LLT Journal, e-ISSN 2579-9533, p-ISSN 1410-7201, Vol. 23, No. 2, October 2020 LLT Journal: A Journal on Language and Language Teaching http://e-journal.usd.ac.id/index.php/LLT Sanata Dharma University, Yogyakarta, Indonesia 241 AUTHORIAL PRESENCE IN ENGLISH RESEARCH ARTICLES BY NATIVE AND NON-NATIVE ENGLISH SCHOLARS Adelia Januarto, Tofan Dwi Hardjanto Universitas Gadjah Mada, Indonesia 1adeliajanuarto@mail.ugm.ac.id, 2deha@ugm.ac.id correspondence: adeliajanuarto@mail.ugm.ac.id DOI: doi.org/10.24071/llt.2020.230204 received 9 March 2020; accepted 6 April 2020 Abstract In this corpus-based study, we address the issue of authorial presence in English research articles (RAs) by native and non-native English scholars or writers in the field of second language writing. Our purposes are to compare the frequencies of authorial presence and to examine discourse functions of authorial presence in the native and non-native English scholars’ RAs. To achieve these purposes, 48 RAs were collected from two Scopus-indexed journals, namely Journal of Second Language Writing and Assessing Writing. Our finding suggests that native and non-native English scholars are different in terms of the degree of visibility in which native English writers are more visible than their non-native counterparts in their RAs. Furthermore, our functional analysis of authorial presence indicates that native English writers use more self-references for different purposes such as to describe research procedures, show the organization of the texts, but fewer self- references to guide readers through the texts than their non-native English peers. These differences might be caused by the idea of writer-responsible culture in native English writers’ RAs, and international publication context in non-native English writers’ RAs. Besides, these findings may be invaluable to material designs in English writing, especially in Indonesia, to assist students as novice writers to consider their explicit presence in their RAs. Keywords: authorial presence, research articles, L2 writing Introduction The use of first-person pronouns as the linguistic realization of authorial presence has become a subject of debate for many years in academic writing such as theses, dissertations, and research articles. On one hand, people problematize the existence of first-person pronouns in academic writing because this linguistic realization of authorial presence implies the subjectivity of the writers while academic writing is required to be objective. The use of first-person pronouns on academic writing defocuses readers from the ideational contents which the writer seeks to convey. As a result, some academic manuals (e.g., Johnson, 2016; Macmillan & Weyes, 2007; Rizvi, 2005; Wood, 2001) advocate the avoidance of first-person pronouns which signals authorial presence in their academic writing in favor of objectivity or neutrality. The avoidance of such authorial presence is commonly realized by the frequent uses of passive voice and non-human subjects mailto:adeliajanuarto@mail.ugm.ac.id mailto:deha@ugm.ac.id LLT Journal, e-ISSN 2579-9533, p-ISSN 1410-7201, Vol. 23, No. 2, October 2020 242 which are claimed to focus readers on the object under study, not the writers. On the other hand, such impersonal strategies, especially passive voice, which seems to convey objective tones are not free from criticisms as well. Dunleavy (2003) and Hinkel (2003) on the impersonal style on academic writing suggest that eliminating first-person pronouns contributes to vagueness and wordiness of a sentence, as the subject of a sentence is unclear, especially in the case of the agentless passive voice. As a consequence, this area in academic writing indicates that the use of first-person pronouns in academic writing is problematic. Since written academic discourse is now widely seen as the embodiment of writer-reader interaction (Hyland, 2001, 2005), self-mention which represents social interaction plays two important roles in written academic discourse. First, first-person pronouns enable a writer to construct his/her impression of credibility as a researcher in a particular area. The credibility is usually materialized by the use of explicit self-mention I or we so the writer indicates that s/he holds the responsibility for his/her view on the materials. Second, first-person pronouns may also serve to promote solidarity to the readers by bringing readers into the research. The pronouns selected to achieve such purpose are usually inclusive we in which the writer treats the audience of the article equally as if both participants hold the same view related to the interpretation of materials. Studies to date have shown that first-person pronoun indicating explicit authorial presence varies depending on some aspects such as writers’ nativity and disciplinary convention. It is now well-accepted that not only do disciplines influence the use of first-person pronouns, previous studies also indicate that nativity also contributes to the difference, such as authority and visibility. To name a few, previous studies (e.g., Al-Shujairi, 2018; Behnam, Mirzapour, & Mozaheb, 2014; Carciu, 2009; Dontcheva-Navrátilová, 2013; Işık-Taş, 2018; Martínez, 2005) have investigated authorial presence about nativity in English RAs of various disciplines such as chemistry, applied linguistics, sociology, and business management. Overall, the findings of their studies suggest that non- native English writers use more or fewer self-references than their native English counterparts do so that the results may be inconsistent. Regarding this issue, we aim to compare how native and non-native English scholars utilize self-references in their English-medium RAs, and to understand the discourse functions of each self-reference signaling presence in their RAs. In analyzing the authorial presence, we draw on Biber, Johansson, Leech, and Conrad’s (1999) grammatical categorization of first-person pronouns and Tang and John’s (1999) discourse functions of first-person pronouns. The underlying reason for choosing Biber et al.’s (1999) categorization is because it covers all English grammatical cases which enable us to explore and collect all types of English first-person pronouns in our corpus. Besides, we consider Tang and John’s (1999) model in comparison to others (e.g., Hyland, 2002; Kuo, 1999) due to its more general taxonomy which is potentially more accommodating for new sub-categories of discourse functions such as to show an intra-textual relation and to highlight the focus of discussion. LLT Journal, e-ISSN 2579-9533, p-ISSN 1410-7201, Vol. 23, No. 2, October 2020 243 First-Person Pronouns as the Realization of Authorial Presence The first-person pronoun grammatically belongs to a subgroup of noun class indicating a speech role as or referring to the speaker/writer (Heath, 2004; Huddleston & Pullum, 2005). The speech role itself denotes what position a participant in an interaction takes, such as a speaker, hearer, or non-participant referent that can be categorized into first, second, and third-person respectively. In general, the first-person pronouns which are the linguistic resource to manifest authorial presence can be seen in Table 1 as follows, but the pronoun we is problematic. Table 1. English first-person pronouns (Biber et al., 1999) Person Case Nominative Accusative Possessive Reflexive Determiner Pronoun Singular I me my mine myself Plural we us our ours ourselves The plural we have several referents such as a speaker and the audience, people in general, a group of speakers without the audience. The first two referents consider the addressee and the writers as one group while the last does not, and thus the prior belongs to inclusive we while the latter belongs to exclusive we. Unlike the singular pronoun I whose referent is clear, i.e. the speaker/writer, the pronoun we, therefore, is vague concerning its multiple referents. First-person pronouns as the linguistic realization of authorial presence in academic writing are complex, as they do not only refer to the referent but also inform multiple identities which the pronouns perform (e.g., see Işık-Taş, 2018; Károly, 2009; Tang & John, 1999). Regarding the identities signified by the linguistic resources, Tang and John (1999) propose a taxonomy of discourse functions of self-mentions in academic writing (see Table 2) by building on Ivanič’s theory on aspects of writer identity (1998). Their taxonomy of discourse functions encompasses six roles or identities, such as representative, guide, architect of the essay, recounter of the research process, opinion holder, and originator. Table 2. Identities/functions of self-references, adapted from Tang and John (1999) Identities Explanation Examples Least powerful Representative To represent a group of people or people in general “The English that we know today reflects many centuries of development (Script 6).” Guide through the essay To help readers in the interpretation of the text “Let us now look at some examples of J[amaican] C[reole] compared to standard British English (SBE) (Script 26).” Architect of the essay To show how the text is organized and outline the materials “In this essay, I will discuss the bastard status of English […] (Script 16).” LLT Journal, e-ISSN 2579-9533, p-ISSN 1410-7201, Vol. 23, No. 2, October 2020 244 Most powerful Recounter of the research process To recount research procedures or methods “I tape recorded a conversation with each co-researcher […] (Ivanic, 1998).” Opinion-holder To express personal views or attitudes “I agree with Fairclough (1992b) […] (Ivanic, 1988). Originator To convey knowledge claims “To me, the phrase embodies the whole evolution process of the language to its present day status (Script 8).” Research Methods This corpus-based study consisted of 48 English RAs (24 RAs written by native writers and 24 RAs written by non-native writers) in the field of second language writing. A small specialized corpus was considered for our study because it allows corpus researchers to see a particular type of discourse (Flowerdew, 2004), i.e. written academic discourse in research articles. L2 writing was selected as our samples because previous studies have not dealt with authorial presence in this field while this field might not just inform us what their studies demonstrate about L2 writing but also how expert writers in L2 writing who aware of authorial presence use this option in practice. In addition, the English RAs which were employed as our data were obtained from Assessing Writing and Journal of Second Language Writing, which were chosen based on three criteria, i.e. the scope of L2 writing, English-medium publication, and Scopus-indexed status. For the article selection, moreover, we applied some criteria, such as empirical articles in favor of its AIMRD (Abstract, Introduction, Methods, Results, and Discussion) structure, native and non-native writers, and the number of writers (single and multiple authorship). The AIMRD structure allows us to see what a self-reference serves concerning where the personal pronoun or other self- referring term is found in a particular section such as showing their responsibility for the selection of procedures in the Method section. Moreover, since the concept of the nativity is problematic here, following previous studies (Carciu, 2009; Işık- Taş, 2018; Mur-Dueñas, 2007), we considered writers’ nativity from their affiliated institutions which are located in English-speaking countries. As a result, the corpus samples of native writers can be seen as follows. Table 3. The size of NW and NNW corpus Journals Articles Total (in words) Assessing Writing (ASW) 24 172,051 (mean: 7,169) Native Writers (NW) 12 86,451 Non-native Writers (NNW) 12 85.600 Journal of Second Language Writing (SLW) 24 194,430 (mean: 8,101) Native Writers (NW) 12 98,648 Non-native Writers (NNW) 12 95,782 TOTAL 48 366,481 (mean: 7,635) LLT Journal, e-ISSN 2579-9533, p-ISSN 1410-7201, Vol. 23, No. 2, October 2020 245 Since we aim at exploring native and non-native English tendency in using authorial presence and examining discourse functions of authorial presence, both quantitative and qualitative methods were employed. The quantitative analysis for the first objective assisted with WordSmith Tools 5 (Scott, 2008) to see the frequency of personal pronouns in each journal and each group of writers. We searched authorial presence with first-person pronouns according to Biber et al.’s (1999) categorization and self-referring terms such as writer, writers, researcher, researchers, author, and authors in the corpus. Each frequency of the authorial presence was normalized to 100,000 words. Next, the chi-square (χ2) test was also done with Minitab 19.2 (Minitab, 2019) to examine whether the differences between frequencies of self-references in NW and NNW RAs were statistically significant. After finishing the calculation of the frequency distribution, the discourse functions of authorial presence were calculated and explained based on Tang and John’s (1999) classification so that we could see the functions of self- references with their co-text more comprehensively. Findings and Discussion Native and Non-native Writer’s Differences in the Corpus Table 4. Frequency Distribution in the NW and NNW corpus (per 100,000 words) Self- references NW NNW TOTAL Raw Norm. Raw Norm. Raw Norm. I 151 81.58 67 36.94 218 59.48 my 20 10.81 13 7.17 33 9.00 me 24 12.97 9 4.96 33 9.00 mine 1 0.54 0 0.00 1 0.27 we 341 184.23 307 169.26 648 176.82 our* 183 98.33 129 71.12 312 85.13 us 18 9.72 30 16.54 48 13.10 researcher* 30 16.21 53 29.22 83 22.65 author 2 1.08 0 0.00 2 0.55 TOTAL 770 415.99 608 335.20 1378 376.01 Note: NW = native writers, NNW = non-native writers, norm. = normalization Table 4 shows the degree of authorial presence in NW and NNW RAs by the frequencies of self-references. As seen in Table 4, normalized frequencies of self- references in English RAs indicate that NWs use self-references more than NNWs do. These self-references in NW and NNW RAs occur respectively 415.99 and 335.20 cases per 100,000 words. This quantitative result, in general, suggests that NWs with the linguistic resources show a slightly higher degree of visibility. Moreover, our result for the significance test on self-references across two groups of writers shows that the difference is statistically significant (df = 6, χ2 = 23.91, p = 0.001). In our findings, furthermore, the frequencies of the subjective pronoun I in the two groups show the most striking difference. A possible explanation for more than twice occurrences in NNW RAs might be that the writers consider this LLT Journal, e-ISSN 2579-9533, p-ISSN 1410-7201, Vol. 23, No. 2, October 2020 246 singular pronoun impolite in the sense that they do not involve readers (Myers, 1989) so they choose to use more plurals to be inclusive. The quantitative analysis further shows that the self-reference researcher(s) is more prevalent in NNW RAs. This might because NNWs who use this third-person reference want to give an impression of objectivity without completely detaching themselves from the texts, or the uses of this resource might be associated with writing practices in their particular academic cultures. In general, despite inconsistent with Isik-Tas’ (2018) and Behnam et al.’s (2014) findings in the field of sociology and chemistry, our overall result for NW and NNW differences in using self-references in L2 writing RAs agrees with Dontcheva-Navrátilová’s (2013) finding which shows the higher frequency of self-references in NW RAs than in NNW in the context of international publication. Given the fact that applied linguistics and the field of L2 writing are closely related (Silva & Leki, 2004), the similarity between our quantitative finding and Dontcheva-Navratilova’ is hardly surprising. Therefore, in line with Dontcheva-Navratilova, the lower frequency of self-references in NNW might also indicate the influence of their non-Anglophone academic literacies on their L2 writing RAs. Discourse Functions of Self-references in NW and NNW Corpus In this section, we demonstrate whether or not self-references stated in Table 4 were employed to achieve similar discourse functions in NW and NNW RAs. In addition to the quantitative analysis presented in Table 5, we also explain the discourse functions of self-references qualitatively to provide a more complete picture of authorial presence in our corpus. Here, we acknowledge that it is difficult to provide a qualitative analysis with a contrastive approach since both groups use the same devices and similar functions; what makes them different is to what extent self-references are used to serve discourse functions. As a consequence, our explanation of discourse functions of authorial presence relies on a qualitative approach without any intention to compare NWs and NNWs. Moreover, before going further to Table 5 on the different discourse functions of self-references between NWs and NNWs, we will briefly remind what first-person pronouns and other self-references can serve in RAs based on Tang and John’s (1999) taxonomy. Firstly, first-person pronouns, especially inclusive, can function as a representative when they are used to denote a group of people, and as a guide through the RA when the inclusive pronouns referring to both writers and their readers serve to navigate readers in understanding the text. Similar to guide, the architect of the RA is also concerned with directing readers but one main difference of architect from guide lies on the fact that architect is the text-oriented whereas guide through the RA is reader-oriented. Thus, personal pronouns and self-referring terms (the researcher(s), the author(s), and the writer(s) are considered as an architect when they are utilized to purely organize text. Additionally, the two high-risk functions, viz, opinion-holder and originator. Between these two and even the rest types, the originator is the most face- threatening function because authorial presence is associated with knowledge claims and findings which are new to the disciplinary literature so the writers are LLT Journal, e-ISSN 2579-9533, p-ISSN 1410-7201, Vol. 23, No. 2, October 2020 247 explicitly exposed to readers’ negative evaluation. Opinion-holder, on the contrary, is concerned with the writers’ role to state a personal view. Table 5. Discourse Functions of Self-references in NW and NNW Corpus Functions/Identities NW NNW Raw Norm. Raw Norm. Representative 15 8.13 10 5.51 Guide through the RA 17 9.18 70 38.59 Architect of the RA 200 108.05 146 80.49 Recounter of the Research Process 379 204.76 275 151.61 Opinion-holder 92 49.70 68 37.49 Originator 67 36.20 39 21.50 TOTAL 770 415.99 608 335.20 Note: NW = native writers, NNW = non-native writers, norm. = normalization Table 5 shows preferences on what authorial identities NWs and NNWs seek to construct based on the discourse functions of self-references. As seen in the table, we can note that there are three differences between NWs and NNWs. NWs use more self-references to recount research procedures (204.76 vs. 151.61 occurrences) and structure their RAs (108.05 vs. 80.49) than NNWs. However, self-references in NW RAs which are used to guide readers through their RAs are less prevalent than in NNW RAs (9.18 vs. 38.59). These suggest that while NWs employ a more personal approach in organizing texts and describing their research process to show their more concerns on the textual organization and procedural decisions, NNWs are more facilitative to their readers given the higher frequency of self-references to help readers in the interpretation of their scholarly texts. Furthermore, the more prevalent self-references to structure texts and describing research procedures in NWs are likely attributable to the idea of Anglophone writer-responsible culture (see Hinds, 2001) to clarify messages and show their responsibility for selected procedures, whereas the high frequency of self- references to guide readers in NNW RAs might be influenced by international publication context in which NNWs make attempts to persuade readers through guiding them to their preferred interpretation. Dontcheva-Navrátilová’s (2013) study, however, shows the opposite in which NWs use more self-references for guiding readers and structuring texts but fewer self-references for recounting research procedures. She explains that guiding readers and structuring texts agrees with NW academic literacy which highlights the reader-oriented character. She, furthermore, believes that the underlying reason for more prevalent self-references in NNW RAs is that NNWs opt for creating a more personal approach when recounting research procedures. Regarding Dontcheva-Navrátilová’s findings, the differences between our findings and hers are surprising. First, given the fact that the fields of study on which we focus are different, i.e. applied linguistics and L2 writing, our difference in terms of discourse functions of self-references in English RAs, particularly on recounting research procedures and helping readers in interpreting RAs, might be LLT Journal, e-ISSN 2579-9533, p-ISSN 1410-7201, Vol. 23, No. 2, October 2020 248 to some extent influenced by the disciplinary factor. Thus, our difference might reflect Xia’s (2018) study which demonstrates discipline as one contributing factor for differences of discourse functions. Second, we believe that the similarity of findings between ours and Dontcheva-Navrátilová’s finding in terms of organizing scholarly texts could be caused by the shared awareness of the importance of signposts to indicate the logical structure of RAs in the field of L2 writing and applied linguistics. Representatives First-person pronouns in English RAs by both native and non-native scholars are related to their discourse functions which represent what identities are constructed by the scholars. In the corpus, writers construct the least face- threatening identity, ‘representatives’, by using sources such as inclusive we and inclusive our. Inclusive plural pronouns here are chosen to represent people in general (example 1) and writers’ disciplinary community (example 2 and 3). (1) The launch of the digital age has ushered in a growing demand for our capacity to produce, manipulate, and interpret visual and graphical representations of information (Lowrie & Diezmann, 2007). (ASW12NNEI) (2) That is, we have not as yet fully understood in detail how the features of effective writing develop among students and therefore, we have not been able to articulate that development. (ASW13NEI) (3) […] a further exploration of the effects of intertextual processing manifested in discourse synthesis and multiple-text comprehension in both L1 and L2 settings could advance our understanding of the multifaceted nature of integrated writing and offer insights into instructional and assessment practices. (SLW19NNEI) These examples show that how RA writers position themselves to construct ‘representatives’ through inclusive our and we with linguistic resources which involve cognition (e.g. understood and understanding) or ability (e.g. capacity) in the Introduction section. These uses of inclusive plural pronouns in the examples are related to typical ‘moves’ of Introduction (see Dudley-Evans, 1986, p. 135; Swales & Feak, 2014, p. 331) in RAs such as establishing a research territory by suggesting the relevance of their studies to either the society or their disciplinary community (1); highlighting a knowledge gap (2); and suggesting the significance of their own study (3). Aligning themselves to the society or their disciplinary community through pronouns in the Introduction section helps them emphasize the persuasive effect when creating a research space so they can promote the novelty of their researches to the readership. Guide through the RAs Writers make themselves explicitly visible in texts through the use of we, but this pronoun, in addition to the writers, also refers to readers. The reader- inclusive pronouns in the English RAs are aimed to draw readers’ involvement in LLT Journal, e-ISSN 2579-9533, p-ISSN 1410-7201, Vol. 23, No. 2, October 2020 249 the interpretation of the texts, and with this linguistic resource. In these following examples, we show the role of inclusive we and our in guiding readers to the writers’ preferred reading. (4) An outstanding finding was that unlike previous studies, our study revealed a significant relationship between the holistic rating of composition quality and complexification at multiple levels of syntactic organization: the sentential, the clausal, and the phrasal level. (ASW08NNED) (5) Aggregating the distinct decision-making behaviors, we can detect a progression of the raters’ distribution of attention to textual features corresponding to the official assessment criteria. (ASW05NER) (6) For example, in Table 7 we can see that, keeping WM at the average, Grade 7 students were 3.86 times more likely to score higher on Task 4 Listen-Write than Grade 6 students [….] (SLW14NER) In example (4) to (6), the inclusive pronouns, which occur in the Results and Discussion section, play important roles in suggesting how their RAs are better read. In the case of example (4), a single writer uses inclusive our to seek cooperation from and at the same time to involve her readers in the interpretation so they can arrive at the same conclusion about the data. In other words, the writers seek to navigate their readers to their preferred interpretation of evidence. Architect of the RAs Unlike previous discourse functions of self-references that focus on the use of inclusive pronouns, discourse function which constructs the ‘architect’ persona is realized through exclusive self-references such as I, exclusive we, and the researchers. The architect of the RA itself by definition denotes a writer’s role to compose and organize a scholarly text (Tang & John, 1999, p. S28). Furthermore, as noted in Table 2, this identity is constructed when writers organize their own texts and outline what they aim to discuss by involving the aforementioned self- referring devices. In our corpus, we found that this identity is involved in several ways to construct ‘architect’: section structure, cross-reference, and outline of the RA. (7) Qualitative results from DSE1, DSE2, and DSE3 typify raters’ attention to textual features; so, in the results section, I focus on these findings. (ASW01NEM) (8) As we mentioned previously, many rubrics do not consider evidence of fairness, or how independent raters utilize the rubric (Hawthorne, Bol, & Pribesh, 2017). (ASW15NED) (9) The researchers wanted to see whether (1) modeling was more effective than self-practice, and (2) collaboration was more effective than working alone in enhancing students’ detection, revision, and commenting skills. (SLW09NNEI) LLT Journal, e-ISSN 2579-9533, p-ISSN 1410-7201, Vol. 23, No. 2, October 2020 250 From example (7) – (9), self-references are involved in signposts, expressions showing text structures. In particular, the self-reference I in example (7) is used to structure the text by highlighting what is being the center of the writers’ interest in the section Results. In addition to section structure, the pronoun we in example (8) is used when the writers show the internal relation between one part of the text to the other. The self-reference the researchers along with the verb in example (9) serves to give an overview of their problems in general. Recounter of the Research Process Writers exploit pronoun I, exclusive plural pronoun we, and the researchers to describe research procedures. With this alignment to research procedures, they seek to show their role as ‘recounters’ who did the research process and made their procedural decision. In our corpus, moreover, we found that self-references, which mainly occur in the Methods section, can help writers in explaining their methods, especially data collection (example 10 and 11) and data analysis (example 12 and 13). (10) Before the program, I spent four weeks observing 20 sessions of the CE course in the five teachers classes (four sessions per teacher). (SLW07NNEM) (11) From the twenty-one students who attended the first class, I asked four multilingual writers to attend follow-up individual one-hour interviews in November/December 2016 (Interview 1). (SLW17NEM) (12) For multi-trait average scores, the researchers employed the same G study approach as the holistic scores since a single score averaged over multi-trait sub-scores was used for analysis. (ASW18NEM) (13) An additional coder who was a university English teacher with an Australian Master’s Degree in Teaching English as a Second Language was invited to work with us on examining the student drafts and WCF. (ASW24NNEM) The uses of authorial presence, which are realized through self-references in example (10) and (11), enable them to underscore their unique personal roles in the process of collecting data. Next, in example (12) and (13), self-reference the researchers and us are used to describe data analysis procedures, but interestingly, instead of using personal pronoun we, the writers in example (12) use the third- person perspective through the researchers to mention themselves when describing and justifying his methodological decision. It seems that writers as narrators of their research use this alternative self-reference in the Methods section because they aim to highlight their role as researchers who chose the methodological approach with the detachment of their research as their preferred choice so they can maintain their objectivity. Opinion-holder By making selves explicitly visible in the RAs through self-references in expressing personal opinions or attitudes, writers expose themselves to criticisms LLT Journal, e-ISSN 2579-9533, p-ISSN 1410-7201, Vol. 23, No. 2, October 2020 251 because this identity, opinion-holder carries a highly face-threatening degree (Işık-Taş, 2018, p. 31). And unlike ‘recounters’ which typically occur in the Methods section, this identity we found in the corpus is widespread in Introduction, Methods, and Discussion. (14) The shared processes identified in our data deserve consideration in developing and using scores from integrated assessment. (ASW14NED) (15) We argue that much of the research on writing assessment has omitted an important element: fairness. (ASW15NEI) (16) As discussed below, we believe that the developmental progression proposed by Biber and colleagues represents a useful means of hypothesizing the syntactic development of maturing academic writers in English [….] (SLW16NEI) (17) We assumed that a writing task that required summarizing ideas in a source reading passage, of the kind that now features in most major English proficiency test [….] (SLW22NNEM) In our corpus, we identified that writers use exclusive we and our to make themselves visible and at the same time to express their personal opinions on their own method (14), previous studies (15), a theory (16), and a topic of discussion (17). Furthermore, in the conveyance of opinions, the self-references which typically co-occur evaluative verbs such as deserve, argue, believe, and assume are employed to strengthen the persuasive effect of the evaluations. Originator Writers use linguistic resources such as first-person pronouns to express their knowledge claim about a subject matter. In the corpus, exclusive pronoun we are commonly used with verbs found while exclusive pronoun our are frequently used with study and research to indicate knowledge claims. (18) For each statistic, we found the value at the 90th percentile for each data set, and then took the average of the 90th percentile values over all the samples. (ASW16NEM) (19) In our study, there was no indication in any of the cases that students were expecting their teacher not to provide WCF. (ASW24NNER) (20) By contrast, in our research the teachers provided much more indirect feedback than direct feedback, with individual practice seen to be determined by whether the teachers believed they or the students were responsible for learning. (SLW13NED) (21) Despite the clear differences in the design of both interventions, we only found trends towards significant differential development. (SLW24NNED) As seen in these examples, the pronouns we and our occur in the Method, Results, and Discussion sections to report their findings and to emphasize their role as people who contribute findings to their disciplinary community, and at the LLT Journal, e-ISSN 2579-9533, p-ISSN 1410-7201, Vol. 23, No. 2, October 2020 252 same time, showing presence when conveying findings indicates the ownership of the findings. As a consequence, making selves visible can enable them to gain recognition for their personal contribution. To serve this function, however, making selves explicitly visible is risky to their own face because stating findings or claims is the most assertive function (Hyland, 2002, p. 1103). Conclusion and Implications In this study, we have provided evidence in the field of L2 writing that degree of visibility is influenced by nativity, and authorial presence is useful to achieve various purposes in their academic writing. Regarding NW and NNW differences in making themselves explicitly visible, our quantitative analysis suggests that even though there is a small difference between the two groups in which NWs are more visible in internationally reputable journals, our statistical measurement yields a significant difference. Such differences could be caused by the influence of NNW cultures on their English RAs. Moreover, we also note that authorial presence in NW and NNW RAs shows differences in the use of self- references to recount research procedures and to structure RAs. Two possible factors are contributing to the differences of such discourse functions such as Anglophone academic culture, viz, writer-responsible for the first two functions. Moreover, we also note that NNWs are more facilitative in terms of navigating their readers throughout their texts than NWs given NNW higher frequency of self-reference for guiding readers. What we have found in our study about authorial presence between NW and NNW RAs, in general, can contribute to our understanding of how NWs and NNWs are different in international publication context even in one discipline. Nonetheless, given the relatively small numbers of samples, generalizing the findings to a broader area should be done with caution. Future studies, therefore, are suggested to validate our findings with larger samples from multiple disciplines to show whether there is a cultural and disciplinary interplay. It is also possible to complement future studies with interviews to obtain “insider ‘emic’ approach” (Swales, 2019) on why academic writers use self-references in their RAs. Furthermore, our findings here may contribute to the English academic writing courses, particularly in Indonesian higher education. As we have shown, expert writers in L2 writing explicitly show their visibility through self-references despite the different degrees about their nativity. Thus, this study cannot only exemplify the extent to which Indonesian novice writers can use self-references in English writing but also show them how the expert writers use self-references to achieve particular functions. References Al-Shujairi, Y. B. J. (2018). What, which and where: examining self-mention markers in ISI and Iraqi local research articles in applied linguistics. Asian Englishes, 1-15. doi:10.1080/13488678.2018.1544699 LLT Journal, e-ISSN 2579-9533, p-ISSN 1410-7201, Vol. 23, No. 2, October 2020 253 Behnam, B., Mirzapour, F., & Mozaheb, M. A. (2014). Writer's presence in English native and non-native speaker research articles. Procedia-Social and Behavioral Sciences, 98, 369-374. doi:10.1016/j.sbspro.2014.03.428 Biber, D., Johansson, S., Leech, G., Conrad, S., & Finegan, E. (1999). Longman grammar of spoken and written English. London: Longman. Carciu, O. M. (2009). An intercultural study of first-person plural references in biomedical writing. Ibérica, 18, 71-92. Dontcheva-Navrátilová, O. (2013). Authorial presence in academic discourse: Functions of author-reference pronouns. Linguistica Pragensia, 23(1), 9-30. Dudley-Evans, T. (1986). Genre analysis: an investigation of the introduction and discussion sections of M.Sc dissertations. In M. Coulthard (Ed.), Talking about text (pp. 128-145). Birmingham: English Language Research, University of Birmingham. Dunleavy, P. (2003). Authoring a Ph.D.: How to Plan, Draft, Write and Finish a Doctoral Thesis or Dissertation. Houndmills: Palgrave MacMillans. Flowerdew, L. (2004). The argument for using English specialized corpora to understand academic and professional language. In U. Connor & T. A. Upton (Eds.), Discourse in the professions: Perspectives from corpus linguistics (pp. 11-33). Amsterdam: John Benjamins Publishing. Heath, J. (2004). Person. In G. Booij, C. Lehmann, & J. Mugdan (Eds.), Morphologie: Ein internationales Handbuch zur Flexion und Wortbildung (Vol. 2, pp. 998-1015). Berlin: Walter de Gruyter. Hinds, J. (2001). Reader versus writer responsibility: A new typology. In T. Silva & P. K. Matsuda (Eds.), Landmark essays on ESL writing (pp. 63-72). New York: Routledge. Hinkel, E. (2003). Teaching academic ESL writing: Practical techniques in vocabulay and grammar. Mahwah: Lawrence Erlbaum Associates. Huddleston, R., & Pullum, G. K. (2005). Introduction to the Grammar of English. Cambridge: Cambridge University Press. Hyland, K. (2001). Humble servants of the discipline? Self-mention in research articles. English for Specific Purposes, 20(3), 207–226. doi:10.1016/S0889- 4906(00)00012-0 Hyland, K. (2002). Authority and invisibility: authorial identity in academic writing. Journal of Pragmatics, 34, 1091-1112. doi:10.1016/S0378- 2166(02)00035-8 Hyland, K. (2005). Stance and engagement: a model of interaction in academic discourse. Discourse Studies, 7(2), 173-192. doi:10.1177/1461445605050365 Işık-Taş, E. E. (2018). Authorial identity in Turkish language and English language research articles in Sociology: The role of publication context in academic writers' discourse choices. English for Specific Purposes, 49, 26-38. doi:10.1016/j.esp.2017.10.003 Ivanič, R. (1998). Writing and identity. Amsterdam: John Benjamins. Johnson, A. P. (2016). Academic Writing: Process and Product. Maryland Rowman and Littlefield. LLT Journal, e-ISSN 2579-9533, p-ISSN 1410-7201, Vol. 23, No. 2, October 2020 254 Károly, K. (2009). Author identity in English academic discourse: A comparison of expert and Hungarian EFL student writing. Acta Linguistica Hungarica, 56(1), 1-22. Kuo, C.-H. (1999). The use of personal pronouns: Role relationships in scientifc journal articles. English for Specific Purposes, 18(2), 121–138. doi:10.1016/S0889-4906(97)00058-6 Macmillan, K., & Weyes, J. D. B. (2007). How to write dissertations and project reports: Pearson Education. Martínez, I. A. (2005). Native and non-native writers’ use of first person pronouns in the different sections of biology research articles in English. Journal of Second Language Writing, 14(3), 174-190. doi:10.1016/j.jslw.2005.06.001 Minitab. (2019). Minitab 19. State College, PA: Minitab, Inc. Mur-Dueñas, P. (2007). ‘I/we focus on…’: A cross-cultural analysis of self- mentions in business management research articles. Journal of English for Academic Purposes, 6(2), 143-162. doi:10.1016/j.jeap.2007.05.002 Myers, G. (1989). The pragmatics of politeness in scientific articles. Applied linguistics, 10(1), 1-35. doi:10.1093/applin/10.1.1 Rizvi, M. A. (2005). Effective Technical Communication. New Delhi: Tata McGraw-Hill. Scott, M. (2008). WordSmith Tools Version 5. Liverpool: Lexical Analysis Software. Swales, J. M. (2019). The futures of EAP genre studies: A personal viewpoint. Journal of English for Academic Purposes, 38, 75-82. doi:10.1016/j.jeap.2019.01.003 Swales, J. M., & Feak, C. B. (2014). Academic Writing for Graduate Students: Essential Task and Skills. Ann Arbor: The University of Michigan Tang, R., & John, S. (1999). The ‘I’in identity: Exploring writer identity in student academic writing through the first person pronoun. English for Specific Purposes, 18, S23-S39. doi:10.1016/S0889-4906(99)00009-5 Wood, N. (2001). The Health Project Book: A Handbook for New Researches in the Field. London: Routledge. Xia, G. (2018). A cross-disciplinary corpus-based study on English and Chinese native speakers’ use of first-person pronouns in academic English writing. Text & Talk, 38(1), 93–113. doi:10.1515/text-2017-0032