Discourse Markers in the Genre of Formal Letters Writtenby Intermediate Students of English as a Foreign Language

This article involves a quantitative computer-assisted study of the use of discourse markers in the genre of formal letters written by a group of students of English as a foreign language (EFL) at the intermediate level of EFL proficiency. The aim of the study was to examine how the group of EFL students uses English discourse markers in a written task involving the genre of formal letters. The study employed a quantitative computer-assisted methodology of computing the frequencies of discourse markers by software program WordSmith (Scott, 2008). The results of the data analysis indicated that the group of EFL students at the intermediate level of EFL proficiency exhibited awareness of the genre conventions associated with the usage of discourse markers in the task. The EFL students’ genre awareness was manifested by the use of stylistically neutral discourse markers (e.g., also, but, however) concurrently with those discourse markers that were typically associated with a formal register of the English language (e.g., furthermore, hereby, therefore). These findings and their didactic implications to the teaching of EFL writing are further discussed in the article.


Introduction S O C I O L I N G U I S T I C S / S O C I O L I N G V I S T I K A
This article involves a computer-assisted study that aims to establish how students of English as a foreign language (EFL) use English discourse markers (further in the article referred to as DMs) in a written task associated with the writing of a formal letter.It is argued in the study that the use of DMs exemplifies EFL students' genres awareness concurrently with being a token of their pragmatic competence.Following Swales (1990), genre in the present study is regarded as "a class of communicative events, the members of which share some set of communicative purposes . . .recognised by the expert members of the parent discourse community" (Swales, 1990, p. 58).In accordance with Tardy (2009), this study adopts the notion of genre awareness that is comprised of formal knowledge about the structural elements of genre (e.g., discursive, lexical, and syntactic conventions associated with a particular genre), process knowledge (e.g., the processes of composition of written texts for a particular audience), rhetorical knowledge (e.g., the intended purpose of a written text), and subject-matter knowledge (e.g., discipline-specific knowledge).The study follows the definition of pragmatic competence formulated by Ifantidou (2011, pp. 332-333), who views it as an ability to identify, reflect, and use linguo-pragmatic knowledge in the written task production in a foreign language.
Guided by the afore-mentioned notions of genre (Swales, 1990), genre awareness (Tardy, 2009), and pragmatic competence (Ifantidou, 2011) as the theoretical background of the study, I argue that DMs can be analysed as genre-appropriate indexes used by the group of EFL students (further referred to as participants), who are enrolled in the university course "Effective Communication in English (ENEC14)" (Stockholm University, 2017) that is designed for an intermediate level of EFL proficiency, i.e., B1-B2 levels according to the Council of Europe's "Common European Framework of Reference for Languages: Learning, Teaching, Assessment" (The Council of Europe, 2011).Extending this argument further, I hypothesise that the use of DMs by the participants in the task of formal letter writing would provide relevant indexes that would be reflective of the genre-specific conventions of formal letter writing.Specifically, I hypothesise that the participants' genre awareness and the level of pragmatic competence would be associated with the use of stylistically formal English DMs in the written task involving formal letter writing.
The study further presented in this article has several novel aspects.First, given that research on the use of DMs in formal letter writing in EFL is underrepresented, the study will offer new insights into the role of DMs in the written task executed by intermediate EFL students.Second, the study is contextualised in the teaching situation of a stand-alone EFL course offered to non-linguistics majors at a university in Sweden.It should be emphasised that the study is not conducted as an experimental procedure.Instead, it capitalises on the classroombased teaching within the context of the actual EFL course.Third, in accordance with Cheng (2007) the afore-mentioned EFL course can be characterised as an English for Specific Purposes (ESP) module, since the students enrolled in the course are non-linguistics majors whose educational background involves a diverse range of disciplines, from engineering to journalism.Assuming that "research on how students analyze and produce genres in ESP genre-based writing classes is still underrepresented in the literature" (Kuteeva, 2013, p. 85), a scientific inquiry into the use of DMs in the genre of formal letter writing within EFL/ESP settings appears novel and relevant.Fourth, another novel aspect of this study involves the focus on Swedish university learners and ESP.This article is structured as follows: First, I will outline previous research on DMs in written tasks in EFL/ESP.Second, I will provide an overview of the state-of-the-art research involving DMs in the genre of formal letter writing in relation to EFL/ESP.Third, the present study will be introduced and discussed in detail.Lastly, the article will offer conclusions in conjunction with the linguo-didactic implications relevant to the teaching and learning of formal letter writing in the field of EFL/ESP.

Research on DMs in Written Tasks in EFL/ESP
There has been a substantial increase in research on DMs in a variety of linguistic subfields (Waltereit 2002: 987), where DMs are referred to by different names, for instance, discourse connectives, discourse operators, pragmatic connectives, sentence connectives, and cue phrases (Fraser, 1999, p. 93).A meta-analysis of previous research literature suggests that to-date there is no definition of DMs specific to the fields of EFL and/or ESP (Gablasova et al., 2017;House, 2013;Müller, 2005).In particular, in a study on the use of the English DM so by EFL learners, Buysse (2012, p. 1764) defines DMs as "optional linguistic items that fulfil an indexical function, in that they connect an utterance to its co-text and/or the context".Interestingly, similar definitions of DMs are proposed in non-EFL linguistic subfields, e.g.discourse studies, and pragmatics.For instance, Schiffrin (2005, p. 57) defines DMs as "sequentially dependent elements that bracket units of talk (…), i.e., nonobligatory utteranceinitial items that function in relation to ongoing talk and text".Fraser (1990) regards DMs as "a class of expressions, each of which signals how the speaker intends the basic message that follows to relate to the prior discourse" (1990, p. 387).Similarly, Hennecke (2017, p. 356) defines DMs as "polysemous, polyfunctional, and syntactically flexible lexical items that occur mostly in sentence-peripheral positions".As evident from the above-mentioned definitions of DMs in EFL (Buysse, 2012) on the one hand and in discourse studies and pragmatics (Brinton, 1996;Fraser, 1990;Hennecke, 2017;Schiffrin, 2005) on the other hand, in these linguistic subdisciplines DMs are seen as optional or nonobligatory expressions that connect one stretch of discourse with another, thus joining the sentence-final discursive chunk in sentence 1 (S1) with the sentence-initial chunk in another sentence 2 (S2).
In discourse and EFL studies, DMs are thought to form a heterogeneous group of lexical items (Fraser, 2015;Schiffrin, 2005) that include "adverbs (consequently, honestly), prepositional phrases (in fact, on the contrary), particles (even, only), subordinating conjunctions (because, though), coordinating conjunctions (and, but), predications (you know, mind you), etc." (Ranger, 2015, p. 155).As indicated by Fox Tree (2001; 2010), the heterogeneity of DMs can be explained by multiple functions they perform in oral and written discourse.Specifically, these functions "can be conceptualized as a long list of varying roles, including contributing to local coherence of adjacent phrases, assisting in turn-taking or repair, or contributing to social solidarity" (Fox Tree, 2010, p. 270).Amongst a plethora of functions that DMs appear to perform in discourse, Schiffrin (2005) refers to that of cohesion.This contention is echoed by Fitzmaurice (2004), who indicates that such DMs as well, you know, I mean, etc. "oil the wheels of conversational exchange" (Fitzmaurice, 2004, p. 428) in oral discourse.Similarly, DMs in written communication facilitate textual cohesion (Fraser, 1999).Hence, it can be assumed that DMs are an important discursive means that contributes to cohesion both in oral and written discourse in the speaker's first language (L1) and second (L2) or a foreign language, such as EFL.
Previous literature (Bikelienė, 2017;Buysse, 2010;Fox Tree, 2010;Povolná, 2012) provides evidence of the difference in the range as well as functions of DMs used by L1 and L2 speakers.In this regard, Fox Tree (2010) argues that the use of DMs in L2 increases with the learner's proficiency in L2.However, the increase is non-linear and, potentially, is associated with learning difficulties, since an L2 or an EFL learner may already possess a number of DMs in their L1 (Fox Tree, 2010).Prior studies suggest that whilst some DMs are identical between the languages, e.g., uhm in English and Dutch (Fox Tree 2001), there is typically no direct correspondence between DMs in the learner's language pair combination.Presumably, the need to acquire and use DMs appropriately in one's L2 and/or in EFL may pose significant difficulties for an L2 and/or EFL learner.Arguably, these difficulties may map onto the learner's performance in written tasks in L2 and/or EFL.In this regard, Aijmer (2002, p. 3) indicates that incorrect use of DMs by non-native speakers, e.g.EFL learners, may lead to misunderstandings, and, potentially, impede the successful written task production in EFL.
The use of DMs in EFL written tasks executed by adult university students has been examined in several corpus-based studies from an applied linguistics perspective (Bikelienė, 2017;Dülger, 2007;Kapranov, 2017;Martínez, 2004;Povolná, 2012;Rahimi, 2011).By means of using various corpora, these studies focus on the role of DMs in essay writing in EFL by Lithuanian L1 university students (Bikelienė, 2017), academic writing in EFL by Turkish L1 university students who major in English (Dülger, 2007), academic writing by Swedish L1 pre-service teachers of English (Kapranov, 2017), expository writing by Spanish L1 university students (Martínez, 2004), academic writing in Masters theses by Czech L1 university students (Povolná, 2012), and argumentative and expository essays by EFL students whose L1 is Farsi (Rahimi, 2011).In particular, Bikelienė (2017) investigates the use of the DM then by Lithuanian L1 university students of English.Bikelienė (2017) reports that the DM then has been identified in one out of four files in the corpus of EFL essays (Bikelienė, 2017, p. 106).By means of contrasting the written and spoken EFL corpora of Lithuanian L1 EFL students, Bikelienė (2017) indicates that the DM then is used "significantly more often in spoken interactions than in writing whereas there were no considerable differences in the overall frequency among written text types" (Bikelienė, 2017, p. 106).Moreover, there is a statistically significant difference between the use of the DM then in oral speech by Lithuanian L1 EFL students and the use of then in EFL academic essays.
Whilst Bikelienė (2017) focuses on the investigation of one DM (then), in a corpus-assisted study by Dülger (2007) a cumulative number of DMs in EFL essays is presented and the frequencies of individual DMs are not discussed.Dülger (2007) reports an increase in the frequency of DMs that are used by the EFL majors, who appear to employ more DMs after they have completed a course on EFL writing.Notably, the increase is concurrent with the effect of those EFL writing courses that focused on the process of writing rather than on the final product of writing.In this regard, Dülger (2007, p. 264) indicates that "the total number of discourse markers increased from 803 to 1054 in the essays written after the processoriented courses".
Similarly to Dülger (2007), Kapranov (2017) has found an increase of the use of DMs in the corpus of the final EFL academic essays contrasted with the essay drafts written by Swedish L1 university students.In particular, the increase involves a more frequent and genreappropriate use of DMs that are associated with the formal register of academic writing in English.To illustrate the point, Kapranov (2017) reports that whilst the DMs generally, just, like, probably, usually, and well are present in the essay drafts, they are absent from the final essays.Instead, the participants in the study (Kapranov, 2017, p. 32) employ such formal DMs, as although, assuming, nevertheless, rather, thereafter, and thereby.
Whereas the study by Kapranov consists of the corpus of the essay drafts and the final essays, the corpus compiled by Martínez (2004, p. 71) involves 78 expository essays written by Spanish L1 university students of the first year of study.Martínez (2004) indicates that elaborative DMs are the most frequent (45.2%)DMs in the corpus, followed by contrastive DMs.Amongst the frequently used elaborative DMs that have been identified in the corpus, Martínez (2004, p. 71) mentions also, besides, in addition, moreover, that is (to say), likewise, and, for example, to sum up, and in short.It has been found that those EFL students whose essays received higher marks use a more varied range of DMs, whilst their peers whose essays were given lower marks repeat the same DMs (Martínez, 2004).Martínez (2004) points to the "difference in the use of the elaborative, contrastive and topic relating DMs between the better writers and the weaker ones" (Martínez, 2004, p. 77).
In contrast to the afore-mentioned studies (Bikelienė, 2017;Dülger, 2007;Kapranov, 2017;Martínez, 2004) that involve essay corpora, the investigation of DMs by Povolná (2012) is based on a corpus of 15 Master's theses written by EFL majors.Povolná (2012) has discovered that Master's students whose L1 is Czech appear to employ causal DMs followed by contrastive DMs in their theses.Povolná (2012) reports that the most frequent causal DMs used by the students are as, because, and since.Povolná (2012) observes that there is a quantitative difference in the use of English DMs that can be, arguably, ascribed to the topic of the Master's thesis, e.g.linguistics, literature, and EFL methodology, respectively.Additionally, Povolná (2012) notes qualitative and quantitative differences in the use of DMs by the individual Master's students (Povolná, 2012, 146).
By means of investigating a corpus of argumentative and expository essays written in English by Iranian university students, Rahimi (2011) has found that the frequency of DMs is higher in argumentative essays with 1176 occurrences, whilst the use of DMs is lower in expository essays with 965 occurrences.Rahimi (2011) reports that the most frequently used DM in all assays is the elaborative DM and, which is represented by 1227 occurrences in the corpus of both argumentative and expository essays.Other more frequently used DMs are or, so, but, also, and because (Rahimi, 2011, p. 72).Rahimi (2011) concludes that EFL students whose L1 is Farsi "do not use a wide range of DMs and that they use some particular elaborative markers like "and" in a significantly higher manner than other markers" (Rahimi, 2011, p. 75).

Previous Research Involving DMs in the Genre of Formal Letter Writing
Prior to proceeding to the outline of previous research involving DMs in the genre of formal letters, it is pertinent to elucidate the notion of genre.As posited by Ifantidou (2011, p. 331), genre is typically viewed through the lenses of social constructs and formulaic prototypes.
In applied linguistics and EFL/ESP, genre can be regarded as a socially embedded process characterised by communicative purposes and socially recognised goals (Ifantidou, 2011, p. 331).In EFL/ESP studies, research seems to follow the definition of genre by Swales (1990), who regards it as "a class of communicative events, the members of which share some set of communicative purposes" (Swales, 1990, p. 58), which are recognised in the given discourse community and constitute the rationale for the genre (ibid.).Swales (1990, p. 58) posits that the rationale determines the schematic structure of discourse, as well as its content and styles constraints.
Specifying the definition of genre by Swales (1990), Cheng (2006, p. 77) posits that genre is associated with social interactions, rhetorical contexts, and such formal properties as structure and style (Cheng, 2006, p. 77).This contention is echoed by Nunan (2007), who suggests that a socially constructed written text (e.g., a narrative, a description) has its "own characteristic structure and grammatical form that reflects its social purpose" (Nunan, 2007, p. 209).In this regard, Chovanec (2012, p. 5) suggests that genres are seen as "configurations of expected forms and meanings that a given community associates with particular situations and acknowledges as distinct from each other." Based upon the above-mentioned approaches to genre (Cheng, 2006;Chovanec, 2012;Nunan, 2002;Swales, 1990), it is possible to assume that formal letters, for instance, covering letters in the job application process, letters of complaint addressed to a business entity, dismissals, promotions, recommendations (Kaur & Singh, 2013), letters by Chief Executive Officers (CEOs) that usually are a part of corporate annual reports (Hyland, 1998) constitute a genre, or at least, a subgenre in its own right.This assumption is further supported by Yates (1989), who indicates that formal letters, in particular, business letters, are remarkably formulaic.They are characterised by clichés, i.e. expressions that are frequently used in formal letter writing (Yates, 1989).Arguably, formulaic expressions facilitate the process of writing a formal letter, since they offer a repertoire of genre-appropriate and stylistically elevated lexical means the writer may choose from (Jenkins & Hinds, 1987).This view is elaborated by Kaur and Singh (2013), who posit that the author of a formal letter "has to be careful in choosing the right words and the right tone, and focus on the purpose of the correspondence" (Kaur & Singh, 2013, p. 187).Presumably, DMs could be considered one of the genre-appropriate "right words" (Kaur & Singh, 2013) that are associated with formal letters.
Previous research on the use of DMs in formal letter writing examines financial disclosures, CEOs' letters to shareholders, and corporate letters (Garzone, 2005;Hyland, 1998;Kapranov, 2016).In these types of formal business communication, DMs play a useful rhetorical role (Camiciottoli, 2010).For instance, Hyland (1998) reports that DMs therefore and nevertheless are used in business discourse to guide the readers' attention in the direction desirable to the company's management (Hyland, 1998).Hyland (1998)

The Present Study: Hypothesis and Specific Research Aims
As mentioned above, there is insufficient research in EFL/ESP studies on the use of English DMs in formal letter writing.The present study seeks to address this issue within the theoretical premises formulated by Swales (1990), Tardy (2009), and Ifantidou (2011).This study is embedded in a university EFL course designed for non-EFL/non-linguistics majors.
In a sense, as previously noted in the introduction, this is an ESP course tailored to the needs of a range of students enrolled in such university programmes as business administration, chemistry, engineering, journalism, etc.
It is assumed in the study that the use of DMs in the task of formal letter writing executed by the participants would provide relevant indexes that would be associated with the genrespecific conventions of formal letters in the English language.Specifically, it is hypothesised that the participants' genre awareness and the level of pragmatic competence would be associated with the use of stylistically formal English DMs in the written task.This hypothesis is guided by previous research findings (Kapranov, 2017) that indicate that the range of formal DMs, such as hence, presumably, therefore, etc. would increase with the development of EFL proficiency.Arguably, the participants would be able to write a formal letter using the genreappropriate formal DMs.
Following the hypothesis, the specific research aims of this study are to i) identify DMs in the task of formal letter writing and ii) establish the frequency of DMs used by the participants in the task.Additionally, the study seeks to establish whether or not there would be quantitative differences in terms of the use of formal DMs and informal DMs in the task.Following Nunan (2008, p. 58), a certain degree of variation in the genre structure that involve DMs pertaining to different registers would be expected in the formal letters written by the participants.Presumably, the frequency of DMs in the task would be suggestive of a potential range of variation that involves informal, stylistically neutral, and formal DMs.Hewings (2005).The students are expected to attend at least 80% of the seminars.The written tasks during the course involve the writing of a short argumentative essay, a formal letter, and a genre project of the student's choice.
The written task of formal letter is given in the middle of the course after the introductory seminar on the genre-specific conventions associated with the formal letter writing in the English language.Typically, the introductory seminar focuses on several types of formal letter writing, especially on a covering letter, and a letter of complaint.At the introductory seminar, the students are explicitly taught that job applications consist of a covering letter and the obligatory enclosures, such as, for instance, CV and resume.The students are made aware of the main functions of a covering letter, its structure, specific clichés, grammatical features, and DMs that should be used for the purposes of coherence.It is explained to the students that the covering letter should involve the applicant's qualifications and work experience.
In terms of the writing of a letter of complaint, it is explained that it should be written in the formal register of the English language.The students are provided with explicit suggestions as far as the level of formality in vocabulary is concerned.It is pointed out that formal lexical items, inclusive of the formal DMs, are expected in this task.The explicit instruction of formal letter writing involves the focus on grammatical constructions relevant to the genre.Additionally, the students are reminded of being polite in their formal letter of complaint in order to avoid unnecessary confrontation.
Typically, the task of formal letter writing is expected to be completed at home and later submitted on the student platform for the subsequent evaluation by the course teacher (who is the author of this article).The students are required to indicate that they submit their own work, i.e. they testify in writing that the submitted text is not plagiarised.The time of the task execution is two weeks starting from the introductory seminar on formal letter writing.The students are instructed that their formal letter should be between 300 and 500 words inclusive of all the necessary enclosures, if any.

The Context of the Study
Participants, Procedure, and Method In total, 18 participants took part in the study.Initially, the group was comprised of 19 participants, however one participant was factored out from the study on the grounds of the participant's status of a native speaker of English.All 18 participants were deemed to be at English B1-B2 levels of proficiency (The Council of Europe, 2011), since the course eligibility criteria required English B1-B2 or an equivalent of at least 60 university credits in an academic subject taught in English.The participants' real names were coded to ensure confidentiality.The following codes were applied, e.g., P1, P2, P3, etc., where P denoted "participant" plus the corresponding number.The participants' descriptive statistics were summarised in Table 1 below:   As far as the procedure in the present study was concerned, it involved the following steps.First, the participants attended a course seminar on formal letter writing.At the seminar, the course teacher explained the structure, purpose, and lexico-syntactic characteristics of the genre of formal letter writing.The participants were provided with a sample of one covering letter, and another sample of a letter of complaint.The samples were analysed in class focusing on the genre-appropriate clichés, syntactic forms, and DMs associated with the formal writing style.Then, the participants were randomly divided into two sub-groups.Each sub-group was assigned one task, i.e. either a covering letter or a letter of complaint, to be written at home and submitted within two weeks after the introductory seminar on the formal letter writing.The DMs used in the sample of a covering letter were and, also, since, therefore, and yet.In the sample of a letter of complaint, the following DMs were introduced, e.g.although, and, but, consequently, furthermore, however, and still.
The present study followed the methodological premises described in Povolná (2012), where the quantitative analysis was based upon the frequency per 1,000 words.Following Povolná (2012), in this study the frequency of the DMs yielded by WordSmith (2008) was subsequently analysed qualitatively.It should be specified that the corpus consisted of two sub-corpora, 9 covering letters and 9 letters of complaint, respectively.All covering letters (total N = 9) written by the participants were collapsed into one file and analysed in WordSmith (Scott, 2008).Similarly, the same procedure was applied to the participants' letters of complaint (total N = 9).

The data
The corpus of the study was comprised of formal letters written by 18 participants, the total number of words = 5,677.Given the small size of the corpus (i.e., 5,677 tokens), the present study should be considered a case study that would show some tendencies for future extensive research with more participants and statistically significant results.To reiterate, due to the size of the corpus the present study operated on the level of the descriptive statistics.The descriptive statistics involved all 18 formal letters (both the sub-corpora of letters of complaint and covering letters) and were given in Table 2 in the form of mean values and their standard deviations (STD) that involved mean number of words, sentences, and paragraphs, respectively.The application of the computer programme WordSmith (Scott, 2008) to the two sub-corpora, i.e., the sub-corpus of the covering letters and letters of complaint, respectively, has yielded descriptive statistics presented in Table 3.It is noted in previous studies that EFL writing is one of the most challenging areas for students (Ahmed, 2016), where "particular tasks require specific writing strategies beyond the general writing ability" (Upton & Connor, 2001, p. 314).Extending this contention further, it seems that the execution of the task of formal letter writing by the participants requires both pragmatic competence and genre awareness.Judging from the results of the data analysis (see Table 3), the participants' formal letters in both the sub-corpora are marked by the presence of formal and stylistically neutral DMs.Informal DMs, such as, for instance, like, you know, well, etc. have not been identified in the corpus.These findings are suggestive of the participants' genre awareness as posited by Tardy (2009), who indicates that it is comprised, among other variables, of formal knowledge about the structural elements of genre, such as discursive, and lexico-syntactic conventions associated with a particular genre.The absence of the informal DMs in the present corpus is also indicative of the participants' pragmatic competence, which according to Ifantidou (2011) is manifested as an ability to use linguopragmatic knowledge in the EFL written task production.

Results and Discussion
As seen in Table 3, the participants' formal letters involve a number of DMs that are typically associated with the genre of formal writing, e.g.consequently, furthermore, hereby, moreover, since, and therefore.In total, the frequency of these DMs is 15 per 1,000 words.Whilst these DMs do not constitute the most frequent DMs in the present corpus (see Table 3), it is, nevertheless, possible to observe that they are used concurrently with the stylistically neutral DMs, such as and, as, also, etc.In this regard, it should be noted that the present findings are in contrast with the contention that a certain degree of informality and the use of stylistically informal register is to be expected in an EFL learners' corpus (Nunan, 2008).As evident from Table 3, the participants in the study do not resort to the use of informal DMs in both the sub-corpora.In the corpus, however, there are two instances of the use of DMs that might be regarded as informal, specifically the DM hopefully occurs once in the covering letter, and the DM unfortunately in the letter of complaint (the number of occurrence = 1).Presumably, these two DMs could be seen as conveying emphasis rather than as indexes of informal style, as illustrated by the following excerpt taken form the participant's letter of complaint: (1) Dear Patrick, I am writing to express my disappointment with a recent experience I had at your store.I purchased a Dell Inspiron N5437 laptop (Core i3 4010U Silver, product ID: 33.013.455)from your main showroom.Unfortunately, after 3 days the laptop has stopped functioning because the power adapter provided with it is out of service.Thus, I cannot charge my device and it is completely of no use.(Participant P1, female) Whilst the DMs hopefully and unfortunately could be potentially treated as pertaining to the informal style of writing, the majority of the DMs summarised in Table 3 are associated with the neutral (also, and, besides, etc.) and formal (moreover, since, therefore, etc.) registers of the English language usage.The use of both the formal and stylistically neutral DMs is exemplified by the covering letter written by Participant P7: (2) Dear Lisa, I am writing to apply for the position of a Global Communication Support Coordinator, which was announced on the Arbetsförmedlingen (ID 2457104).I am interested in this great opportunity, since my working and education background fits the job very well.
Firstly, I have a global background for this global position.My original country is China and I had been working in New Zealand from 2009 to 2013 before moving to Sweden last year.Secondly, my working experience with media is closely relevant to the position, especially the experience with news website Skykiwi.com.I was fully experienced in website and forum maintenance, daily news writing, media cooperation as well as brand promotion.Besides, I participated in magazine editing and marketing, and then got a chance to work for a famous cosmetics company.
Moreover, my educational background makes me confident for this job.I got a bachelor degree in media management.The aim of the programme was to give solid knowledge in media production and marketing management.Thus, it is a habit for me to think more from a marketing perspective when doing media jobs.Also, I want to emphasize my Chinese and English language skills as another advantage for the job.(Participant P7, female) Arguably, the present findings support the hypothesis in this study, where it has been assumed that the participants would be able to write a formal letter using the genreappropriate formal DMs.To reiterate, the afore-mentioned hypothesis is based upon previous research (Kapranov, 2017), where it has been found that the course teacher's feedback in conjunction with the explicit mode of instruction have resulted in the EFL students' ability to produce academic essays using genre-appropriate DMs.In this study, similar effects of the explicit instruction in terms of the use of DMs are observed.Moreover, the results of the present study appear to support the findings in Kapranov (2017), where such formal DMs, as although, rather, and therefore have been reported as a consequence of implicit instruction.Specifically, the frequencies of these DMs in the present study are 0.1% (although), 0.1% (rather), and 0.1% (therefore).
Taking the results of the present study into consideration, it can be assumed that intermediate EFL students are able to exhibit pragmatic competence (Ifantidou, 2011) and genre awareness (Tardy, 2009) of the genre-appropriate DMs in the task of formal letter writing, provided that they have been exposed to the explicit mode of EFL instruction.However, these findings should be taken with caution, since the frequency of the formal DMs is relatively low, 15 in total per 1,000 words.It means that the participants seem to embed the use of the formal DMs into a more frequent use of neutral DMs, such as and, as and also.In other words, whilst the use of formal DMs does take place against the background of the absence of informal DMs in the corpus, the prevalence of stylistically neutral DMs suggests that the participants' writing in the task is marked by variance.
Arguably, a more frequent use of the formal DMs would be an index of the participants' pragmatic competence in the task.Whilst the frequency of formal DMs per 1,000 words is relatively low, it is laudable that the participants on the intermediate level of EFL proficiency do not resort to the informal DMs in the task.It can be argued that the absence of informal DMs in both the sub-corpora could be regarded as the index of the participants' pragmatic competence.Another finding that is suggestive of the participants' pragmatic competence in the task of formal letter writing stems from the observation that the participants make use of a wider repertoire of DM in comparison with those that have been exemplified in the two sample letters provided at the introductory seminar, e.g.and, also, since, therefore, and yet in the covering letter, and although, and, but, consequently, furthermore, however, and still in the letter of complaint.
Judging from the results of the data analysis, it is possible to distinguish several quantitatively substantial groups of DMs in the corpus.Based upon the classification of DMs proposed by Fraser (1999) and Povolná (2012), these groups are deemed to be associated with causal, contrastive, and elaborative relationships.The DMs that involve causal relationships are summarised in Table 4.
The findings summarised in Table 4 are in concert with the study conducted by Povolná (2012), where the most frequently used causal DMs are reported to be as, because, and since.Similarly to Povolná (2012), the DM as appears to be frequent (1% and 1%, respectively), followed by the DM because (0.1% and 0.3%), and the DM since (0.1% and 0.2%).As seen in Table 4, the sub-corpus of the letters of complaint appears to be characterised by a

Table 4
DMs associated with causal relationships in the corpus

Table 5
DMs associated with contrastive relationships in the corpus qualitatively more diverse repertoire of causal DMs (as, because, consequently, since, so, then, therefore, thus) in contrast to the sub-corpus of the covering letters, where the DMs consequently, if, and therefore are not present.Quantitatively, the causal DMs involve 50 occurrences per 1,000 words in the sub-corpus of the covering letters, whilst their number in the sub-corpus of the letters of complaint equals 57.Whilst the paired-sample t-test has not yielded any statistically significant results between the number of occurrences of DMs in these sub-corpora, it is possible to observe that the sub-corpus of the letters of complaints tends to be quantitatively more associated with the causal DMs.At the same time, this observation does not apply to the DM then whose frequency (0.1%) appears to be equally distributed between the sub-corpora.This finding seems to support previous research results obtained by Bikelienė (2017), who has found no considerable differences in the frequency of the DM then in a variety of EFL essays.Quantitatively, the sub-corpus of the covering letters involves 20 occurrences of contrastive DMs per 1,000 words, whilst the analogous measure in the sub-corpus of the letters of complaint equals 24 DMs.However, the repertoire of contrastive DMs used by the participants in the covering letters is qualitatively wider (see Table 5).
Another group of DMs that is substantially represented in the corpus involves elaborative relationships.This group of DMs is summarised in  The findings presented in Table 6 seem to support the results reported by Martínez (2004), who points to a prominent role of elaborative DMs in academic writing by EFL university students.Similarly to Martínez (2004, p. 71), the elaborative DMs in this corpus are also, and, besides, furthermore, and moreover.As evident from Table 6, the DM and appears to be the most frequently used (3%, and 2.5%, respectively) in the group of elaborative DMs.This finding is in agreement with the study by Rahimi (2011), where it is reported that the elaborative DM and is the most frequent DM in the corpus of both argumentative and expository essays.
In addition to the above-mentioned groups of DMs (i.e., causal, contrastive, and elaborative), the results of the data analysis indicate that there are two quantitatively smaller groups of 1) commentary DMs, such as hopefully (0.1%) and perhaps (0.1%) that occur in the covering letters, and the DM unfortunately (0.1%) that has been identified in the letter of complaint; 2) sequencing DMs, e.g.firstly (0.1%), lastly (0.1%), and secondly (0.1%) that are present in the covering letters.
Whilst the results of the data analysis exhibit a certain degree of variability in terms of the quantitative distribution and the qualitative repertoire of DMs, it is, nevertheless, observed that the participants avoid informal DMs and employ genre-appropriate DMs in the task.It follows that the genre of formal letters written by the participants involves a complex discursive space that is marked by the presence of stylistically neutral and formal DMs.
Paraphrasing Hyland (2003), and Myskow and Gordon (2009), it could be argued that this discursive space in EFL classroom settings has benefitted from a genre-based approach to the teaching and learning of DMs by means of the formal letter writing task that is timely and relevant to the particular needs of EFL students.

Conclusions and Linguo-Didactic Implications
The article discussed the use of DMs in the task of formal letter writing executed by the participants, who were EFL university students at the intermediate level of proficiency.The application of the quantitative computer-assisted methodology (Scott, 2008) to the corpus of the formal letters yielded the results that were indicative of the participants' awareness of the genre conventions associated with the use of DMs in the task.The participants' genre awareness was manifested by the use of stylistically neutral DMs (e.g., also, but, however) concurrently with the formal DMs (e.g., furthermore, hereby, therefore).Given that no informal DMs were identified in the corpus, that finding suggested that the participants exhibited genre awareness and pragmatic competence.Those findings should be approached with caution, since the number of participants was limited.However, even with the size of the present data, it was possible to formulate the following linguo-didactic suggestions that were relevant to EFL students at the intermediate level of EFL proficiency: i) explicit instruction of genre-based conventions of formal letter writing; 2) explicit instruction of the use of the genre-appropriate DMs in formal letter writing; 3) ample possibilities of formal letter writing embedded into a genre-based approach.
Concurrently with the linguo-didactic implications, it should be noted that the present study might offer avenues for further research that would involve a more substantial number of participants and, potentially, would lead to statistically significant results.Another avenue of future research would involve a juxtaposition of informal letters or e-mails written by the participants spontaneously and formal letters that would be written after the introduction of the topic of formal letter writing during an EFL course.

N
Descriptive Statistics of the Data Mean Values 1 Mean N of Words 315.3 (STD = 69.7) 2 Mean N of Sentences 14.2 (STD = 3.8) 3 Mean N of Paragraphs 5 (STD = 1.5)Explanation of the abbreviations: N = number; STD = standard deviation Santos (2002), 2010;Hyland, 1998;Kapranov, 2016)areholders are employed to support the CEOs' claims and draw the shareholders' attention to the conclusions that are desirable to the corporate management.Similarly,Garzone (2005)has found that DMs yet, although, and however in CEOs' letters to the stakeholders create a persuasive rhetorical effect.The use of DMs for rhetorical purposes in the British Petroleum CEO's letters to shareholders is reported inKapranov (2016).It has been found in the study that British Petroleum (BP) employs the elaborative DM and in order to facilitate "the effect of BP as a partner, an integral part in the triangle 'society -energy sector -BP', with every actor in this triangle being concerned by the challenged posed by climate change"(Kapranov, 2016, p. 225).Judging from the previous literature(Camiciottoli, 2010;Hyland, 1998;Kapranov, 2016), it can be generalised that DMs can be regarded as indexes associated with the genre of letter writing, especially a formal or business letter.This observation is further supported by DosSantos (2002), who suggests that a genre-based analysis of formal letters offers a valuable perspective on genre that, perhaps, should be taught to EFL students (Dos Santos, 2002, p. 187).Assuming that the role of DMs in formal letter writing has not been profoundly elucidated in EFL/ESP literature, it seems pertinent to explore how EFL students would use genre-appropriate DMs in the written tasks of formal letter writing within the context of their EFL/ESP course.In the subsequent sections of this article, I will present and discuss a study that aims at examining the use of English DMs in the written task of formal letter writing by a group of EFL students.
It should be perhaps reiterated that this study is embedded into an EFL university course offered at a major university in Sweden.This study does not involve a controlled experimental procedure.Instead, it reflects an actual classroom situation with the realistic EFL tasks that are executed within the framework of the course "Effective Communication in English (ENEC14)" (Stockholm University, 2017).This course aims to facilitate the students' ability to use English for academic and professional purposes (Stockholm University, 2017).The course involves weekly seminars throughout the semester that cover a range of genres of both written and spoken communication, e.g. business correspondence, reports, academic papers, oral presentations (ibid.).By the end of the course, students are expected "to explain the need for stylistic variation in different kinds of text; recognise various genres and be aware of their aims; and compare different text styles and genres" (Stockholm University, 2017).The books used during the course are Longman Academic Writing Series 4: Essays by Oshima and Hogue (2014), Mastering the Academic Word List by Schmitt et al. (2011), and Advanced Grammar in Use by

Table 2
Descriptive statistics of the data