Pragmatic functions of I in academic discourse: linguistic research articles

Pragmatic functions of Iin academic discourse: linguistic research articles Veronika Raušová (Praha) ABSTRACT This study presents comparative research focused on the frequency and usage of the 1st person singular pronoun Iin linguistic research articles (RAs) written in English by native speakers of English and native speakers of Czech. Two specialized corpora, together comprising 80 RAs, were compiled for the purposes of this study. The study shows that in comparison to previous research, the use of Igradually increases in RAs of Czech authors but the device is still underused compared to RAs of native English speakers. The underuse is linked to longstanding traditions of Czech academic writing. Moreover, the pragmatic functions of propositions featuring Ias an explicit authorial reference in the discourse (Stating opinions and claims, Discourse organisation and guidance, Research process recounting) were linked to the tenets of the Politeness theory, to show benefits and risks of exploiting implications that they carry. KEYWORDS comparative study, explicit authorial reference, personal pronoun I, politeness theory, research articles 1. INTRODUCTION Writing asuccessful research article (RA) is askill every researcher in any academic field should strive to master because even the best research may be rendered worthless when it is not communicated appropriately to the intended audience. The purpose of RAs is rarely to simply inform using cold hard facts. Their purpose is mostly to negotiate claims with the intended readership to persuade them of their validity. In order to do that, the authors need to establish interaction between themselves and the readers. One way to achieve acertain level of writer-reader interaction is to employ the 1st person singular personal pronoun Ias the means of introducing authorial voice into their writing. This pronoun and its pragmatic roles within the academic discourse are the focus of this study. Iaim to compare the ways and frequency of its use by native English authors of RAs in the field of linguistics and native Czech authors writing linguistic RAs in English. Furthermore, Iwould like to relate the pragmatic roles to the Politeness theory (Brown and Levinson 1987) and assess the effects they might have on the faces of the parties participating in the interaction. In this way, Iwould like to explain not only in what manner and how often the authors decide to use explicit authorial reference but also what probably motivates them to choose personal reference over other strategies and what are the risks and benefits they must bear in mind when employing it. The assumption that there are differences between the two groups of authors (either quantitative or qualitative) in their use of the personal pronoun Iis based VERONIKA RAUŠOVá 169 on the distinct conventions of academic writing from which the authors draw and the assertion that “the English texts of non-native speakers of English are inevitably influenced by the traditions of academic writing in their native tongue (DontchevaNavratilova 2014a: 43)”. Finally, the importance of this comparison lies in the fact that the English language is “the lingua franca of science in the international context (Dontcheva-Navratilova 2014a: 50)” and in order for the non-native authors writing in English to persuade Anglophone academia of the validity of their work, they need to satisfy the audience’s expectations when it comes to the conventions of Anglophone academic writing. The following section will offer amore detailed account of the existing conventions of Anglophone and Czech academic writing and their influence on the usage of the pronoun Iin research articles. 2. CONVENTIONS OF ACADEMIC WRITING There are different conventions governing the way an academic text should be structured and the way information should be conveyed to the readership stemming from traditional approach to academic writing based on language communities (e.g. Vassileva 1998; Tang and John 1999; Dontcheva-Navratilova 2013, 2014a, 2014b) and also from common practice in aparticular discipline (e.g. Kuo 1999; Hyland 2008). Since Iam comparing RAs within the field of linguistics, the interdisciplinary differences are not relevant here. The focus is on the distinct features of Anglophone and Czech traditions. Anglophone academic writing is described as featuring marked authorial presence, being interactive and dialogic, exhibiting strict discourse norms, employing explicit discourse organisation and overall being reader-oriented. Czech academic writing, in comparison, is characterized by backgrounded authorial presence, monologic discourse with afocus on terminological and conceptual clarity, by the lack of explicit discourse organisation and by being primarily oriented on the writer. These differences, with afocus especially on the way Czech authors establish writer-reader interaction have influence on the features they employ in their RAs written in English (Dontcheva-Navratilova 2014b: 41). The preference to background authorial presence and focus on the topic rather than on the interaction with the reader could be the main reasons why “the frequency of occurrence of personal attribution decreases towards the east to German or Czech academic communities and why it is generally higher in the Anglophone academia (Čmejrková et al. 1999: 47)”. In addition, Chamonikolasová notes that “In describing their research methods, or their intentions and conclusions, Anglophone authors usually indicate their identity by the use of pronouns of the 1st person singular and (that) the use of the 1st person singular in Czech texts is generally uncommon; it occurs only in more recent publications (Chamonikolasová 2005: 82)”. The way authors employ the pronoun Ito explicitly mark their presence within the discourse results in anumber of pragmatic functions, which Iwill describe in the next section. 170 LINGUISTICA PRAGENSIA 2/2018 3. THE PRAGMATIC ROLES OF I The use of personal pronouns in academic texts allows the authors to break out of the constraints of the impersonal style, which is still generally favoured in academic discourse since it lends the propositions an air of objectivity and grants the authors some distance from their claims and stances. Personal reference forces the authors to take personal responsibility for their propositions and usually conveys “the desire to both strongly identify oneself with aparticular argument and to gain credit for an individual perspective (Hyland 2005: 181)”. This represents the motivation why an author would use explicit authorial reference in atext. Moreover, each use of Ican be analysed as having an additional pragmatic function. The taxonomy of functions used for the purposes of the present study is based on Harwood’s brief recount of functions that include: organizing text and guiding the reader through the argument, stating personal opinions and knowledge claims, and recounting experimental procedure and methodology (Harwood 2005: 344). These functions are included in the widely employed Tang and John’s (1999) model in Figure 1 to show how they relate to it. Even though the present classification is rather simple in comparison to other taxonomies (Ivanič 1998; Hyland 2002; DontchevaNavratilova 2013) it is sufficient for the purposes of the present study. No ‘I’ ‘I’ as representative ‘I’ as guide ‘I’ as architect ‘I’ as recounter of research process ‘I’ as opinion holder ‘I’ as originator Discourse organisation and guidance Recounting research process Stating opinions and claims Least powerful authorial presence Most powerful authorial presence Figure 1.Atypology of possible identities behind the 1st person pronoun in academic writing (Tang and John 1999: S29) with the current classification in bold type. In the following section, Iwill briefly describe the three categories of the proposed taxonomy. At this point, it is important to stress that the pronouns cannot be analysed in isolation and that the decisive factor in determining the pragmatic function of aconstruction with asubjective pronoun heavily depends on the context in which it appears. Also, while the categories below are presented as distinctive, it is also important to note that the pragmatic functions may overlap. VERONIKA RAUŠOVá 171 3.1 STATING OPINIONS AND CLAIMS Hyland notes that the decision to use Iin aproposition conveying opinions and claims “is aclear indication of the perspective from which astatement should be interpreted, enabling writers to emphasize their own contribution to the field and to seek agreement for it (Hyland 2005: 181)”. (1) Iargue that their treatment is superficial because, despite appearances, it relies solely on asociological, as opposed to an ethical, orientation to develop aresponse. (Hyland 2005: 181) 3.2 DISCOURSE ORGANISATION AND GUIDANCE The second category comprises cases where the author embodies the roles of an architect and aguide (see Tang and John 1999: S28) who accompanies the reader throughout the text. The implications of an overt authorial presence in this category is to establish ashared position with the reader within the time and place of the particular RA and offer aguided tour through the text, highlighting important information, drawing the reader’s attention where necessary, and preparing the ground for arguments by indicating where and when the argumentation will occur. (2) In this essay, Iwill discuss the bastard status of English from the pre-English period (– AD 450) to Middle English (c.1100–1450) (Tang and John 1999: S28) 3.3 RECOUNTING RESEARCH PROCESS The last category covers the propositions where the authors “describe or recount the various steps of the research process (Tang and John, 1999: S28)”. While the present category might look similar to the category of Discourse organisation and guidance, there is one crucial feature that distinguishes them. While the previous category includes propositions conveying what the author is doing, has done, or will do within the text, the current category subsumes propositions recounting what the author did outside of the text. (3) To trace these areas Ifirst conducted an analysis of 139 papers on the Web of Science which included ‘metadiscourse’ among their keywords using the visualising analysis programme CiteSpace (ENCORP_PR256) Having described the taxonomy of functions used in this study, Iwould like to note that each time the authors choose to use explicit authorial reference in their texts, they inevitably enter apragmatic framework with specific implications. Iimplies adirect involvement of the author and often also an implicit involvement of the reader, whether it is used in aproposition expressing the author’s opinions or claims (1), in aproposition serving to organize discourse (2) or aproposition referring back to the research process (3). The direct involvement of the author further implies facethreatening acts (FTAs) both to the writer’s and the reader’s faces as well as negative politeness strategy towards the reader, while the implicit involvement of the reader 172 LINGUISTICA PRAGENSIA 2/2018 implies employment of apositive politeness strategy towards the reader. Iwill elaborate on this notion in the next section. 4. POLITENESS THEORY Based on the Politeness theory developed by Brown and Levinson (1987) the authors, as well as the readers, have positive and negative faces, which is afigurative expression for their desire to be agreed with and to be unimpeded by others, respectively (Čmejrková et al. 1999: 53). For my purposes here, Iwill interpret the face wants of the participants in interaction in academic discourse in the following way: — Writer’s positive face is the desire to be agreed with — Writer’s negative face is the desire to avoid criticism — Reader’s positive face is the desire to participate — Reader’s negative face is the desire to decide independently Myers (1989) was first to apply Brown and Levinson’s model to written texts and academic writing in particular, in order to explain the interactions between writers and readers. He argued that “scientific discourse consists of interactions among scientists in which the maintenance of face is crucial (Myers 1989: 5)”. The use of the 1st person pronoun Ieither threatens the participants’ faces to some degree and/or functions as aredressive strategy. In general, explicitly marked authorial presence will constitute aFTA to writers’ positive face (risk of disagreement) as well as to their negative face (risk of inviting criticism). In relation to readers, the FTAs are aimed towards their positive face (not being included) as well as to their negative face (speakers imposing their inference). Now let us consider Hyland’s notion: “reference to the writer explicitly marks astatement as an alternative view rather than as adefinitive truth; allowing the reader to choose the more persuasive explanation (Hyland 1996: 20)” which “invites the reader to participate in adialogue (Hyland 1996: 18)”. Consequently, positive politeness strategy towards the reader evoked by the use of 1st person pronoun Ithus lies in the conscious, yet implicit involvement of the readers into the discussion which minimizes the threat to their positive face (author invites the reader to participate in the inferential process), while the negative politeness strategy towards the reader lies in the fact that the writer does not present the claim as categorical, which mitigates the threat to the reader’s negative face. To attenuate the FTAs to the writer’s positive and negative faces it is often necessary to involve additional devices, such as hedges (epistemic verbs, adverbs, etc.). This account is mainly applicable to propositions from the Stating claims and opinions category as described above but it applies to the remaining categories as well in the sense that “some uses of Iare lowrisk, discrete instances of authorial intervention, while other uses, such as when the writer makes aclaim, carry much greater threat to face, and are potentially points at which the writer exposes themselves to attack by the audience (Harwood 2005: 344)”. VERONIKA RAUŠOVá 173 To conclude this section, it is important to note that the authors have at their hands more explicit and possibly more effective devices to achieve agreement, to avoid criticism, and to invite the reader to participate. However, there is areason for explicit authorial presence in the discourse using I, which is mainly “emphasizing what you have done, what is yours in any piece of research (Hyland 2005:181)”. Such an emphasis would be quite difficult to achieve while employing other devices and it depends entirely on the authors’ conscious choice whether to take the risks described above with aparticular proposition or to opt for safer strategies (passive constructions, abstract rhetors, inclusive pronouns, questions, etc.) 5. MATERIALS AND METHODS This study required acompilation of two specialized corpora, consisting of only single-authored research articles (RAs). The first corpus (CZCORP) comprises 40 RAs written in English by 40 Czech authors, extracted from Czech peer-reviewed academic journals with atotal word count of 201 797 words. The second corpus (ENCORP) includes 40 RAs written by 40 native English speakers which were published in renowned, peer-reviewed, international academic journals. The second corpus reached 307 947 words. The native language of the selected authors was verified from information available in the RAs themselves as well as data obtainable on the internet. Drawing data from different journals might present arisk for consistency of the material due to varied requirements each journal imposes on the submitting authors. While this is true for overall formatting of the text, Ihave not encountered any limitations regarding the style of aRA, except for explicit suggestion not to use defamatory language in the journal Applied Linguistics1 and to avoid sexist language in journals of the Elsevier publishing house2. Hopefully, the explicit authorial reference does not qualify as such. The following tables offer the composition of the corpora used for this study. The CZCORP comprises seven different journals primarily because it was impossible to obtain single-authored linguistic RAs by 40 distinct authors from alower number of journals. The scarcity of suitable RAs results in an additional compromise regarding the year of publication of the selected RAs. There are 6 RAs published before the year 2010, which is not ideal and has to be accounted for when drawing conclusions about the data. Nevertheless, Iassume that the data are still comparable with previous research and may be used to observe gradual change. Dontcheva-Navratilova’s (2014a) research has revealed differences between RAs of Czech authors written in English published locally compared to international publications in terms of the frequency of personal pronouns. (Dontcheva-Navratilova 2014a: 49). However, her data also reveal that the differences are mostly caused by individual authors who either underuse or overuse acertain pronoun in aparticular RA. This is relevant for further justification of the selection of Czech journals 1 https://academic.oup.com/applij/pages/General_Instructions 2 https://www.elsevier.com/journals/journal-of-pragmatics/0378-2166/guide-for-authors 174 LINGUISTICA PRAGENSIA 2/2018 from which Idraw my data. Iassume that when writing aresearch article in English aCzech author would intend to aim at Anglophone audience and would not tailor the discourse consciously for it to fit local conventions of academic writing simply for the fact that the journal is published locally. This assumption is crucial for interpreting the data because underuse of the 1st person pronoun Iin CZCORP can consequently be attributed to the underlying influence of the conventions of Czech academic discourse. Otherwise, it would stem simply from adeliberate adjustment of writing style to the expectations of the local audience (see Section 7.for asuggestion of further research). It was necessary to manually clean the texts of unwanted material such as abstracts, footnotes, tables, references, or block quotations. Ihave decided to leave shorter direct quotations as well as paraphrases in the corpora, since the expected frequency of the pronoun Iin such an environment is very low. The resulting texts Journal Publisher Year of publication No. of articles Brno Studies in English Faculty of Arts, Masaryk University, Brno 2004–2016 16 Linguistica Brunensia Faculty of Arts, Masaryk University, Brno 2016 6 AUC Philologica 1— Prague Studies in English Charles University, Karolinum Press 2017 6 AUC Philologica 3/2013— Prague Studies in English XXVI Charles University, Karolinum press 2013 5 Ostrava Journal of English Philology Department of English and American studies, Faculty of Arts, University of Ostrava 2014–2017 3 Linguistica Pragensia Faculty of Arts, Charles University 2013–2014 3 Studies in Applied Linguistics Faculty of Arts, Charles University 2011 1 Table 1.Composition of the CZCORP Journal Publisher Year of publication No. of articles Applied Linguistics Oxford University Press 2013–2016 10 Lingua Elsevier 2016–2017 10 Language Sciences Elsevier 2010–2017 10 Journal of Pragmatics Elsevier 2016–2018 10 Table 2.Composition of the ENCORP VERONIKA RAUŠOVá 175 were converted into plain text and loaded into AntConc (Anthony 2017). Both the corpora were searched for instances of 1st person singular pronoun I.The results were again manually sorted and all irrelevant occurrences (e.g. Iused as anumeral, Iappearing in acitation, Iused in examples, etc.) were removed. Finally, Ihave conducted an analysis of the pragmatic functions Irepresented in each of the instances and classified them according to the taxonomy described in Section 3.Due to the different size of the corpora, the results were normalized per 1000 words and the statistical significance of the results was determined using the log-likelihood statistical function3 at p < 0.01, which means that any LL value higher than 6.63 signifies astatistically significant result. 6. ANALYSIS In this section, Iwill present acomparison of results obtained from my corpora with the previous research and then offer aquantitative and qualitative analysis of the roles the authors perform in the discourse by using Iin their RAs with regards to the politeness theory. Corpus size 1st person singular pronoun I Relative frequency per 1000 words log-likelihood p < 0.01 ENCORP 307 947 664 2.2 150.41 significant difference CZCORP 201 947 163 0.8 Table 3.Overall results The quantitative analysis shows statistically significant underuse of the 1st person pronoun Iin research articles of Czech authors, which confirms the assumption made in the introductory section. Similar results were reached in previous studies which Idescribe next. The comparison between RAs written by Czech authors in English and RAs of native English speakers in terms of the frequency of their use of personal pronouns has been conducted by Dontcheva-Navratilova (2014a), who observed the frequency of 1st person pronouns (I, me, we, us, our) in acorpus consisting of 12 single-authored RAs for each group, published between 2001–2012. In her study, the frequency of the personal pronoun Inormalized per 1000 words reaches 0.4 in the corpus of Czech authors and 1.68 in the corpus of native English speakers (Dontcheva-Navratilova 2014a: 49). In the same year, in aresearch including 15 RAs and chapters in internationally published books written by Czech linguists in English between 2002–2012, the frequency of Iin the texts reached 0.37 per 1000 words (Dontcheva-Navratilova 2014b: 54). While comparing the previous findings with the results shown in Table 3, Imust agree with the claim that “Czech authors are affected 3 Log likelihood calculator available at: http://ucrel.lancs.ac.uk/llwizard.html was used to compare the data and determine significance of the results 176 LINGUISTICA PRAGENSIA 2/2018 by the tendency in Anglo-American academic writing towards an increased use of personal forms (Dontcheva-Navratilova 2014b: 55)”. Nevertheless, while the frequency of Iin RAs written in English by both groups of authors seem to be on the increase, Czech authors writing in English still fall short of native speakers of English. 6.1 FREQUENCY OF PRAGMATIC FUNCTIONS Table 4 shows the overview of the functional analysis of Iin linguistic RAs of Czech authors writing in English and native speakers of English. While the frequencies are higher in all the categories in the ENCORP data, statistically significant differences were achieved only in the categories of Discourse organisation and guidance and Stating opinions and claims. The category of Other was not included in the calculation since it includes incomparable instances. ENCORP frequency per 1000 words CZCORP frequency per 1000 words log-likelihood p<0.01 Discourse organisation and guidance 337 1.1 68 0.3 99.15 significant difference Stating opinions and claims 269 0.9 66 0.3 60.98 significant difference Recounting research process 42 0.1 27 0.1 0.01 difference not significant Other 16 x 2 x x Table 4.Pragmatic functions and their frequency The individual categories rank the same in terms of the frequency of their use in both corpora; however, each category shows underuse in the CZCORP. The Discourse organisation and guidance category is the most frequent function and it can be explained by the expected interactivity and dialogic nature of aRA. To establish an interaction with the reader the writer will often step into the discourse to navigate the reader through the text and the argumentation. The category of Stating opinions and claims that closely follows might owe its high frequency of use to the need to maintain authorial presence in the text to assert alevel of authority as well as to imply subjectivity in order to invite readers in the inferential process. The least frequent function in my data is Recounting research process. The reason why authors resort to Iso infrequently when talking about the research process and methods used prior to writing the RA might be due to other strategies being preferred to represent the scientific methods more objectively (passives, abstract rhetors) and also due to the fact that some of the RAs might focus on theoretical issues. The following sections will offer adescription of each category as well as example sentences from both corpora. VERONIKA RAUŠOVá 183 Ivanič, R. (1998) Writing and Identity: The Discoursal Construction of Identity in Academic Writing. Amsterdam: John Benjamins. Kuo, Ch. (1999) The Use of Personal Pronouns: Role Relationships in Scientific Journal Articles. English for Specific Purposes 18, 121–138. Myers, G. (1989) The pragmatics of politeness in scientific articles. Applied Linguistics, 10/1, 1–35. Tang, R. and S.John (1999) The “I” in identity: Exploring writer identity in student academic writing through the 1st person pronoun. English for specific purposes 18, 23–39. Vassileva, I. (1998) Who am I/who are we in academic writing? International Journal of Applied Linguistics 8, 163–185. Veronika Raušová Department of English Language and ELT Methodology Faculty of Arts, Charles University nám. Jana Palacha 2, 116 38 Praha 1, Czech Republic [email protected]