Introduction

Even though writing is considered as a complicated and difficult task, being a writer is an important journey. The latter helps writers to discover themselves, their ideas, and the world around; and it creates their professional identities as academics, researchers and scholars (Murray & Moore, 2006). Nunan (1991) states that writing is not only based on connecting words together, but also it is a resulting product of this process. It is controlled by the context conditions in which it occurs (audience & genre). In academia, writing a discourse does not only rely on constructing correct grammatical sentences, but also demands the construction of a cohesive text which requires the use of conjunctives or discourse markers (DMs) that establish a logical link between the text parts (Hamed, 2014).

Discourse markers are linguistic items such as ‘furthermore’, ‘because’, ‘however’ and ‘then’ which construct links between discourse segments. Their importance stems from their role in creating cohesive and coherent texts. In academic writing, discourse markers aid in producing an effective piece of discourse and facilitate the communication between readers and writers. The misuse of DMs can affect directly the communicative process and may lead to misunderstanding the writer’s thoughts (Rahimi, 2011).

In the postgraduate context in Algeria, writing a research paper in English is a challenging task. Postgraduate students confront difficulties when they produce their texts in a foreign language. Since French is the language of instruction at the Algerian university, it is very difficult for postgraduate students in scientific branches to produce an accurate written discourse in English. Discourse markers’ use in this kind of texts is a significant problem in this case. It is an issue that may face those students as they are unaware of academic writing’s basics. In the present study, the randomly selected students belong to scientific streams. They are considered as intermediate in English academic writing skills. For this reason, studying DMs’ use in their discourse can be a worthy attempt to determine the quality of their writing. Therefore, the present paper tries to describe and analyze the extent to which these postgraduate students utilize DMs in their academic writing in English. For this reason, the two researchers try to answer the following research questions:

• To what extent do postgraduate students use DMs in their academic writing in English?

• Which kind of DMs do postgraduate students use?

• How do postgraduate students utilize DMs in their research papers?

1. Literature Review

Various researches were conducted in the last years in order to explore and study the nature of discourse markers as important linguistic items. They casted light on those expressions from different angles. The focus of the previous studies was on the meaning and the function of discourse markers in different texts. Halliday & Hasan (1976) state that conjunctive elements are “cohesive not in themselves but indirectly, by virtue of their specific meanings; they are not primarily devices for reaching out into the preceding (or following) text, but they express certain meanings which presuppose the presence of other components in the discourse” (p. 226). They focus mainly on the semantic meaning of DMs. For this reason, they classify conjunctive elements into four groups in which each group construct a conjunctive relation, additive, adversative, causal and temporal.

Schiffrin (1987) shifted the study of DMs from the semantic to the pragmatic focus. She examined discourse not only as a stretch of language, but also as a form of social interaction. For that reason, she tended to adopt both quantitative and qualitative procedures, and combine concepts and modals that are innate from linguistic and sociology to analyze those items. She, in this sense, defines DMs as “sequentially dependent elements that bracket units of talk’’ (p.55); i.e., they are utterance-initial connectives that work in relation with proceeding talk or text.

Furthermore, Fraser (1999) describes DMs as “a class of lexical expression drawn primarily from the syntactic classes of conjunctions, adverbs, and prepositional phrases” (p. 1). He states that the function of these items is to create a kind of relationship between two segments, a prior segment S1 and an introducing one S2. Their meaning is pragmatic rather than conceptual since it is negotiated by their context. In his research, he creates a new taxonomy of DMs in which he categorizes them into four groups, elaborative, contrastive, inferential and causative. Elaborative markers elaborate a parallel relationship between the message segments. Contrastive DMs, however, contrast the two parts. Inferential and causative markers describe an inference and reason relation between the message parts respectively. Fraser (2005) suggests a new subclass or temporal DMs which create a time relation between segments. Furthermore, Fraser (2009) focuses on the three primary main classes (elaborative, contrastive and inferential) to study discourse markers.

Various works were achieved on the use of DMs in both spoken and written discourse. Matinez (2004) conducted a study about the use of DMs in the expository compositions of Spanish undergraduates. She adopted Fraser’s taxonomy (1999) as a model of analysis. By using a quantitative procedure, Martinez constructed a relation between the frequency of DMs and the quality of written discourse. Similarly, Jalilifar (2008) investigated DMs in descriptive composition of 90 Iranian English language students. He asserted Martinez’s results (2004), and distinguished between graduate and undergraduate students’ academic writing. On the other hand, Castro (2009) worked on spoken discourse in which she casted light on the use and function of DMs in EFL classroom discourse or interaction. She determined the occurrences, the frequencies, and the function of DMs when they are used by non-native teachers. Her work results revealed the importance of those markers in accomplishing the textual and interpersonal functions that are involved in increasing the coherence of classroom discourse.

Moreover, Modhish (2012) achieved a research on DMs use in Yemeni EFL learners’ compositions. He adopted Fraser’s taxonomy (1999) as a model to analyze his corpus. His findings showed a less strong relation between DMs’ frequency in learners’ papers and their writing quality. However, Ali & Mahadin (2016) utilized a comparative study to analyze the written discourse of intermediate and advanced Jordanian EFL learners. They relied on Fraser (2009) as a model of analysis. Their study outcomes revealed the difference between advanced and intermediate use of DMs in terms of frequency and the variety of use in each subcategory of those markers. Also, Tadayyon & Frahani (2017) explored DMs in academic papers. In the same line with Ali & Mahadin (2016), they adopted also a comparative study to investigate DMs’ use in published academic papers of Iranian and English postgraduate students. The work results indicted the high rate of DMs in Iranian students’ writing in comparison with their English counterparts’. These outcomes emphasized the previous studies results concerning the frequency of DMs and the quality of the written discourse.

The review of the existing literature review shows that the majority of previous studies concerning this topic focus on EFL learners as the main sample of study. They tackle mainly the quantitative side of the issue which is represented by the frequency of DMs in the selected corpus and forgetting about analyzing the way those linguistic items are used. The present study is an endeavor to describe the academic written discourse of postgraduate students in scientific streams. This will be made by detecting the use of DMs which are the elements that ensure the coherence of any written discourse and determine its quality.

2. Methodology

The present work aims to investigate the use of discourse markers in postgraduate students’ academic writing. In order to accomplish the previously mentioned objective, seven (07) research papers are selected randomly as the corpus of the study. They have been written by postgraduate students enrolled in scientific streams at Kasdi Merbah University- Ouargla, Algeria. These students belong to four (04) scientific branches, namely, Applied Biology, Procedures Engineering, Electrotechnical Engineering and Informatics. All of them took technical English courses in ‘Licence’ and Master Degrees. To carry out this study, two methods are adopted: quantitative and descriptive. The former is designed to determine discourse markers’ frequency. It is very useful since it is based on the measurement of quantity (Kothari, 2004). In this case, the two researchers rely on SPSS (Statistical Package for the Social Sciences) version 19 to analyze data and design diagrams. The descriptive method, on the other hand, is adopted in order to report and describe the occurring of DMs in the target research papers and to determine the way they are utilized by the chosen sample.

In order to classify discourse markers into subclasses, the two researchers rely on Fraser’s taxonomies (1999, 2005, and 2009) as models of the study. It is assumed that they are the most comprehensive to represent and describe DMs since Fraser has categorized them into functional categories. According to Fraser (1999), DMs can be elaborative such as ‘and’, ‘besides’, ‘in addition’, ‘also’, ‘as well as’, ‘as well’, and ‘correspondingly’; contrastive such as ‘however’, ‘but’, ‘in contrast’, ‘although’, ‘whereas’, and ‘nevertheless’; or inferential like ‘so’, ‘therefore’, ‘in conclusion’, ‘as a consequence’, ‘for that reason’, and ‘because of’. Also, he adds the forth category that includes other DMs such as ‘because’, ‘since’, and ‘as’. Besides the previous subclasses, Fraser (2005) suggests temporal DMs like ‘then’, ‘after’, ‘before’, ‘as soon as’, ‘when’ and ‘meanwhile’ which are removed later in Fraser’s taxonomy (2009).

3. Findings and Discussion

After analyzing the collected data, the results indicate the occurrence of 381 DMs in the selected postgraduate students’ research papers. These DMs are subdivided into five categories:

Figure 1: Discourse Markers’ Classes

Figure 1 represents the frequency of DMs’ categories in the selected research papers. It reveals that elaborative DMs are the most frequent (41.1%) followed by inferential (22.1%), contrastive (19.7%), and temporal DMs (12.1%). The last subcategory, causative DMs, (5%) has the lowest frequency.

In addition, as it is mentioned in Figure 2 below, postgraduate students utilized different elaborative DMs. ‘And’, ‘in addition’, ‘also’, ‘furthermore’, ‘indeed’, ‘in other words’, ‘as well as’, ‘as well’, ‘or’, ‘for example’, ‘that is to say’, and ‘in fact’ are the most frequent markers in this category. ‘And’ (42.3%) and ‘also’ (23.1%) have the highest frequency. ‘Indeed’ (8.3%), ‘in addition’ (7.7%), ‘as well as’ (7.7%), ‘as well’ (3.2%), ‘similarly’ (2.6%), ‘that is to say’ (1.3%) and ‘in other words’ (1.3%) represent the average in the corpus. However, ‘in fact’, ‘for example’, ‘or’ and ‘furthermore’ seem to have the least percentage (0.6%):

Figure 2: Elaborative Markers

Figure 3 below describes the occurrence of inferential DMs in the target articles in which ‘due to’ (29.4%) and ‘because of’ (15.3) are the most frequent. Whereas, ‘so’ (9.2%), ‘thus’ (8.2%), ‘therefore’ (8.2%), ‘in this case’ (7.1%), ‘consequently’ (5.9%) and ‘hence’ (5.9%) are in the middle. In addition, markers such as ‘for that reason’ (3.5%), ‘so that’ (2.4%), ‘in conclusion’ (2.4%) and ‘then’ (2.4%) can be found in the chosen papers with low frequencies:

Figure 3: Inferential Markers

Based on the results that are mentioned in Figure 4, it is very clear that ‘however’ (28%), ‘on the other hand’ (24%), and ‘but’ (20%) are the most frequent contrastive DMs. Also, ‘in comparison’ has an average recurrence (9.3%). However, the low score in this subcategory is represented by ‘whereas’, ‘even though’, ‘although’ with (4%), ‘while/whilst’ (2.7%), and ‘nonetheless’, ‘nevertheless’ and ‘in contrast’ (1.3%).

Figure 4: Contrastive Markers

As it is observed in the graph below (Figure 5), temporal DMs are represented by different markers. ‘Then’ (23.9%), ‘after’ (23.9%), ‘when’ (20%), and ‘while’ (13%) show a high frequency which are followed by ‘finally’ (8.7%) with an average one. However, ‘firstly’ (4.3%), ‘before’ (2.2%) and ‘subsequently’ (2.2%) have the least percentages in this subclass:

Figure 5: Temporal Markers

Figure 6 below shows the three main DMs in this subcategory in which ‘because’ (57.9%) represents the highest frequency followed by ‘as’ (36.8%) and ‘since’ (5.3%) as less frequent markers:

Figure 6: Causative Markers

The research outcomes reveal the presence of 381 DMs that are used by postgraduate students in their academic papers. This high rate indicates the overuse of DMs; on one hand, and the misuse of them on the other in the target corpus. This can be explained by the lack of native-like competence as those students are non-native speakers. It also shows the lack of academic writing awareness among the majority of the target sample. After the categorization of the selected DMs, it seems that elaborative markers (41.1%) are the most frequent. Martinez (2004) explains that it is necessary in expository writings to construct and elaborate ideas and thoughts which hinge on the utilization of parallel links between segments that are indicated by elaborative markers. In addition, the inferential (22.1%), the contrastive (19.7%), the temporal (12.1%) and the causative DMs (5%) represent the less frequent markers. This result reveals the difficulty to utilize those markers because of the limited background knowledge which leads those postgraduate students to avoid them in their writing.

Furthermore, the obtained results indicate the variety of DMs’ use in each subclass. In the elaborative category, the use of ‘and’ (42.3%) and ‘also’ (23.1%) is noticeable in which they represent the most frequent markers. Modhish (2012) notes that the overuse of these two markers is due to the tendency of avoiding making errors. This can be explained by the transfer of L1 since both ‘and’ and ‘also’ are frequent in Arabic written discourse. In addition, the analysis of inferential markers in the selected papers shows that postgraduate students rely intensely on ‘due to’ (29.4%) and ‘because of’ (15.3) when they construct an inference relationship between thoughts. Since ‘due to’ and ‘because of’ have the same meaning, the chosen students deal with them as synonyms. The intensive use of these DMs creates a redundant composition. The obtained outcomes are in line with Ali & Mahadin (2016) who observes the overuse of the two markers in intermediate Jordanian students’ L2 writing. Moreover, contrastive DMs stand in the third place in which their use is restricted in ‘however’ (28%), ‘on the other hand’ (24%), and ‘but’ (20%). In this case, it is very clear that those postgraduate students are not aware of the function of other contrastive DMs. Temporal and causative markers are represented as the less frequent categories adopted by the selected corpus respectively. The former is characterized by the overuse of ‘then’ (23.9%), ‘after’ (23.9%), ‘when’ (20%), and ‘while’ (13%) since they are the most useful logical connectives in the scientific discourse that is based necessarily on different procedures’ steps. On the other hand, causative markers reveal the dominance of ‘because’ (57.9%) as the highest score. It is a logic result since most postgraduate students are not familiar with the function of other markers when they construct a causal relation between segments.

The two researchers also detected the misuse of some DMs in the target research papers. The results are presented in the selected examples that are extracted from the selected corpus:

1. “And the most challenge in this system is only matching the sun dependent intermittent and diurnal power supply with the time dependent power demand of the residence.” (Appendix C-Paper 4)
   In this example ‘and’ is used at the beginning of the sentence which is incorrect, while, instead of that, the student is able to use ‘in addition’ as a relevant DM.

2. “Single diode model is widely used due to having precise simulation accuracy and calculation simplicity and fairly emulates the PV characteristics.” (Appendix C-Paper 4)
   It is obvious that there is a causative relationship between the two segments. For that reason, it is more suitable to use ‘since’ instead of ‘due to’ to express a reason relationship.

3. “To prove that the used model is valide for different condition, the tests on the panel were to fix the temperature for differnet solar irradiance level and on other hand the irradiance was fixed for averge of temperature in fig.10 to fig.13.” (Appendix C-Paper 3)
   The example above represents a combination of two types of DMs, elaborative and contrastive. It seems to be wrong since the first is used to elaborate the relation between segments and the second to contrast it. Thus, it is explained as a kind of L1 transfer.

4. “Packaging materials play an important role in preventing the deterioration of food quality due to adverse environmental influences such as microbial contamination, oxygen and moisture and also serve to prolong the shelf life of the product. packed up.” (Appendix C-Paper 5)
   As it is observed in (d), the participant combines two discourse markers of the same subclass (elaborative). In this case, ‘and’ and ‘also’ serve the same function which creates a redundant expression.

5. “The maximum ambient temperature is found at 13 h30 min with 28 °C, the irradiance value was E=557W/m²on the other hand, the cell temperature was 38°C.” (Appendix C-Paper 3).
   In this example, the student tends to elaborate not to contrast ideas; thus, ‘on the other hand’ is inappropriate. In this case, the use of the elaborative marker ‘and’ is the most appropriate.

6. “ But these connections also expose lots of sensitive data.” (Appendix C-Paper 1).
   The example (f) is not correct since it is inappropriate to put ‘but’ at the beginning of the sentence as it functions as a coordinative conjunction which has related sentences. ‘However’, in this case, can be used.

7. “The pre-trained model can then be used to detect and thus prevent unwanted attacks or intrusion in the IoT system.” (Appendix C-Paper 1).
   In this expression, the paper’s writer combines two markers (and/thus) which creates a kind of redundancy since in this case ‘and’ is sufficient to convey the intended message.

Conclusion

The main objective of the present paper was to investigate the use of discourse markers in postgraduate students’ academic writing. These students are required to publish their papers in English and they sometimes face some difficulties when they use DMs in writing a coherent discourse. This problem can affect the quality of their research papers and their accuracy. The obtained results of the present study revealed the overuse of DMs in the selected corpus. Also, they indicated the high frequency of elaborative markers since they are the most appropriate to elaborate ideas and thoughts in this kind of discourse. Inferential, contrastive, temporal and causative; on the other hand, are the least frequent because of the limited background knowledge on those DMs and their functions. In addition, postgraduate students tended to use a variety of DMs in each subclass that is characterized by the overuse of some markers at the expense of others. This can be explained by the misuse of those items, L1 interference, and the ignorance of some DMs’ functions. Consequently, the paper’s outcomes interpreted the quality of the selected papers and the proficiency level of postgraduate students in academic writing in English. The present research work recommends implementing teaching the basics of academic writing in English in the scientific major classes. Also, it suggests a well-designed syllabus of academic writing for postgraduate scientific major students’ curriculum during their first academic year of their doctoral studies which may broaden their knowledge of DMs and may help them to write a cohesive and a coherent discourse.

Academic writing is often considered as a challenging task. Discourse markers are also significant items that ensure the accuracy and the coherence of any text. The misuse or the overuse of these items can affect the quality of the written discourse. Thus, the present paper aims at investigating the use of discourse markers in postgraduate students’ academic writing. For this reason, the two researchers select randomly seven (07) academic papers in English as a corpus of the study. These chosen papers are written by postgraduate students enrolled in scientific branches at Kasdi Merbah University-Ouargla- Algeria. They belong to four branches, namely Applied Biology, Procedures’ Engineering, Electrotechnical Engineering and Informatics. In order to accomplish the paper’s objective, quantitative and descriptive methods are adopted. In addition, in order to categorize discourse makers in subclasses, the two researchers rely on Fraser’s taxonomies (1999, 2005 & 2009) as models. The obtained results reveal the high frequency of elaborative, inferential and contrastive markers. Furthermore, they show the overuse and the redundancy of some discourse markers in the in the selected corpus. These outcomes may explain the proficiency level of the target sample and determine their discourse quality.

Academic writing, discourse markers, elaborative markers, inferential markers, contrastive markers, postgraduate students, written discourse.

La rédaction académique est souvent considérée comme une tâche difficile. Les marqueurs de discours sont également des éléments significatifs qui assurent la fiabilité et la cohérence de tout texte. Le mésusage ou la surutilisation de ces éléments peut affecter la qualité du discours écrit. Ainsi, le présent article vise à enquêter sur l’utilisation des marqueurs de discours dans l’écriture académique des étudiants de troisième cycle. De ce fait, les deux chercheurs sélectionnent au hasard (07) sept articles académiques en anglais comme corpus de l’étude. Ces articles sont rédigés par des étudiant de troisième cycle inscrits dans les filières scientifiques de l’Université Kasdi Merbah –Ouargla-Algérie. Ils appartiennent à quatre branches, à savoir la biologie appliquée, génie de procédées, génie électrique et l’informatique. Afin d’atteindre l’objective de l’étude, des méthodes quantitatifs et descriptifs sont adoptées. De plus, afin de catégoriser les marqueurs de discours en sous-classes, les deux chercheurs s’appuient sur les taxinomies de Fraser (1999, 2005, 2009) comme modèles. Les résultats obtenus révèlent la fréquence élevée des marqueurs élaboratifs, inférentiels et contrastifs. De plus, ils montrent la surexploitation et la redondance de certains marqueurs discursifs dans le corpus sélectionné. Ces résultats peuvent expliquer le niveau de compétence de l’échantillon cible et déterminer la qualité de son discours.

Ecriture académique, marqueurs de discours, marqueurs élaboratifs, marqueurs inferentiels, marqueurs contrastifs, doctorants, discours écrit.

تعتبر الكتابة الأكاديمية إحدى أصعب المهمات, كما أن علامات الخطاب تعد عناصر مهمة تضمن دقة وتماسك أي نص. يمكن أن يؤثر سوء استخدام هذه العناصر أو الإفراط في ذلك على جودة الخطاب المكتوب وبالتالي تهدف هذه الورقة إلى التحقيق في استخدام علامات الخطاب في الكتابة الأكاديمية لطلاب الدراسات العليا. لهذا السبب تم اختيار (07) أوراق بحثية باللغة الإنجليزية كمجموعة الدراسة. تمت صياغة هذه الأوراق من قبل طلبة الدراسات العليا (ما بعد التدرج) المسجلين في الفروع العلمية بجامعة قاصدي مرباح ورقلة. ينتمي هؤلاء الطلبة إلى أربعة فروع وهي البيولوجيا التطبيقية, هندسة الطرائق, الهندسة الكهربائية والمعلوماتية. ومن أجل تحقيق هدف هذه الورقة تم اعتماد الأساليب الكمية والوصفية لتحليل النتائج. إضافة لذلك من أجل تصنيف علامات الخطاب تم الاعتماد على تصنيفات فريزر (1999,2005,2009) كنماذج. وقد كشفت النتائج عن التكرار العالي للعلامات التفصيلية, الأستنتاجية و التباين, علاوة على ذلك فهي تظهر مدى الاستخدام المفرط والتكرار لبعض علامات الخطاب في الأوراق المختارة مما يفسر مستوى إتقان العينة للكتابة الأكاديمية باللغة الإنجليزية و تحدد جودة خطابهم.

الكتابة الأكاديمية, علامات الخطاب, علامات تفصيلية, علامات استنتاجية, علامات التباين, طلاب الدراسات العليا, خطاب مكتوب.