Errors in the Translation of Discourse Markers from English into Persian in Movie Subtitles

Reza Kafipour

American Journal of Educational Research

Errors in the Translation of Discourse Markers from English into Persian in Movie Subtitles

Reza Kafipour

English Department, Shiraz University of Medical Sciences, Shiraz, Iran


Discourse markers are a part of language which are much neglected in the Persian language although they play a very important role in our daily conversations. This study aims at researching and investigating errors which occurr in the translation of discourse markers (DMs) from English into Persian in the subtitles of drama movies. To conduct this study, five movies were chosen. Each individual movie was selected from a single decade from 1970 onward. After the selection of the movies, six commonly used DMs which were mentioned in [6] were searched and obtained by the aid of computer softwares (AntConc© And adobe acrobat reader™). These DMs were categorized based on their functions in the original work. The obtained DMs in this study were analyzed carefully by the researcher and errors in translation were identified. The errors were categorized according to [6] model including omission, literal translation and translation by a different DM. The results and chi-square showed that omission is the most prevailing error made by Iranian translators in the translation of English subtitles into Persian followed by literal translation and translation to another DM. In today’s Persian language, using some expressions like midooni (I know), manzooram ineke (I mean) and etc. in today’s Persian language indicates that English DMs are being directly copied into Persian which may cause interference.

Cite this article:

  • Reza Kafipour. Errors in the Translation of Discourse Markers from English into Persian in Movie Subtitles. American Journal of Educational Research. Vol. 4, No. 15, 2016, pp 1100-1105.
  • Kafipour, Reza. "Errors in the Translation of Discourse Markers from English into Persian in Movie Subtitles." American Journal of Educational Research 4.15 (2016): 1100-1105.
  • Kafipour, R. (2016). Errors in the Translation of Discourse Markers from English into Persian in Movie Subtitles. American Journal of Educational Research, 4(15), 1100-1105.
  • Kafipour, Reza. "Errors in the Translation of Discourse Markers from English into Persian in Movie Subtitles." American Journal of Educational Research 4, no. 15 (2016): 1100-1105.

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

At a glance: Figures

1. Introduction

There are some elements of language which are usually unclear. These elements can be seen in language use, and most of the times, they are ignored linguistically and pragmatically. Regarding the dynamicity of language, the speakers and the listeners are always acting in an interrelated context. Language is not just syntax and semantics. Sometimes, some other elements should be taken into consideration. These elements are even more important and affect the discourse and consequently the speech. In fact, these elements play a very important role in our daily speech and we are not even conscious about using them. They can connect the speech parts and fill the gaps among the different parts of language. When we talk about pragmatics, more complex phenomena are popped up to the mind. Pragmatics is a level above syntax and semantics in the hierarchy of morphology. In order for the speech to be precise, the pragmatic elements should be pondered too.

Audiovisual translation makes the use of subtitling widely in order to transfer the message due to the fact that it is less costing. The textual version of the dialogues available in the films and TV programs is called subtitling. Subtitles are usually displayed at the bottom of the screen and they can be in the form of both translation and the original language. The purpose of the latter is to help people who have hard of hearing to understand what is going on. Having shared the characteristics of literary translation, the models used for assessing the translation of literary works can also be used for assessing the translation of subtitles ([2, 17, 36]).

The main function of audiovisual translation is to elicit the same response from the target language audience, as that which was elicited by audience from the source language audience and to preserve similar effect. Audiovisual texts are expressive texts which preserve different functions and actions. These texts are highly fabricated and pre-fabricated by their authors, both linguistically and semiotically. In fact, discourse markers are some units without which, spoken and/or written dialogues are incomplete and they cannot be indispensable to some ordinary components.

Discourse markers are a very ambiguous part of language. Some of them like well, right, and now can have some other functions. They can serve as adverbs, inserts and etc. So, identifying a discourse marker itself is a very challenging task, let alone translating it. Discourse Markers play a very important role in discourse, in order to preserve the its cohesion. They are used to convey the intended meaning of the speaker. But the problem which arises here is that whether they serve the same function in the target language when it is translated or not.

Every language uses its own discourse markers in order to preserve the cohesion of the discourse. Unfortunately, and to the best of the researcher's knowledge, only a few researches, if not say at all, has been conducted about the discourse markers in the Persian language. So, doing such a research about the translational procedures taken for the translation of the discourse markers in Persian language is deemed quite important. The results of this study can benefit the professional and freelance translators, to better understand the ways in which discourse markers are translated and help them get acquainted with the common errors while translating the discourse markers.

1.1. Objectives

The purpose of this study is to investigate the errors made by Persian translators in translating the discourse markers, with regards to [6] model of translating discourse markers.

2. Literature Review

So much have been written and said about translation since its golden age in 1960’s. Although the practice of translation dates back to first century BCE and in the works of Cicero and Horace, and then developed by St. Jerome (fourth CE), the translation studies have just merged as a new discipline after world war II [22]. The new concepts like translation workshops were introduced into this discipline. These workshops mostly dealt with literary translation (focus of this thesis) and were intended for the introduction of new translations into target cultures and to develop the principles of the new discipline.

By the development of this discipline, it became more and more of a science, a title given by Nida in his 1964 book, and then supported by other scholars such as Wills, Koller, and the Leipzig school scholars such as Kade and Nubert. Even the name of the new discipline was being determined, with candidates like translatology [22]. Like any other disciplines which have become a science, later on, the translation studies were approached academically too. So, more and more scholars tried to shed light on the today’s new science. One of the first efforts to shape it into an academic framework was that of Holmes and Toury. In his paper ‘the name and nature of translation studies’, [16] he tried to make the translation studies a unique and distinct discipline. It was an effort which was then appreciated by Gentzler and Snell-Hornby. [12] describes Holmes’ paper as “generally accepted as the founding statement for the field” and [31] agrees too. Indeed, they made this discipline (which was evolved from dispersed older disciplines) a distinct discipline, and gathered the old disciplines together.

2.1. Discourse Markers

During the past 20 years, the study of Discourse Markers (hereafter; DMs) has seen a dramatic growth. Many articles and books are published in the field, every year. The meaning of this term is so ambiguous that different scholars provide various meanings. So, it can be labeled by a variety of terms. They can be termed cue phrases [19], discourse connectives [5], discourse operators [27], discourse particles [30], discourse signaling devices [25], phatic connectives [4], pragmatic connectives [32, 35], pragmatic expressions [7], pragmatic formatives [8], pragmatic markers ([9, 10, 29]), pragmatic operators [1], pragmatic particles [23], semantic conjuncts [26], and sentence connectives [14].

Some of the more important definitions will be briefly elaborated here. One of the first references to DMs was that of [20], when they answered a question by Rhoda which was started with well. They discussed that well, as a discourse marker, points back to a previous topic which is a shared knowledge among the participants of that language. In fact, when it is used in initial position, it acts as a joint for an unstated topic in the discourse. Another account of DMs is that of Levinson who in his book, entitled Pragmatics, states that in every language there are some elements or structures which link an utterance to a prior discourse. For example, when words like but, therefore, in conclusion, well, beside and etc. are in the initial position in the utterance, most of the times, they seek to set such a relationship. All of those words share the property that at least one component of truth-conditional relations exists in them. In fact, they indicate that how an utterance, using these words, is related to a prior discourse or is the continuation of another prior discourse [20].

Zwicky is another important figure in the field of DMs who considers them as a class. He indicates that among the great group that is named 'particles', there is a significant class of words regarding the grammaticality, in English and other languages. These terms have been named discourse particles or interjections; however, he calls them discourse markers. Regarding the meaning, distribution and prosody, discourse markers can set a class. But again, it should be noted that they are independent words and not critics [37]. From Zwicky's statement, one may conclude that the DMs group still does not form a class. Theese markers are independent units of the sentence which usually operate at a pragmatic level and beyond the scope of function words. The position of DMs is usually at the beginning of the sentence and they serve as a joint in the conversation. DMs are prosodically separate and independent from their surrounding co-text by intonations, pauses, or even both. It should be noted that all discourse markers share a pragmatic function (e.g. when used in a discourse to join a previous utterance to a current one) and not solely a semantic meaning.

The first theoretical work and maybe the most detailed one in the field of DMs is that of [29] who speaks about some elements which are "sequentially-dependent units of discourse". These elements are termed 'discourse markers' by her and include expressions like you know, but, now, so, because, and, well, then, or and oh. She indicates that DMs cannot be classified linguistically and even paralinguistic and non-verbal gestures are possible DMs. She further elaborates that to delimite these markers in a class of linguistic conditions; first of all, the common characteristics of these items should be identified. Finding these shared and common characteristics in such complex language relations is a very difficult task in English. A large body of language which is typologically diverse should be analyzed in order to identify other ambiguous linguistic schemes, upon which these markers can be drawn. Most of the times, the discourse marker comes in the initial position in the sentence. It should be able to be used in different levels of discourse. From a syntactic point of view, it should be able to be detached from the rest of the sentence. It has to have several prosodic outlines. And finally, it should be able to operate at both global and local levels.

The scope of [29] work is somewhat narrow as she focused only on 11 expressions among which, just oh and well do not have a meaning, although she indicates that all the DMs have a 'core meaning'. To extend her idea of DMs, she suggests a number of other cases to be counted as DM; examples are, deictics such as ‘here’ and ‘there’, meta-talk such as ‘this is the point’ and ‘what I mean is’, interjections like ‘gosh’ and ‘boy’, perception verbs like ‘see’, ‘look’, and ‘listen’ and quantifier phrases such as ‘anyway’, ‘anyhow’, and ‘whatever’.

2.2. What Is Related by Discourse Markers?

Whatever discourse markers are called (discourse connectives, discourse operators, or cue phrases), they have one thing in common. They create a kind of association among the discourse segments to which they belong (S2) and the previous discourse segments (S1) [11]. In fact, they have a two-sided function, one lying in the first segment of the expression and the other in the second segment. However, there are some issues which should be mentioned about DMs. First, as [11] states "segments related by a DM need not be adjacent". The second issue is that a DM does not require introducing S2, but it is possible to appear in medial or final place as well. The third issue deals with the grammatical status of discourse markers. Sometimes, the DM joints two autonomous sentences, S1 and S2 while the position in which two autonomous clauses are connected by the assistance of a coordinate conjunction is another issue. Furthermore, as the DM is syntactically a subordinate conjunction, it is not able to establish a sentence which stands alone. In this case the previous autonomous phrase is needed to be presented [11]. Another problem is coherence. In order for the expression to be coherent, the interpretation of both segments (S1, S2) should be compatible with the DM used, not just their semantic readings.

2.3. Models of Analysis of Audiovisual Texts

Very few scholars and researchers have worked on the field of translation studies and its relation to audiovisual texts. So, the models which are going to be mentioned here are not primarily 'all' models which exist in the field. These models can be considered as some methodological insights in the field of audiovisual translation. [6] considers them as 'models'; because most of them end up as foundations for translational research.

The articles in Babel from 1960's initiated the researches in the field of audiovisual translation. These articles approached audiovisual texts from a translational view and primarily differentiated this type of texts, compared to other types of translation. They mainly focused on the constraints a translator faced when translating audiovisual texts into another language. The notable works which worth mentioning are those of [25], and [18]. The latter work is especially important since it "clearly notes the fundamental constraints that operate during the process of translation of audiovisual texts" [6]. The idea of translation priorities which presents the balance of the positions based on constraints, and adds another dimension to the translation of audiovisual texts, was introduced by [28] model.

Another approach which seeks to incorporate both translation theory and discourse analysis is the model which is based on the analysis of genres and types of audiovisual texts. The works by [13] and his focused works on a specific genre or [24] are worth mentioning. These studies mainly focus on the accordance of source texts to the audiovisual genre to which they belong. In fact, according to [13], the generic characteristics of the audiovisual text should be recognized and established. Pedagogical purposes should also be taken into consideration when we approach audiovisual translation. Such insights could be seen in the works of [3] who discuss the audiovisual translation from a didactic perspective. Putting aside the concepts of fidelity to the source text and studying audiovisual text, regarding the poly systemic methodology, the descriptive studies have also affected the field of audiovisual texts and their analysis.

Regarding the important works in the field, [6] names the works of [33] and [32]. These works impact the analysis of textual and linguistic elements of the target text. Besides, the macro-textual elements are introduced, too. These structures set the social, historical and economic framework within which the target text is composed and translation operations occur. These elements also contain the translated product which presents the ways in which translation is produced and the linguistic register uses the relationship between different genres and translations. In terms of historical studies, the works by [34] and [15] are so important. They both discuss macro-textual questions (economic, reasons of censorship and political issues) as well as Micro-textual questions.

2.4. Subtitling

Subtitling is the most widely studied audio visual translation mode. It comprises different subtypes, but it can be generally defined as the linguistic practice that shows a written text on screen purporting to convey the dialogues, the iconic elements or other material on the soundtrack such as songs [32]. The specialized characteristics —regarding the number of characters or the time of display— vary depending on the media used and the target public. The most common type is interlinguistic subtitling, usually displayed in open captions. As its name states, it involves the transfer from one language into another or other languages. Thus, it includes the bilingual subtitling —common in Finland or Belgium, for instance— where each block is made up of two lines, each in a different language. Thus, space constraints are even stricter in bilingual subtitling. Intralingual subtitling (often in closed captions) consists in the transfer of the dialogues into text in the same language, but with the necessary synchronization constraints. It is similar to the SDH, but not exactly the same: intralingual subtitling is aimed at L2 learners and people with slight listening disabilities, and does not involve extra linguistic features.

3. Methodology

3.1. Material

All of the movies chosen for this study were selected from among the drama films, in order for the genre to be specified. To study the discourse markers in the movies in a diachronic premise, five movies were chosen. All of these movies were released from 1970 onward. The first one was chosen from 1970's, second one from 1980's, third one from 1990's, fourth one from 2000's and finally the fifth one from the current decade. These movies were chosen from the five decades in order to study the development of discourse markers in a diachronic manner. It should be noted that the researcher was unbiased regarding the selection of the specific movies, and there were no special reasons behind the selection of the movies, except that they all belonged to drama genre. The movies include Taxi Driver (1973) written by Paul Schrader, Raging bull (1980) written by Paul Schrader, Shawshank Redemption (1994) written by Frank Darrabont, The Departed (2006) written by William Monnahan and Bad Teacher (2011) written by Lee Eisenburg and Gene Stupnitsky.

3.2. Data Collection

The researcher chose six discourse markers, namely you know, you see, look, oh, I mean, and now. The rationale behind choosing these discourse markers is that they are all among the ten highly used discourse markers according to Santa Barbara Corpus of Spoken American English. The more important reason is that these discourse markers are the only markers for which, the function, usage, and pragmatic meaning were provided in the exclusive work of [6]. This work is the benchmark for this study. One of the main concerns of this study, which was also a limitation upon it, was the matter of validity and accuracy of the subtitles. So, the researcher contacted the national TV of Islamic Republic of Iran, IRIB, to make sure if there are any official subtitles. Unfortunately, the answer was no, as they kept only the dubbed versions of the movies in their archive. So, in this study, the researcher tried to choose from among the best subtitles available on the internet sources. These websites are rated by Persian users as the most valid ones in terms of subtitles of the foreign movies. Moreover, in case of several subtitles for a single movie, the subtitle which has been highly rated by users was the one selected.

Finding the movies was a difficult task. The researcher found three newer movies in online sources and bid for them. But finding the two earlier movies was really troublesome as they were a bit outdated. All of the movies were bought and downloaded from online sources, and they were carefully viewed and analyzed by the researcher. The subtitles were downloaded too, and mounted on movies. The discourse markers and their translation were looked for, carefully. Although it was a very difficult task, the researcher found and downloaded the electronic versions in both PDF and TXT formats. The PDF format was required for adobe acrobat reader™ and wordsmith® which are computer programs, used for word processing. The TXT format was required for AntConc© which is a concordance software, and a big help to the study.

Firstly, the downloaded files from the internet were loaded in adobe acrobat reader™ and wordsmith®, in order to find the six discourse markers intended for the study. These DMs were typed in the search bar one by one and the programs did well in finding them in the whole text. About 180 discourse markers were found in these five movies. These samples were compared to their Persian equivalents which were either printed (in paper) or in electronic version, to see what errors have been made while translating them. To ensure that all of the discourse markers have been found, the TXT files were again loaded, this time in AntConc©. This software is a more comprehensive and advanced software for fulfilling the task, since it is able to search for case sensitive words and word clusters, providing the user with the page number and co-text (a couple of sentence before and after the highlighted word).

3.3. Data Analysis

The yardstick, against which the results of this study have been analyzed, is the invaluable work of [6]. He has analyzed six commonly used discourse markers, regarding their function and pragmatic features. But, it is important to note that he failed to propose an error categorization in terms of error analysis in the translation of DMs. He solely considers the deviation from the pragmatic meaning as an error and does not provide any solutions. In the present study, a categorization of the errors is offered under which, the mistakes during the process of translation of the DMs into Persian could be categorized. These errors are omission, literal translation and translation by a different DM. Regarding the statistical procedures, Chi-square was used to check if there was any significant difference in statistical values. The six discourse markers now, oh, you know, you see, look and I mean were examined in this study. ‘Now’ is commonly used as a marker of transition and has a deictic function to it. It indicates the speaker's progression through the discourse. The next DM is ‘oh’ which is sometimes used as a repair tool in the conversation. It may be used as a DM in question-answer adjacency pairs. Its third function is using it as an intensifier. Another DM researched in this study was ‘you know’ which could be used to express either confidentiality or shared knowledge. The next DM was ‘you see’ which functions just like ‘you know’. ‘Look’ has also two functions. It can be either used as a marker of digression or reference. And finally, the last DM investigated in this study was ‘I mean’ which is used for clarification. The following table could be a better illustration of the model used in this study:

Table 1. The Discourse Markers and Their Functions

4. Results

After identifying errors in translation of movie subtitles and categorizing them as explained earlier, the overall results of the data analysis could be best illustrated in Table 2 and Table 3 along with Figure 1.

Table 2. The Overall Distribution of Errors and Data Analysis

As depicted in Table 1, omission error (n=87) was the most frequently occurring error in the translation of movie subtitles from English into Persian followed by literal translation (n=31) and translation using a different DM (n=15). To find out if this finding is statistically significant, chi square was run as indicated in the following table.

If we take the probability level as p=0.05, there are huge deviations from this value, in terms of chi-square. This significant value shows that the results have not been obtained by chance and they are valid and reliable. There were three categories of errors defined for this study. The total number of errors was 133, which was divided by three, in order to obtain the expected frequency which was 44.3. The highest value belongs to omission. If we take the initial o for omission, the £ -o =24. This value proves that values in Table 2 are statistically significant. Therefore, it can be concluded for sure that omission is the most prevalent error made by Persian translators while translating discourse markers followed by literal translation and translation using a different DM. Furthermore, the following figure can illustrate the difference visually based on percentage of each error.

Figure 1. Percentage of Error Distribution

5. Discussion and Conclusion

The high percentage of omission in the error categorization indicates that Persian language is not yet a Discourse-marker-rich language. In fact, the Persian translator has no choice but omitting these discourse markers and not translating them from English into Persian. Most of the times, he/she has to omit it, because of the lack of appropriate equivalents or the lack of knowledge on the part of Persian translators. This fact proves that the most common error in translating discourse marker is omission. Literal translation is another common error, made by the Persian translators. This again reflects the limited number of discourse markers in Persian. Although it may seem trivial, but this very fact proves that the Persian translators have imported many discourse markers from English, directly and without any changes. The result is that we sometimes hear some expressions like manzooramine (I mean), midooni (you know), and etc. in today's Persian language.

The results confirm that movie translation is not as easy as it gets. Semiotic features of movie translation have made this genre a unique one. In the case of discourse markers translation, facial expressions, pauses, stress, intonation and gestures play an important role, and each can help the audience to better understand the message. The spectators, while watching a subtitled movie, have to synchronize the image and the text which distracts his/her attention. So, the dubbed version of a movie is far more desirable in terms of message transference. In that case, even omission may not count as an error, as the dubber sound and actor/actress image mix to convey the same meaning as intended in the original text.

The results of this study are useful for professional and freelance translators. They could get acquainted to the strategies and procedures used for translating discourse markers and common errors made while translating discourse markers. English language teachers can also be granted from the results of this study. They can teach discourse markers regarding the secondary literature represented in this study. Linguists can also use the highlights of this study, researching and investigating the development of discourse markers both in English and Persian, synchronically and diachronically.

The researcher selected 6 out of 10 discourse markers according to Santa Barbara Corpus of Spoken American English. Future studies can investigate all 10 discourse markers. Moreover, this study just investigated a limited number of movies which were selected. The other researchers are recommended to include an extensive number of movies to see if the results differ. On the other hand, considering mother tongue of the translators when the translator speaks Persian as the official language and some other local languages as Turkish and Kurdish may produce interesting and fruitful results.


This paper has been extracted from a master thesis conducted ar Islamic Azad University, Science and Research Branch, Marvdasht, Iran.


[1]  Ariel, M. (1994), Pragmatic operators: The Encyclopedia of Language and Linguistics, Oxford: Pergamon Press.
In article      PubMed
[2]  Baker, M. (1992), In Other Words. A Coursebook on Translation, London and New York: Routledge.
In article      View Article
[3]  Bartina, F. & Espasa, E. (2005), Audiovisual translation. In: Martha Tennet (Ed.), Training for the new millennium, (pp. 83-100), Amsterdam/ Philadelphia: John Benjamins.
In article      View Article
[4]  Bazanella, C. (1990), Phatic connectives as intonational cues in contemporary spoken Italian. Journal of Pragmatics, 14(4): 629-o47.
In article      View Article
[5]  Blakemore, D. (1987), Semantic constraints on relevance, Oxford: Blackwell.
In article      
[6]  Chaume, F. (2004), Discourse markers in audiovisual translating. Meta, 49, 843-855.
In article      View Article
[7]  Erman, B. (1992), Pragmatic expressions in English: A study of you know, you see, and I mean in faceto-face conversation. Stockholm: A lmqvistandWiksell.
In article      
[8]  Fraser, B. (1987), Pragmatic formatives, in: J. Verschueren and M. Bertuccelli-Papi, eds., The pragmatic perspective, 179-194. Amsterdam: Benjamins.
In article      View Article
[9]  Fraser, B. (1988), Types of English discourse markers. Acta Linguistica Hungarica, 38, 19-33.
In article      
[10]  Fraser, B. (1990), An approach to discourse markers. Journal of Pragmatics, 14, 12-14.
In article      View Article
[11]  Fraser, B. (1999), Contrastive discourse markers in English. To appear in: Y. Ziv and A. Jucker, edsPragmatics and beyond: Discourse markers. Amsterdam: Benjamins.
In article      
[12]  Gentzler, E. (2001), Contemporary Translation Theories: 2nd revised edition. Clevedon: Multilingual Matters.
In article      PubMed
[13]  Grice, P. (1999), Studies in the ways with words. Cambridge, MA: Harvard University Press
In article      
[14]  Halliday, M.&Hasan, R. (1976), Cohesion inEnglish. London: Longman.
In article      PubMed
[15]  Gutiérrez Lanza, C. (1999), Translation and censorship of cinematographic texts in Franco's Spain: dubbing and subtitling English-Spanish (1951-1975), Doctoral Thesis, University of León (Spain), Oct. 1999.
In article      
[16]  Holmes, J. S. (1988), Translated!: Papers on Literary Translation and Translation Studies, Amsterdam: Ropodi
In article      
[17]  House, J. (1977), A Model for Translation Quality Assessment, Gunter Narr Verlag: Michigan.
In article      
[18]  James, D. (1988), Some aspects of the syntax and semantics of interjection.
In article      
[19]  Knott, A. & Dale, R. (1994), Using linguistic Phenomena to motivate a set of coherence relations. Discourse Processes, 18, 35-62.
In article      View Article
[20]  Labov, W. and Fanshel, D. (1977). Therapeutic discourse. New York: Academic Press.
In article      
[21]  Levinson, S.C. (1983), Pragmatics, Cambridge: Cambridge University Press.
In article      
[22]  Munday, J. (2009), Routledge companion to translation studies, London and New York: Routledge.
In article      
[23]  Ostman, J. (1995), Pragmatic particles twenty years after. In B. Warvik, et al., Organization in discourse Proceedings.
In article      
[24]  Permpikul, M. (2001), Contrastive discourse markers in Thai, Boston Polanyi, The syntax of discourse, 3,261-270.
In article      
[25]  Polanyi, L. & Scha, R. (1983). The syntax of discourse. Text, 3,261-270. Quirk, R, Greembaum, S. Leech, J & Svartvik, J. (eds.). a comprehensive grammar of the English language. London: Longman
In article      View Article
[26]  Quirk, R, Greembaum, S. Leech, J & Svartvik, J. (1985), a comprehensive grammar of the English language, London: Longmanniver sity.
In article      
[27]  Redeker, G. (1995), Ideational and Pragmatic markers of discourse structure. Journal of Pragmatics, 14, 367-381.
In article      View Article
[28]  Rouchota, V. (1996), Discourse markers: what do they link. In: John Harris and Philip Black, eds., UCL Working Papers in Linguistics, 8, 199- 214.
In article      
[29]  Schiffrin, D. (1987), Discourse markers, Cambridge: Cambridge University Press.
In article      View Article  PubMed
[30]  Schorup. (1985), Common discourse particle in English conversation: Like, well, y'know, NewYork: Garland.
In article      
[31]  Snell-Hornby, M. (2006), The Turns of Translation Studies: New Paradigms or Shifting Viewpoints, Amsterdam: John Benjamins.
In article      View Article
[32]  Stubbs, M. (1983), Discourse analysis, The University of ChicagoPress: Chicago
In article      
[33]  Su, J-C. (2004), Contrastive discourse markers in Mandarin. Unpublished manuscript, Boston University.
In article      
[34]  Sweetser, E. (2000), From etymology to pragmatics, Cambridge: Cambridge University Press.
In article      
[35]  van Dijk, T. A. (1979), Relevance assignment in discourse comprehension. Discourse Processes, 2, 113-126.
In article      View Article
[36]  Waddington, C. (2001), Different Methods of Evaluating Student Translations: The Question of Validity, Translators' Journal, 46, 311-325.
In article      View Article
[37]  Zwicky, A. (1985), Clitics and particles, Language, 6, 283-305.
In article      View Article
  • CiteULikeCiteULike
  • MendeleyMendeley
  • StumbleUponStumbleUpon
  • Add to DeliciousDelicious
  • FacebookFacebook
  • TwitterTwitter
  • LinkedInLinkedIn