English speaking backgrounds from those who were from Chinese speaking backgrounds as putting them in the same group would cause results to be severely mixed and inaccurate: it would be harder to draw out patterns and figure out conclusions due to the various language exposure that these students faced. Calculations were conducted in terms of percentages, mostly descriptive statistics in nature for this paper, for it was deduced that the number of students involved may not warrant inferential statistics. Hence, the data is clearly presented in terms of percentages in a manner that reflects clearly and immediately topical Theme patterns and other tested writing features that were inherent in the writing of the students.

In this paper, the terms SE, SC and PRC students will be used in reference to the students from the three different categories as stated above. The terms SE, SC and PRC essays will be used with reference to the essays produced by the respective students.

13 of the Singaporean students from the English – Speaking (SE) background participated in the study. Another 17 Singaporean students indicated that they emerged from a Chinese-Speaking (SC) background. Therefore, there were altogether 30 Singaporean students who participated in this study. All the students were of age eighteen to twenty.

All the PRC students indicated that they came from a Chinese-Speaking background. Many of them came from different provinces or cities of China; for instance Shandong province, or Jilin and Beijing. There were altogether 30 PRC students participating in this study. Refer to Table 1.1 below for the distribution of subjects with reference to their linguistic background. All the students were of age eighteen to twenty.

Table 1.1. Distribution of subjects based on their linguistic background

The researcher requested that the co-ordinators of the English Language Centre at the National University of Singapore for sixty essays of an expository nature, the title of which was “A woman’s place is in the home”. The reason for choosing such a topic is that despite recent emancipation of women, the salary scales of women in certain Asian countries are still lagging behind those of men. Moreover, in parts of China, it is a fact that many rural girls are still being denied of privileges where education and career are concerned. Students find such a topic writable if they are especially relevant to the country they are residing.

As mentioned in the previous section, 30 essays were written by Singaporean students and 30 essays were written by PRC students. This was carried out in a classroom setting under the time constraint of one hour. The students were informed by the co-ordinators that their names would not be revealed in this study, thus assuring them of participant anonymity.

The essays were analysed mainly using the framework of Systemic Functional Grammar. In this paper, three main issues pertaining to examining the essays would be discussed in detail. The theoretical framework employed in this study is systemic functional grammar. This includes the analysis of (1) lexical density and nominalization, (2) realisation of theme, (3) the ability to write epistemically. One main hypothesis of this study was that essays written by Singaporean students would generally be found to satisfy the expectations of the required expository genre more adequately compared with those written by PRC students. These expectations include the manifestation of a high level of rhetorical competence, epistemic quality and a higher incidence of nominalization and lexical density in the essays, amongst other factors. Another main hypothesis of this study was that Singaporean students from an English-Speaking Background (SE students) generally produce expository essays of a better quality compared to Singaporean students from a Chinese-speaking background (SC students).

3.1. Method of Analysis

Most of the analyses involved a simple calculation of the proportion of an entity in relation to the total number of clauses. The independent variables in this study include lexical density, nominalization, types of themes and types of modals. The dependant variable, on the other hand, refers to the percentage of the independent variable that is analysed. For instance, in order to find the percentage of nominalization in an essay, the number of nominalised clauses in one essay was expressed as a percentage of the total number of clauses.


The average percentage use of nominalization in one category of students, e.g., PRCs, was the sum of all the percentages of nominalised clauses in PRC essays, divided by the total number of PRC essays, i.e., 30.


However, the method of calculating modality and themes was different from the above method. For instance, a positive modal operator with a high degree of probability such as certainly was expressed as a percentage of the total number of modal operators in the essay.

The average percentage of positive modal operators for each category of students was calculated in the same manner as stated above in (b).

This method of calculation permitted a quantification of linguistic items among categories such as modality

3.2. Methodological Concerns

The sample size for each group could undoubtedly have been larger. However, the research was limited by the fact that only sixty essays could be secured from the Singaporean and PRC groups of students, as a larger data size could have affected the teaching curriculum of the lecturers involved. The procuring of exactly 15 SE essays and 15 SC essays was another minor impediment, as only 13 SE essays and 17 SC essays could be acquired. However, despite these limitations, a proper evaluation of the available data could probably be carried out due to the fact that qualitative analyses and interpretations could be gleaned from this data size. The results should be relatively stable because a representative sample size should ideally constitute at least 10% of the population size. The total number of Singaporean students who enrol in the proficiency course, as well as the PRC students who enrolled in the English course, does not exceed 600.

4. Findings

4.1. Lexical Density

Lexical density (LD) is defined as the number of lexical items/content words per clause in [1]. The examples below illustrates the calculation of lexical density. The symbol / is used to indicate clause boundaries.

Formula (I)

Extract from PRC1:

1(i) /So we can draw a conclusion/. /“A woman’s place is in the home.”/

/This statement is completely wrong, especially in today/. LD = 11/ 3 = 3.67

In the above example (1), there are 11 lexical items (in bold) distributed among 3 clauses, thus yielding a lexical density of 3.67.

Extract from SE2

1(ii) /Despite the changes in the society now, the controversy about the position of women exists/. /Probably this controversy is less obvious in Singapore/, /as manpower is vital to her development/. LD = 15/3 = 5.00

In the above example 1 (ii), there are 15 lexical items (bolded) distributed among 3 clauses, thus yielding a greater lexical density than the clause in Example (1). True to expectations, SE essays had the highest average lexical density (LD = 5.47), followed by SC essays (LD = 4.73), and then PRC essays (LD = 3.61). This suggests that the essays of the English-speaking Singaporeans contained the highest, and those of the PRC students the lowest, number of content words per clause.

4.2. Nominalization

Nominalization is a structural feature whereby any element or group of elements is made to function as a nominal group in the clause in [1]. The following example illustrates the process of nominalization, whereby verbs encode actions, nouns encode things and causal conjunctions encode logical relations.

Extract from SE2

(2): Although society has changed for quite some time, people are still debating about the position of women. (LD = 8/2 = 4)

As lexical density is proportional to the amount of nominalization that occurs in a text, results emerged as expected. The employment of nominalization, which is a measure of the amount of nominalization that occurs, was highest in the essays of English-speaking Singaporeans (45.96%), followed by those written by Chinese-speaking Singaporeans (39.14%), followed by those written by PRC students (3.12%). However, although there might be a gradation in the results following the trend in lexical density, a precipitous decline was noted in the average utilisation of nominalization by the PRC students. According to [8], there is a propensity for “written language to display a high degree of nominalization”; and indeed, this is its most distinctive characteristic. Once again, the sharp contrast in the employment of nominalization contributed support to the hypothesis that the writing of PRC students would be the most speech-like among the three categories, and that the writing of the SE students would be the most characteristic of the written mode.

4.2.1. Realization of Theme

Topical Theme Among the students’ essays, there were two instances whereby the topical Theme appeared to be absent. The topical Theme was either ellipted or could be considered improperly produced in the clause:

(i) Ellipted Topical Theme (Extract from SC1)

(3) They believed that women should not do rough work like ploughing,

*but should stay at home to look after children.

(ii) Inability to nominalise (Extract from SC3)

(4) *Moreover, keep doing housework is also a boring task.

The percentage use of proper topical Themes was over ninety percent for all three groups of essays by the students. SE essays showed a slightly higher percentage use (94.19%) of proper topical Themes than the SC (90.58%) and PRC essays (91.77%). This means that there was a very slightly higher propensity for SC and PRC students to form either ellipted or improper topical Themes. An implication of this, therefore, is that the start of sentences with topical Themes does not look academic enough, or is too simplistic at the level of academic writing. A constant appearance of improper topical Themes causes the essay to look more narrative than truly expository in nature.

4.2.2.Textual Theme

In this study, the textual theme takes into sole consideration of conjunctive Adjuncts such as and, but, therefore, rather than continuity Adjuncts such as yes, no. The reason is that conjunctive Adjuncts are much more likely to be employed than continuity Adjuncts in written discourse; in this case, an expository essay.

The percentage use of textual Theme was more varied across the three categories of students. PRC students employed the least amount of textual Theme (37.89%), followed by the SE students (41.2%), followed by SC students (50.6%). This means that there was a lesser tendency on the part of PRC students to textualise the Theme of a clause; i.e., their employment of conjunctive relations was not as frequent as that of Singaporean students. The lower usage of textual Themes implied less cohesion in PRC essays. The example below illustrates the example of a textual theme employed by an SE student.

Despite the greater use of textual Themes by SE and SC students than PRC students, it should perhaps be pointed out that both groups of Singaporean students were inclined to conjoin sentences in an inappropriate manner. However, the SC students might be slightly more inclined to do so than the SE students. Both groups of students might have harboured the delusion that the lengthier the sentence, the more complex and profound it would appear and the greater its credibility. Such a writing style is known as juxtaposition consolidation, which has been referred to as:

“the expedient use by some inexperienced writers of ignoring coordination or subordination as ways to link ideas in favour of simply placing ideas next to one another. One form of juxtaposition takes the form of identifying the topic, then following it with a sentence of comment or explanation not grammatically connected to it.”

In [9]

The awkwardness engendered by long sentences in SC students’ essays disrupts the flow of reading and hence affects the coherence of the essays. Such a writing style might stem from general insecurity about one’s writing [8]. The following examples provide illustrations of juxtaposition consolidation:

A Inappropriate employment of Type II Connective so that it is used to conjoin sentences rather than clauses: Extract from SE3

(6) No doubt, women’s are indeed a better choice at doing household chores and bringing up children, however, in today’s highly competitive world, women are also needed in the society.

Note: In place of a comma, a semi-colon or a full-stop should be used before “however”.

Link of ideas without proper grammatical connection: Extract from SC10

(7) Not only today, but even several years ago, when society was dominated mostly by men, there were some women in different parts of the world who came in front as leaders, struggled day and night to prove themselves and show the world that they were capable of doing more than just running a family.

With regards to this issue, lecturers seemed to differ in their opinion. One lecturer commented that the SC students had confessed to punctuation problems, i.e., confusion between the use of comma and semicolon, while another suggested that SC students perhaps lacked the knowledge of the appropriate moment to complete a sentence.

4.2.3. Interpersonal Theme

When a constituent to which one assigns a mood label appears at the beginning of a clause, it is regarded as an interpersonal Theme. The mood categories considered in this study are: the (I) Unfused Finite in interrogative structures, (II) mood, (III) comment, (IV) vocative and (V) polarity Adjuncts. (These four Adjuncts are considered modal Adjuncts). In addition, Wh-words at the beginning of an interrogative are also considered as interpersonal Theme by [9].

The percentage use of interpersonal theme is not very high for students across the three categories. However, there is a slightly higher tendency for interpersonal theme to be used among the SE students (13.50%) than the SC (8.37 % ) or PRC students (6.70%). Instances of interpersonal theme are illustrated below:

8 (I) Mood Adjuncts: Extract from SE2:

8 (II) Comment Adjuncts: Extract from SE5

8 (III) Wh-Words in Interrogatives: Extract from SC1

4.3. Epistemic Writing

“Epistemic writing is the ability to write statements with an appropriate degree of doubt or certainty. Such epistemic comments are important to written academic work where authors have to differentiate opinion from fact and evaluate their assertions in persuasive and acceptable ways” [12] In this paper, an analysis of modality and modulation was performed to ascertain the level of epistemic writing.

4.4. Modalization

Mood Adjuncts include words such as Probably, I believe, I think. Modal Operators, on the other hand, include words expressing degrees of probability such as might. Both sub-categories of simple modalization express judgment implicitly [13]. Below is a list of mood Adjuncts and modal operators classified into degrees of high, median, and low probability.

In this paper, the use of the above Modal Adjuncts is referred to as simple modalization, as exemplified below:

Extract from SE8

(9) Therefore, a woman’s place is not always in the home, but outside the home.

Simple modalization: High degree of probability, negated.

SC and PRC students employed modals with the highest degree of probability most frequently. These included modals such as certainly and always. The proportionate utilisation of modals with high degrees of probability was highest in PRC essays (2.32 %) and lowest in SE essays (0.11 %). It is worth noting that SE students hardly used simple modalization compared to the other two groups of students, especially with regard to high modalization (0.11%). In contrast, SC and PRC students employed median and low degrees of probability with more frequency than SE students. The results suggest a reluctance on the part of SE students to utilise modals in general.

Where complex modalization is concerned, SC and PRC students were more predisposed to employ a median degree of obligation I think (1.79 % and 2.11% respectively) than I believe (0.82% and 1.22% respectively). In addition, “I think” was more commonly employed by both SC and PRC students, compared to SE students. Unfortunately, the employment of complex modalization such as I think is generally frowned upon in written work, even though it has the quality of being epistemic in [12]. Refer to Table (A) in the Appendix for a summary on the results of modalization in the students’ essays.

4.5. Simple Modulation

Simple modulation involves the use of finite verbal operators of obligation such as when the modals must, should, and will are employed. Below is a table which displays the different degrees of obligation expressed by simple modulation.

The use of the above modals is referred to as simple modulation as exemplified below:

Extract from P29: (10) They mustn’t have a lot of words to say.

Simple modulation: High degree of obligation, negated

Extract from P2:

(11) The women who only stay at home should come out to have a new life.

Simple modulation: Median degree of obligation, positive.

Modals of modulation (i.e., will, should, might) were the most frequently used by all three groups of students: an overall percentage of 60.42 was contributed by such modals in the students’ essays. The analysis of modulation revealed that all three categories of students favoured median degrees of modulation. The SE, SC and PRC students were all equally likely to utilise modals such as should, would and could. (5.58%, 15.90%, and 16.30% respectively). This implies that a tendency to use such modals of median degrees of modulation is still rather small (i.e., less than 20 %).

On the other hand, the modals with a low degree of obligation (may) were hardly employed by any of the three groups of students, while only SC and PRC students used the modals of high degree of obligation (must, will). Refer to Table (B) in the Appendix for a summary on the results of modalization in the students’ essays.

In addition to the fact that essays of the PRC students were inundated with modality and modulation to a higher degree than were those of the SE students in many areas, the PRC students’ lack of epistemic writing was also reflected in their choice of expression. They utilised words such as completely, absolutely conveying a high degree of certainty and directness in what they wanted to express. The following exemplify the frequent occurrence of such statements in the essays:

(12) Extracts from PRC essays

“A woman’s place is in the home.”

This sentence is completely untrue/This is absolute nonsense. /You are totally wrong.

Moreover, an average of 50 % of pronouns employed in PRC students’ essays was devoted to the first person pronouns I and We, and to the second-person pronoun You. This contrasts sharply with SC essays where only 27% of pronouns were devoted to first and second person. The SE essays contained the lowest occurrence of first and second person pronouns which comprised only 16.1% of the pronoun distribution.

First and second person pronouns are often discouraged in written academic work, although [14] did point out that they are not invariably characteristic of the spoken mode. In fact, these pronouns were reported to contribute to a high frequency of occurrence in informal letter writing. However, it is generally admitted that employment of these pronouns I, we, you confers on written discourse a quality of dialogue, thereby personalizing the discourse. Such a quality might be appropriate for some forms of written discourse, e.g., a letter, but would be inappropriate in a context where the discourse must be de-personalized, e.g., writing in an academic context, which is addressed to the impersonal other.

5. Discussion

The findings of this study can be summarized using the various aspects of context of situation, namely mode, field and tenor. In the case of field, there was hardly any distinction amongst the groups. One experiential circumstance was the fact that all the students contributed a piece of written work based on the topic “A woman’s place is in the home” under the time constraint of one hour, in the setting of a classroom. The fact that the event type was controlled for (given that the topic was the same for all three groups) ensured that the essays focused upon the same field.

In the case of mode, there were various points of similarity and difference amongst the essays produced by the students. Mode was realized by (1) lexical density (2) nominalization and (3) the employment of topical, textual and interpersonal theme. The lexical density of a text is a function of its place on a register scale which ranges from most active to most reflective [15]. The more inclined a text is to the “language-in-action” end of the scale, the lower will be its lexical density. Since written language is mostly classified as reflective rather than active, expectations are such that the lexical density will increase in value as the text inclines, incrementally, towards written discourse, away from spontaneous speech.

A presumptive conclusion at this stage might be that essays written by PRC students approximate most closely to the spoken mode, compared to those of SC or SE students. True as this may be, this conclusion is inherently problematic. According to [8]. the employment of simple sentences should contribute to a higher lexical density, given the resultant existence of fewer clauses per sentence. Notwithstanding this, the PRC students’ inclination to employ simple sentential structures does not contribute to a higher average lexical density in their essays. Rather, the reason for their lower lexical density is the sparse number of content words per clause in their essays, compared to those of SE and SC counterparts (refer again to Example 1). This follows from the fact that the number of content words is directly proportional to lexical density (refer to the Formula (I): Lexical Density). The infrequent use of content words on the part of PRC students is perhaps attributable to their limited vocabulary, which is in turn ascribable to the fact that they started learning English only at the late Primary or Secondary level.

The same kind of argument could therefore be extended to SC essays, which had a relatively lower lexical density than SE essays. However, the reasons for the slightly lower lexical density in SC essays were twofold. First, SC students might not be as well-equipped with vocabulary items as SE students (due to differences in linguistic background), notwithstanding the fact that there were only a few apparent variations between their essays in this respect. A second but more important reason is that the SC students had a penchant for producing complex sentential structures, which contained more clauses per sentence, thereby reducing lexical density (refer to Formula on Lexical Density above). For instance:

Example B: Extract from SC1:

(13) / You will be surprised to learn / that it is not uncommon to hear of highly-educated Japanese women/ /who stay at home to take over the learning of the children/. LD = 14/3 = 4.67

Both SE and SC students’ essays contained about the same number of content words per clause. However, in contrast to SE essays, SC essays almost invariably consisted of 2 or 3 clauses clustered together forming a very long, grammatically intricate sentence. As the formula is Lexical Density = No. of content words / no. of clauses, the increase in the number of clauses in SC essays caused the lexical density to decrease.

As lexical density correlates with nominalization, it is not surprising that PRC essays have the lowest degree of nominalization and SE essays have the highest. A more interesting point is that SE students employ a higher degree of nominalization than SC students. The reason is due to SE students being much better equipped with the skills of nominalization than their SC counterparts as a consequence of their English-speaking background. SE students utilised English in their everyday life, whether it involved letter-writing, drawing up a shopping list, or diary keeping. These different modes of writing perhaps enabled the SE students to build up a repertoire of writing skills, which would be useful for expository writing at tertiary level.

However, a higher degree of nominalization is not necessarily indicative of productive, well-written work. In any case, nominalization itself can “reduce the clarity and crispness” of a writer’s style, causing a piece of writing to sound “technical” or “scientific” in [12]. Nevertheless, it is the general consensus of [12] and other academics that nominalization does lend a certain formality to written “academic” work. Nominalizations facilitate discussion of processes and qualities and permits effective summary of ideas, both of which are favoured in expository writing.

In the case of themes, the Singaporean students used more textual and interpersonal themes than the PRC students. These results are interesting because (1) A greater proportion of employed textual themes suggests a propensity for Singaporean students to conjoin their sentences in a more complex manner than PRC students; and (2) A greater number of employed interpersonal themes suggests an inclination for Singaporean students to express their judgement regarding the relevance of the content in the essay. In fact, [16] suggested that a higher incidence of interpersonal Theme in a piece of writing indicates that the writer regards the piece of writing more as an argumentative discourse than merely as an (factual) expository one. Yet, interestingly, the author feels that the lower incidence of interpersonal themes in PRC essays does not mean that they are any less capable than their SE counterparts in perceiving the argumentative nature of the essay. In fact, their attempt to express their judgement on the issue was flooded by non-epistemic usage of emotionally charged words to indicate absolute disagreement on the given topic (as illustrated earlier in Section 4.3: Epistemic Writing). The lower proportion of interpersonal themes is therefore more attributable to the fact that they are not used to writing expository essays, and are hence handicapped by their ability to express their judgement in a more detached/objective manner.

Tenor is realized by the relationship between the reader and the writer. The results suggested that was realized differently by PRC and Singaporean students, especially in the areas of mood and epistemic writing. First, PRC and SC students were generally more likely to use modals of higher probability or obligation such as must and will, or certainly and always, compared to their SE counterparts. Second, PRC students were more inclined to employ lexical choices having an extreme degree in meaning such as completely, absolutely. Thus, the results suggest that PRC students displayed much greater emotional involvement than the Singaporean students. They expressed themselves as though they were engaging in oral dialogue with the reader in oral discourse. The Singaporean students, on the other hand, were much more detached and impersonal: they did not address the reader nor did they show much affect in relation to the topic.

Therefore, epistemic writing, or the ability to write formally and academically in a detached and objective manner, was virtually non-existent in PRC essays. This was in sharp contrast with the writing of Singaporean students. This is based on the observation that they utilized expressions conveying an impression of directness and absoluteness, such as (x) This statement is completely untrue. Such an authoritative tone might be considered refreshing insofar as it reflects beliefs held with evident conviction and emotional commitment attributes which are rarely exhibited in written work at tertiary level. Unfortunately, such unmodulated writing is often not well-accepted because it violates genre expectations of a certain level of formality, detachment and objectivity demanded by expository/argumentative writing at tertiary level. In fact, non- epistemic statements such as (x) exemplified above are more likely to be found in spoken discourse, where arguments on the same topic can be engaged in on an emotional level, often without the need to provide the relevant justifications required in written discourse.

There are possible reasons why PRC students should write more emotively than Singaporean students with regard to the topic “A woman’s place is in the home”. Chinese women have been expected to work in the fields since time immemorial. With the upsurge of feminism and liberalism in recent decades (especially since 1949), more and more women have joined the workforce. Whether they worked as farmers or doctors, it is increasingly the case that a woman’s place is not in the home. This implies that Chinese women (and even men) carry the conviction that their place is in society, and therefore write more emotionally about the issue than Singaporeans.

The PRC students possess fewer skills and strategies in stating emotionally-coloured ideas in more objective terms. Nevertheless, personal feedback from the author’s students indicates that the proper way to write Chinese expository essays is to use emotionally-charged words in order to create an impact on the reader. Therefore, it is more likely the case that PRC students would extend this strategy of writing Chinese expository essays, and use it for English expositions as well. This ability to incorporate the so-called “writer’s voice” is undoubtedly an advantageous asset that PRC students possess over Singaporean students. Nonetheless, this asset needs to be honed and modulated so that the impression of subjective personal judgement is tempered by a more objective yet authoritative manner of expressing one’s views.

Finally, the fact that PRC students frequently employed expressions of certainty such as absolute, completely perhaps has to do with the fact that they come from an authoritarian society. In the Chinese culture, laws of society are established with absolute certainty.

Questioning of authority is often not permissible. As language often reflects reality, one might associate the high degree of certainty expressed by PRC students with their socio-cultural background, and a political milieu dominated by dogmatic ideology and the political imperative.

6. Pedagogical Implications: Curriculum Design and Implications

It is essential that the formal aspects of writing be integrated with the writing process to attain professional standards of writing. Although [17] discovered that formal grammar instruction produces few positive results when taught as a school subject, many researchers and practitioners have asserted the necessity for students to pay attention to formal aspects of their writing refer to [18].

The issue in question is therefore not whether language forms and structures are useful, but whether teachers can aid students in recognising the relations between language structures and the roles they play in conveying the appropriate meaning. Such awareness of language as a resource permits students to access what [11] describes as the discourse of power, i.e., the ways of writing by which people organise and influence the world around them. In addition, it permits teachers to carry out professional preparation and planning prior to conducting the lessons.

In order to engage students’ interests and purposes for writing, the author feels that the most important aspects of language pedagogy is to allow the student himself/herself to be able to select some of his/her own material for writing assignments that would be assessed in the curriculum. Although this student-centred approach may lead to controversial issues such as lack of standardisation or teacher control, it may be of great assistance to students who might write better as a result of realising that they are now able to write about a topic that they initiated. Another instructional method whereby writing could be an occasion for learning is when students do not hand in the first draft that they write, but instead, hand it to their peers for evaluation. In this manner, students learn to assess a piece of writing, and in turn, may become more aware of their position as readers when they produce written work. They learn to pick out incoherence, grammatical inconsistency, contradictions and lack of cohesion in their peers’ essays, and as a result, become more sensitised to their own writing in the future, i.e., more self -aware.

Theoretically, many Singaporean students satisfied the requirements of the generic structure required by “A woman’s place is in the home”, by providing a good thesis statement and being able to support their arguments with examples. This is not surprising as they have had exposure to expository writing, for two years in a junior college. However, despite this exposure, many Singaporean students do not write very coherently nor are they able to bring out very creative or well-substantiated examples to illustrate their points, even if they were familiar with the generic structure of the essays. Hence, though it is no longer necessary to explicitly teach the writing of expositions to Singaporean students at tertiary level, it may be necessary to keep the focus on teaching them how to write essays coherently. The PRC students, on the other hand, are mostly starting from scratch where writing of expositions is concerned. For these students, raising a certain amount of genre-consciousness will be helpful, and this can be either taught implicitly or explicitly. Implicit teaching of expositions could involve handing them abundant reading material by telling them what is acceptable and what is not. Explicit teaching of expositions could entail the actual presentation of the generic structure (for instance, the three- or five- paragraph model mentioned in Chapter Two) to the students, and requiring that they learn what is expected of them in the introduction, body, and concluding paragraph.

A third instructional method is to permit students to examine given-new relations in student-written texts and see how information is presented in [18]. For instance, they could be made to realize that other types of topical progression could be employed besides constant and dove-tailing Themes. The more sophisticated patterns of topical progression such as exposition of split Themes or rhemes could be taught by the utilisation of examples to aid instruction of such progressions. Alternatively, students could focus on first and last sentences of a paragraph, or propose the best transition sentences for the beginning of a new paragraph.

As the results revealed that PRC students faced general difficulty with writing epistemically, a fourth pedagogical method towards improving students’ writing would be to teach them how to use personal pronouns in relation to expository/argumentative essays. For instance, speech-like features such as “I” and “you” which constantly appeared in their essays should be reduced, as they might convey an impression of informality. Furthermore, although PRC students should be encouraged to retain their propensity to express their emotions, they should be taught how to moderate such expressions in the context of writing in a tertiary institution. For instance, the students might first be informed that expository essays require opinions to be expressed in an objective manner. They should therefore learn to channel the demonstration of personal affect or emotion towards a topic in a different manner, e.g. byemploying expressions such as “It is my firm opinion/belief that…” or “It is strongly recommended that……” in place of expressions such as “I definitely believe that…..” and “People must be able to understand that a woman’s place….

A fifth pedagogical technique could be to encourage students to examine their word choices and explore how word changes influence writing. In particular, the effect of switching verb groups to nominalised words and phrases, or the influence of compressing clause information into prepositional phrases, participial phrases, relative clauses and adjectives and adverbs can be explored in [18]. In the same manner, students could be encouraged to vary their use of different types of modals, e.g. might vs. will. In this way, their awareness of the hedging valence of different modals might be heightened in [12].

Finally, the students might be encouraged to keep a journal recording their thoughts and feelings in relation to their usage of the language. This would assist them in visualising an imagined audience reading their written work, and perhaps make salient the fact that they are required to write in a manner which enables not only themselves, but also the reader, to understand their written content.

7. Closing Comments

This study has presented an overview of the characteristics of written discourse pertaining to the genre of an expository/argumentative essay “A woman’s place is in the home”. It is hoped that the characteristics of written discourse which appertain to the different categories of students who participated in this study will provide useful insights into the kind of writing that one might generally expect from Singaporean or PRC students. However, these characteristics are not claimed to be representative of all writing produced by Singaporean and PRC students, as there will always be exceptions pertaining to students from each group.


[1]  Halliday, M. A. K. 2013 (In-press) An Introduction to Functional Grammar. London: Routledge.
In article      
[2]  Halliday, M.A.K. and Matthiessen, C. N. I. M. 2004. An Introduction to Functional Grammar. London: Hodder Arnold.
In article      
[3]  Butt, D., Fahey, R. Feez, S. Spinks, S. Yallop, C. 2003. Using Functional Grammar: An Explorer’s Guide. Sydney: Macquarie University (NCELTR).
In article      
[4]  Rose, D. and Martin, J. R. 2012. Learning to Write, Reading to Learn: Genre, knowledge and pedagogy in the Sydney School. London: Equinox.
In article      
[5]  Hyland, K. 1998. Hedging in Scientific Research Articles. Amsterdam: John Benjamins Publishing Company.
In article      
[6]  Teng, S. C. 1998. Linguistic Differences in Texts Produced under Examination and Non-Examination Conditions. Singapore: Department of English Language and Literature, National University of Singapore. Unpublished PhD Thesis.
In article      
[7]  Allison, D. and Wu, S. M. 2000. What Counts as Critical. Proceedings of the First Symposium on Teaching and Learning in Higher Education, ed. By C. M. Wang, K. P. Mohanan, D. Pan, and Y. S. Chee, 107-112. Singapore: CDTL, NUS.
In article      
[8]  Halliday, M. A. K. 1987. Spoken and Written Modes of Meaning. In R. Horowitz and S. J. Samuels (Eds), Comprehending Oral and Written Language, (pp 55-82). San Diego: Academic Press. Inc.
In article      PubMed
[9]  Shaughnessy, M.P. 1977. Errors and expectations: A guide for the teacher of basic writing. New York: Oxford University Press.
In article      
[10]  Raskin, V. and Weiser, I. 1987. Language and Writing: Application of Linguistics to Rhetoric and Composition. Norwood, N. J.: Ablex Publishing Corporation.
In article      PubMed
[11]  Martin, J. R., Matthiessen, C. M. I. M. and Painter, C. 1997. Working with Functional Grammar. New York: Arnold.
In article      
[12]  Hyland, K. and Milton J. 1997. Qualification and Certainty in L1 and L2 Writing. Journal of Second Language Writing, Volume 6, 83-203.
In article      CrossRef
[13]  Eggins, S. 1994. An Introduction to Systemic Functional Linguistics. London: Pinter Publisher.
In article      
[14]  Chafe, W. and Danielewicz, J. 1987.Properties of Spoken and Written Language. Thematic Development in English Texts, In M. Ghadessy (Ed.), Thematic Development in English Text (pp 83-113). London: Pinter Publisher.
In article      
[15]  Ure, J. 1971 Lexical density and register differentiation. In G. Perren and J.L.M. Trim (eds), Applications of Linguistics, London: Cambridge University Press. 443-452.
In article      
[16]  Ku, G. B. 1997. A Description of Recontextualized Knowledge Expressed in the Written Output of Lower Secondary Pupils in Singapore in English and Mandarin. Singapore: Department of English Language and Literature, National University of Singapore. Unpublished Masters Thesis
In article      
[17]  Hillocks, J. R. G. 1986. Research on Written Composition. United States of America: The University of Chicago NCRE Publications c/o Department of Education and English.
In article      
[18]  Grabe, W. and Kaplan, R. B. 1996. Theory and Practice of Writing. United States of America: Addison Wesley Longman Limited
In article      PubMed
  • CiteULikeCiteULike
  • MendeleyMendeley
  • StumbleUponStumbleUpon
  • Add to DeliciousDelicious
  • FacebookFacebook
  • TwitterTwitter
  • LinkedInLinkedIn