© 2015 Narr Francke Attempto Verlag 44 (2015) • Heft 1 C HRISTINE S. S ING * “My own summary is definitely less scientific but more easy to grasp”: ESP writing and disciplinary identity Abstract. This paper examines the construction of writer identities in the ESP writing of business students using English as an L2. Despite the wealth of literature on the topic, little is known about identity in ESP writing, particularly involving ESP undergraduates. The aim of this study is to investigate how these students use pronominal self-reference in order to perform authorial identities. Drawing on a self-compiled specialised corpus, the study adopts a mixed-methods approach of computation and interpretation. It was found that student writers, in assuming weak authorial roles, tend to use self-reference with low-risk functions. The findings also suggest that these students rely on inclusive we, which highlights field-specific uses in their writing. These results may help us to understand how writer identities assist students in their development of a disciplinary identity. 1. Introduction There is no denying that “[w]riting is […] the central activity of institutions” (H YLAND 2013: 95), to the extent that it is the explicit acknowledgement of membership in the academic discourse community. Becoming a proficient writer is thus endemic to academic literacy practices, which “constitute central processes through which students learn new subjects and develop their knowledge about new areas of study” (L EA / S TREET 1998: 157). In acquiring the ‘subject-specific literacies’ (H YLAND 2002b: 352) of a given community, students may eventually develop a disciplinary identity, which is enacted in and through writing. One way of studying identity therefore involves examining the extent to which students inhabit their texts. Research into identity construction in academic writing was sparked by I VANIČ ’s (1998) pioneering study of self-representation in the writing of mature students. Since then, several typologies of writer identities have been proposed. Most studies in the field have focused on graduate or expert writers, addressing disciplinary variation with reference to research articles (RA) and paying little attention to variability within a single discipline. It is this gap that the present study proposes to fill. Its purpose is thus to provide an in-depth analysis of business students’ identity construction in a piece of assessed writing. It will be argued that ESP writing is a genuine showcase for studying identities * Address for correspondence: Dr. Christine S. S ING , Lecturer/ Postdoctoral Researcher, Institut für Englische Wirtschaftskommunikation, Wirtschaftsuniversität Wien, Welthandelsplatz 1, A-1020 W IEN . E-Mail: csing@wu.ac.at Research areas: English for Academic Purposes, Corpus Linguistics, Applied Linguistics. ESP writing and disciplinary identity 83 44 (2015) • Heft 1 as these students are, to a larger extent than those in general English for academic purposes (EAP) settings, forced to assume several, often conflicting, identities. This has important implications for discipline-specific writing instruction in that these students have to live up to quite different literacy expectations. Not all of these identities lend themselves to (corpus) linguistic analysis. While disciplinary identities are discursively constructed, writer identities can be tracked on the textual level by means of identifying pronominal self-reference. Drawing on a selfcompiled specialised corpus - the corpus of Academic Business English (ABE) - this present account sets out to examine all writer-oriented pronouns in the corpus, including section-based analyses of the seminar papers that make up its database. Prior to presenting the main findings, I will first outline the possibilities for selfhood in ESP writing, which will be followed by a brief overview of existing typologies of writer identities. The in-depth study of self-reference will conclude with a discussion of implications for ESP writing instruction. 2. ESP writing and (disciplinary) identity Given the internationalisation of academic practices in Higher Education (HE), “university students around the world are increasingly likely to be using English for their studies, although in many cases their pre-degree preparation will not have included extensive writing practice in English in the relevant genres” (N ESI / G ARDNER 2012: 3). Familiarising student writers with the (specific) purposes of university writing is therefore an essential prerequisite for smoothing the transition from secondary to higher education. University writing thus marks a cultural shift, confronting students with the fact that “someone cannot engage in a discourse in a less than fluent manner. You are either in it or you’re not” (G EE 1996: 155). In order to be immersed in the academic discourse community, students need to learn a disciplinary culture. Yet, this enculturation clearly goes beyond the level of conventions: in developing academic writing skills, students need to create a new identity (H YLAND 2002b: 352), demonstrating that they are accustomed with the genres of particular disciplines. Hence identity construction is a central concern in their literacy development. I VANIČ (1998: 23) distinguishes between four dimensions of writer identity. Three of them - the autobiographical self, the discursive self and the authorial self - relate to the writing individual and interact in text production. The autobiographical self originates from the writer’s previous experiences, including beliefs and attitudes, while the discursive self becomes manifest in the writer’s positioning in a particular discourse. The authorial self, then, determines a writer’s visibility in a text, indicating the extent to which s/ he claims responsibility for its content. The fourth dimension complements the others in that it abstracts from the writing individual and relates to the “prototypical identities available in the socio-cultural context of writing” (I VANIČ ibid.). The latter clearly refers to the options of identity available to students in a particular institutional context. 84 Christine S. Sing 44 (2015) • Heft 1 The ESP setting of a business school showcases such a particular institutional context, suggesting several conflicting identities that students need to adopt and adapt to, whether as novices in the field, apprentice writers or trainee professionals. In doing so, ESP students meet three challenges: First, today’s ESP student is expected to be a proficient L2 user of English. Unsurprisingly, these students are linguistically challenged, struggling “simultaneously with the challenges of general English, academic English and specialised disciplinary language” (P ETERS et al. 2014: 744). Lacking competencies in general English will, quite inevitably, delay literacy development. Second, although writing is “most prominent in business, academic, and professional domains, where it often is used to carry out tasks of relatively high importance” (T ARDY 2012: 6266), writing instruction has rarely taken centre stage in the ESP curriculum. Thus there is generally little knowledge about the target genres in the (future) workplace. Third, these students also experience a culture clash, stemming from the fact that they are rooted in epistemologies different from both educational and professional genres. Turning now to self-representation in writing, from a linguistic perspective, “a writer’s identity is created by, and revealed through, the use or absence of the I pronoun” (H YLAND 2002a: 352). Based on this assumption, several typologies of self-representation in writing have been proposed (T ANG / J OHN 1999; H YLAND 2002a; H ARWOOD 2005b; S TARFIELD / R AVELLI 2006; S HELDON 2009). They differ in two respects: the data used to establish the identities and a focus on either the writing individual or the discourse functions by which identity is enacted. For example, some studies refer to a so-called ‘Methodological I’ (e.g. H ARWOOD 2005b) while H YLAND ’s analysis (ibid.) is grounded in discourse functions such as explaining a procedure. I will build on T ANG / J OHN ’s (ibid.) framework, extending it by means of H YLAND ’s (ibid.) discourse functions. In contrast to T ANG / J OHN , I will argue that genre roles are not to be dissociated from discourse roles in that writer identities may also index disciplinary identities. In their study of undergraduate essay writing, T ANG / J OHN (ibid.) discovered six different identities (cf. Figure 1  page 85). The authors found that the identities behind these pronouns have varying degrees of visibility and assertiveness, showing a cline of authorial presence in writing. The weakest form of authorial presence − apart from the complete absence of author pronouns − is the so-called ‘I as representative’ while roles such as the ‘opinion-holder’ and ‘originator’ are situated at the top-end of the assertiveness scale. It is possible to relate these identities to various discourse functions, such as H YLAND ’s (2002a): Expressing self-benefits; Stating a goal/ purpose; Explaining a procedure; Elaborating an argument; Stating results/ claims. While there is considerable overlap between the roles shown in Figure 1 and these functions, there are also important differences between them. Thus stating a goal is the equivalent of ‘I as architect’, explaining a procedure overlaps with ‘I as recounter of the research process’ and elaborating an argument is congruent with ‘I as opinion holder’. The most powerful role ‘I as originator’ can be linked to stating results/ claims. ESP writing and disciplinary identity 85 44 (2015) • Heft 1 44 Fig. 1: A typology of possible identities behind the first person pronoun in academic writing (adapted from T ANG / J OHN ibid.: 29) There are also important differences between the taxonomies. H YLAND ’s (ibid.) function expressing self-benefits, adding a reflective dimension to identity construction, is not attested in T ANG / J OHN ’s (ibid.) data. There are no equivalents for the roles ‘I as representative’ and ‘I as guide’ either. This is owed to methodology since H YLAND (ibid.) excluded all inclusive uses of the plural forms we, our and us. This is rather problematic since, due to a lack of semantic distinction, the ambivalence between exclusive and inclusive uses cannot be fully resolved (H ARWOOD 2005a). More importantly, the inclusive uses of we appear to be field-specific; hence their relevance to the present analysis. 3. Study and method There are two common approaches to studying discipline-specific writing. First, corpus-based approaches aim to retrieve lexico-grammatical patterns in order to identify field-specific uses of language. These variationist accounts tend to focus on variables such as novice vs. expert practices or English L1 vs. English L2 uses. There is a substantial body of research on disciplinary variation in academic writing, including citation practices (e.g. C HARLES 2006) or shell noun uses (e.g. S ING 2013). Second, ethnographically-oriented studies seek to address the social conditions of the production and interpretation of academic texts (e.g. G NUTZMANN / R ABE 2014: 33). Using interpretative methods, the authors propose an interview-based study of the language demands and attitudes of German researchers using English as an L2 in their writing. The ultimate aim of such an account consists in homing in on precisely those contextual factors amiss in much corpus-based research. This present account intends to reconcile these two approaches, drawing on a selfcompiled specialised corpus, the ABE corpus. “‘Specialised’ corpora are corpora designed for the purpose of creating a sample of specialised language by collecting texts […] of similar text-type or genre” (G AVIOLI 2005: 7). These have the edge over larger corpora whenever detailed, fully contextualised analyses are intended. With spe- 86 Christine S. Sing 44 (2015) • Heft 1 cialised corpora, “the analyst is probably also the compiler and does have familiarity with the wider socio-cultural context in which the text was created […]. The compilercum-analyst can therefore act as a kind of mediating ethnographic specialist informant to shed light on the corpus data” (F LOWERDEW 2005: 329). This inside information will be used when discussing the findings of the corpus study below. Subcorpora Number of papers Number of running words Business 103 236,917 Economics 104 273,455 Finance 104 253,203 Marketing 102 251,156 Total 413 1,014,731 Table 1: ABE Corpus description The specialised language represented in the ABE corpus is “[t]he academic business English required by students on courses in disciplines such as business, finance, accounting and banking [which] has more in common with the study of other EAP disciplines” (D UDLEY -E VANS / S T J OHN 1998: 53). As shown in Table 1, the ABE corpus totals c. 1 million running words, comprising more than 400 seminar papers produced by advanced students of international business administration. The seminar paper is deeply rooted in German-speaking HE (K RUSE 2006). It is an argued text (S IEPMANN et al. 2008: 20), frequently taking on the form of RAs in miniature (E HLICH 2003). In many disciplines, particularly in the social sciences and humanities, it continues to be the standard type of assessed writing. The corpus has been subdivided into four thematic subcorpora, each of which contains approximately 250,000 words. The subcorpora were built on the basis of both internal and external criteria. The former entailed generating a list of topical vocabulary that these texts have in common while the latter related to the thematic focus of the seminar in which the paper originated. 4. Main Findings In the first stage of the analysis the writer-oriented pronouns I, me, mine, my, we, our, ours and us were retrieved. The results are displayed in Figure 2 (  page 87), demonstrating that, with the exception of mine and ours, all pronouns could be accounted for in the data. In order to facilitate cross-corpus comparison, the frequencies were normalised to occurrence per 1,000 words. ESP writing and disciplinary identity 87 44 (2015) • Heft 1 Fig. 2: Frequency of writer-oriented pronouns across subcorpora As can be seen, the most pervasive pronouns across all four subcorpora are I and we. Regarding the frequency of occurrence of the other writer-oriented pronouns - me, my, our and us - there are important differences between the sub-fields at hand. There seems to be a noticeable divide between papers in business and finance on the one hand and economics and marketing on the other. This effect is particularly strong when focusing on the plural forms our, us, we exclusively. The fact that pronominal self-reference is attested in the data is not, however, a reliable indicator for the actual strength of authorial presence in the papers. In order to measure this more accurately, it is useful to also include the pronouns’ range and dispersion. Range concerns the number of texts in which the item appears (P AQUOT 2010: 48) while dispersion is “a statistical coefficient of how evenly distributed a word is across successive sectors of the corpus” (R AYSON 2003). Accordingly, the pronouns can be said to have low range as they only occur in approximately one quarter of the papers in each of the subcorpora. Range is lowest in finance and business, amounting to 20% and c. 24%, respectively. The pronouns occur in 26% of the texts in the marketing and in 30% of the texts in the economics subcorpus. Crucially, these averages include all six writer-oriented pronouns. If we isolate the most frequently occurring pronouns I and we, range increases dramatically to c. 60% on average. Taken together, these findings suggest that the student writers assume low visibility in their texts. Given the average length of 2,200 words, even the most frequent pronouns can be argued to occur as little as two to three times per seminar paper. Range appears to be surprisingly low when compared to T ANG and J OHN ’s (ibid.) analysis of essays produced by English native speakers. They found that writer pronouns occurred in c. 80 per cent of the texts, with an average of 3.41 pronouns per essay. In sum, not only do business students appear to underuse explicit self-reference, the distribution of pronouns also follows sub-field specific patterns. Turning now to the second factor - dispersion - further analysis revealed that authorial references cluster in particular parts of the corpus texts. Figure 3 (  page 88) highlights this patchiness, showing the distribution of the most frequent pronoun I across paper sections. 0 0,5 1 1,5 Business Economics Finance Marketing Frequency per 1,000 words I me my our us we 88 Christine S. Sing 44 (2015) • Heft 1 Fig. 3: Section-based distribution of self-reference using I across subcorpora As shown, I chiefly occurs in the introductory and concluding sections while being particularly infrequent in the body of the seminar paper. Moreover, the distribution of I shows sub-field specific variation, being more pervasive in the business and marketing subcorpora than in the other two. In order to integrate these findings with the typologies mentioned earlier, the following section-based analysis will identify the pronouns’ discourse functions as well as relating them to specific writer identities. 4.1 Discourse functions of writer-oriented pronouns in introductions As illustrated in Table 2, pronominal self-reference fulfils two main functions in seminar paper introductions: organising discourse, which accounts for c. 80% of pronoun uses, and, to a much lesser extent, engaging the reader. Function Total % Business Econom. Finance Marketing Organising discourse 448 80 88 61 89 81 Engaging the reader 77 14 12 20 9 12 Referring to personal experience 13 2 0 15 0 0 Expressing claims 9 2 0 2 1 3 Describing the research process 10 2 0 2 1 4 557 100 100 100 100 100 Table 2: Discourse functions of writer-oriented pronouns in introductions For ease of cross-corpus comparison, the table provides the raw overall frequencies of the pronouns (cf. column ‘total’) as well as percentages. The findings demonstrate that, in constructing their identities as ‘architects’, the student writers largely abide by genre conventions, according to which introductions are to familiarise readers with the topic and to provide an outline of the paper’s structure. Typical corpus examples include: In the second part we will present […]; In this paper I shall first give […], then […]. 0 2 4 6 8 Business Economics Finance Marketing Frequency per 1,000 words Introduction Body Conclusions ESP writing and disciplinary identity 89 44 (2015) • Heft 1 Authorial self-reference thus relies on so-called frame markers (emboldened) (H YLAND 2005), i.e. expressing discourse acts such as sequencing or labelling stages, which tend to be formulaic in nature. Interestingly, the discourse functions identified in the economics subcorpus deviate from those in the other subcorpora. Here the role of architect is much weaker in favour of engaging the reader or referring to personal experience. Building a relationship with the reader is chiefly encoded by the inclusive uses of we, as exemplified in We are living in a fast changing world […]; We all know that […]. These generic uses of we clearly include both writer and reader; in fact, they include a rather large, unspecific group of people. Further uses of we will be discussed below, elaborating on their field-specific uses. Prior to moving on to the discourse functions in the body of the papers, I will briefly comment on the function referring to personal experience, which is only attested in the economics subcorpus. In several papers students provide anecdotal evidence for the relevance of the topic or supporting their motivation for choosing it. For example, students report on their study-abroad experiences, which prompted them to work on a particular topic. 4.2 Discourse functions of writer-oriented pronouns in the main parts As shown in Table 3, the most pervasive discourse function in the body of seminar papers is engaging the reader, which accounts for more than half of the pronoun uses. The data in the table furthermore suggest complementary uses of self-reference in papers dealing with business and finance on the one hand and those concerned with economics and marketing on the other. In the former case, functions such as organising discourse and describing the research process tend to be overused while engaging the reader is clearly overused in economics and marketing. Function Total % Business Econom. Finance Marketing Organising discourse 236 28 34 20 48 20 Engaging the reader 450 52 42 66 24 61 Referring to personal experience 10 1 2 0 1 2 Expressing claims 126 15 11 13 20 15 Describing the research process 36 4 11 1 7 2 858 100 100 100 100 100 Table 3: Discourse functions of writer-oriented pronouns in the main parts Textual organisation in the body of the paper chiefly serves to outline the structure of a particular (sub)section or to announce the next stage in the line of argumentation. Cases in point are: Typologies of Ecopreneurs [=section heading] In this section I shall outline […]; In the following chapter I describe […]. 90 Christine S. Sing 44 (2015) • Heft 1 Moreover, there are few explicit references to the research activity or process. This, at least partially, results from the writing task itself, according to which seminar papers are almost entirely literature-based. Due to this task-related influence, the ‘Methodological I’ does not apply. This said, there are some references to the research activity itself, accounting for four per cent of writer-oriented pronoun uses, such as During my research I came across several approaches […] or In doing my research I was not able to find literature […]. Moving on to the most pervasive function of engaging the reader, it is important that, in ‘writer-responsible cultures’ (D AHL 2004) as represented by English, the writer builds a relationship with the reader. The inclusive uses of we, us and our are engagement markers, helping to establish such a relationship, in which the writer adopts the role of guide - as in phrases such as directly lead us to or Let us consider - or representative. The latter can be argued to cover several communicative functions. Unlike the examples mentioned in section 4.1, where we was shown to represent a large, undefined group, the following uses relate to a specific group, possibly the disciplinary community: […] we need a definition […]; According to NAME (1999) we can call them opportunity costs […]. In the absence of co-authored papers, these inclusive uses of we are particularly interesting. Here, inclusive reference has two functions: It expresses communality, suggesting that there is a mutual understanding between the student author and his/ her reader. More importantly, inclusive we serves as a hedging device, reducing the author’s responsibility for the contents by ‘drawing the reader in’ and/ or attributing the contents to a source. It was surprising to see that only 4% of authorial self-references fulfil the vital function of evaluating existing claims. This role of ‘I as opinion holder’ is the second most assertive role in the typology shown in Figure 1 above. If articulated at all in their writing, students’ views tend to be rather strong and are not harmonised with the contents, thus lacking appropriateness and measure. Typical examples include: […] based on SOURCE - I can support these findings or Although I agree with Kant that […]. If these student writers find it challenging to adjust the strength of their claims to the contents, they seem more comfortable giving recommendations as in I recommend COMPANY continues to […] or […] I have several ideas that could help […]. 4.3 Discourse functions of writer-oriented pronouns in conclusions Table 4 illustrates the main functions of authorial reference in the concluding sections of seminar papers: engaging the reader (example 1), expressing claims (2), describing the research process (3) and organising discourse (4). (1) […] have to deal with it. We should not forget this! (2) […] For this reason I think that cross-cultural […]. (3) […] business model. I have shown that private equity […]. (4) […] In conclusion, I sum up some of the key points […] ESP writing and disciplinary identity 91 44 (2015) • Heft 1 These discourse functions can be related to the conventional, genre-based move realisations identified in H ÜTTNER ’s (2008) study of Austrian student paper conclusions: provide a summary statement/ review; qualify and evaluate the paper/ results; provide a personal reflection; provide a wider outlook/ embed paper. Function Total % Business Econom. Finance Marketing Organising discourse 46 13 14 15 6 17 Engaging the reader 121 34 55 37 24 25 Referring to personal experience 25 7 0 6 2 9 Expressing claims 89 25 20 26 33 23 Describing the research process 63 18 9 15 31 18 Stating results/ claims 12 3 2 1 4 8 356 100 100 100 100 100 Table 4: Discourse functions of writer-oriented pronouns in conclusions Prior to wrapping up, let us briefly consider a final set of examples taken from the finance subcorpus, for which the analysis produced slightly different results. While organising text and engaging the reader are clearly underused, the discourse functions expressing claims and describing the research process are overused, for example what I really consider very useful or I have demonstrated that such standards as the NAME Accords […]. These examples indicate that the papers in the finance subcorpus seem to be organised differently. Indeed, there are several papers dealing with a case study, which entails a structure around the problem-solution pattern. This may have prompted students to articulate recommendations and to describe how they proceeded with the task. 5. Discussion and implications for ESP writing instruction I will now briefly summarise the main findings and relate them to the pedagogical issues raised in section 2. In general, the business students in this study assume low visibility in their writing. Not only are writer-oriented pronouns fairly infrequent in the corpus, they also have extremely low range, occurring in a mere quarter of the seminar papers, except for the pronouns I and we, which have a coverage of c. 60%. Furthermore, the pronouns are not evenly distributed across papers, clustering in the introductory and concluding sections, which may also explain the pervasiveness of less powerful authorial roles such as ‘architects’ rather than ‘opinion holders’. This self-effacing tendency can be ascribed to several factors. First and foremost, the weak authorial presence may result from the students’ ‘autobiographical self’. They 92 Christine S. Sing 44 (2015) • Heft 1 come equipped with rather common (mis-)conceptions about academic writing, regarding it “as some monolithic entity, involving dry, convoluted, distant and impersonal prose” (T ANG / J OHN 1999: 24). According to these beliefs, impersonal style is the defining feature of expository writing, conceiving of academic style as ‘author-evacuated prose’ (G EERTZ 1988). From this perspective, pronominal self-reference is a taboo subject and, more importantly, is regarded as ‘unscientific’ and ‘inappropriate’, particularly coming from students. This ‘textbook stereotype of impersonality’ (H YLAND 2002b: 352) continues to be advocated in introductory books: “There are several points in an academic text where the writer may […] use the first person form to announce his/ her presence. In general, however, it is advisable to let the content do the talking” (S IEPMANN et al. 2008: 13). Another contributing factor are the literacy traditions and epistemologies of the students’ L1, which seem to influence how these students approach an L2 writing task. The amount of metadiscourse used in a text is both languageand field-specific. English certainly represents a ‘writer-responsible culture’, in that writers are expected to ensure the comprehensibility of texts. German, by contrast, relies less on metadiscourse, which explains the relative scarcity of pronominal self-reference in German academic writing (S TEINHOFF 2007). Relating the present findings to the cline of authorial presence introduced above (Figure 1), it is evident that the data in this corpus suggest a slightly different ranking. The section-based analysis in particular has demonstrated that organising discourse, the weakest or ‘low-stakes function’ ( HYLAND ibid.: 353), chiefly occurs in the highly formalised introductory and concluding sections, in which students can be argued to perform standardised, routine functions in their writing, albeit in abidance of genre conventions. Moving up the assertiveness scale, the writer identities of guide and representative are performed by engaging the reader. The latter has been shown to cover a greater variety of functions in this corpus, most notably by means of the inclusive uses of we. It was found that inclusive reference fulfils two main functions: it integrates the reader into the writing to the extent that it appears as a team project and collaborative effort; it functions as a hedging device. For these students, assignment writing clearly is a potentially face-threatening act, whose impact can be cushioned by sharing the responsibility for content or procedural issues with the reader. The most powerful role attested in this data is expressing claims, which is, however, subject to considerable field-specific variation. This function is most pervasive in the finance subcorpus, resulting from the fact that several papers involved case studies, which may prompt this kind of response from students: Most claims are in fact articulated as recommendations, which may be more congruent with the professional roles that these business students aspire to and which are more likely to be those of consultants than scholars (G RUBER 2004). Given the absence of any formal writing instruction prior to producing the assignment, it is surprising to see that students still seem to have acquired genre knowledge, at least concerning the move realisations in particular paper sections. However, this lack of induction programme appears to be the rule rather than the exception in many ESP writing and disciplinary identity 93 44 (2015) • Heft 1 HE settings (P OHL 2014: 128). Instead, genre knowledge tends to be acquired implicitly, by means of ‘the pedagogy of osmosis’ (T URNER 2011: 21). Consequently, students tend to resort to other learning strategies in order to master the writing task at hand. They seek to emulate the expert models they find in the literature they use for their seminar papers, which is rather common in business and economics, owing to the fact that the standardised format of English-speaking RAs using the IMRaD (Introduction − Method − Results − Discussion) structure lends itself particularly well to imitation (L IMBURG 2014). Turning now to revisit the three pedagogical challenges mentioned above, it is evident that these ESP undergraduate writers have effectively constructed another strong identity, the learner self, which enables them to reconcile the conflicting roles and identities experienced in the setting of a business school. The first challenge occurs on the level of language learning, where the students respond to the linguistic challenges of general, academic and business English by means of rote learning. Being L2 users of English, they frequently rely on pre-fabricated, highly formulaic expressions. Second, the status accorded to writing instruction in the ESP curriculum poses yet another challenge. Indeed, students are expected to make a great conceptual leap from the educational genres anchored in the curriculum to the genres that hold importance in the workplace. As a result, the discrepancy between the training text used for writing practice and the professional target genres could not be more apparent. If assignment writing requires students to ‘fictionalise’ (P OHL 2009), this is particularly valid for this ESP setting. Third, the culture clash experienced by ESP undergraduates in their learning environment results from the situatedness of literacy practices. University writing seems to be the locus of glocalising forces, in that institutionalised forms of writing collide with highly standardised target genres. In attempting to model expert practices, students are (mis-)guided by the assumption of uniformity in disciplinary forms of writing. The findings thus seem to suggest that these students have been trained in epistemologies that highlight learning as the accumulation of knowledge. This ‘acquisition metaphor’ ( SFARD 1998: 5) devises the human mind as a container that is consecutively filled with knowledge. The process of learning a language is thus commodified while literacy development emphasises the mechanics of writing, regarding it as a technical skill, dissociated from social and disciplinary practices. The implications for ESP writing instruction are twofold. The acquisitition metaphor should be complemented by the participation metaphor, which views learning as the process of becoming a participant member of the target community ( SFARD ibid.: 6). Second, ESP students should be sensitised to distinct literacy demands, thereby empowering them to switch more easily between roles and practices. ESP writing “must be understood as the crucial process by which students make sense not only of the subject knowledge [...], but also how they can make it mean something for themselves” (H YLAND 2013: 97). It seems that the student writer, in stating that “my own summary is definitely less scientific but more easy to grasp”, claims this authority, promoting an identity of pragmatic consultant rather than scholarly writer. 94 Christine S. Sing 44 (2015) • Heft 1 Literature C HARLES , Maggie (2006): “Phraseological patterns in reporting clauses used in citation: A corpusbased study of theses in two disciplines”. In: English for Specific Purposes 25, 310-331. D AHL , Trine (2004): “Textual metadiscourse in research articles: a marker of national culture or of academic discipline? ” In: Journal of Pragmatics 36, 1807-1825. D UDLEY -E VANS , Tony / S T J OHN , Marie Jo (1998): Developments in ESP. A Multi-disciplinary Approach. Cambridge: Cambridge University Press. E HLICH , Konrad (2003): “Universitäre Textarten, universitäre Struktur”. In: E HLICH , Konrad / S TEETS , Angelika (eds.): Wissenschaftlich schreiben - lehren und lernen. Berlin: De Gruyter, 13-29. F LOWERDEW , Lynne (2005): “An integration of corpus-based and genre-based approaches to text analysis in EAP/ ESP: countering criticisms against corpus-based methodologies”. In: English for Specific Purposes 24, 321-332. G AVIOLI , Laura (2005): Exploring Corpora for ESP Learning. Amsterdam: Benjamins. G EE , James Paul (1996): Social Linguistics and Literacies. Ideology in Discourses. London: Taylor & Francis. G EERTZ , Clifford (1988): Works and Lives. The Anthropologist as Author. Stanford Calif.: Stanford University Press. G NUTZMANN , Claus / R ABE , Frank (2014): “‘Theoretical subtleties’ or ‘text modules’? German researchers' language demands and attitudes across disciplinary cultures”. In: Journal of English for Academic Purposes 13, 31-40. G RUBER , Helmut (2004): “Scholar or consultant? Author-roles of student writers in German business writing”. In: R AVELLI , Louise J. / E LLIS , Robert (eds.): Analysing Academic Writing. Contextualized Frameworks. London: Continuum, 45-65. H ARWOOD , Nigel (2005a): “‘We do not seem to have a theory … The theory I present here attempts to fill this gap’: inclusive and exclusive pronouns in academic writing”. In: Applied Linguistics 26, 343-375. H ARWOOD , Nigel (2005b): “‘I hoped to counteract the memory problem, but I made no impact whatsoever’: discussing methods in computing science using I”. In: English for Specific Purposes 24, 243-267. H ÜTTNER , Julia (2008): “The genre(s) of student writing: developing writing models”. In: International Journal of Applied Linguistics 18, 146-165. H YLAND , Ken (2002a): “Authority and invisibility: authorial identity in academic writing”. In: Journal of Pragmatics 34, 1091-1112. H YLAND , Ken (2002b): “Options of identity in academic writing”. In: ELT Journal 56, 351-358. H YLAND , Ken ( 1 2005): Metadiscourse. Exploring Interaction in Writing. London: Continuum. H YLAND , Ken (2013): “ESP and writing”. In: P ALTRIDGE , Brian / S TARFIELD , Sue (eds.): The Handbook of English for Specific Purposes. Malden, Mass: Wiley-Blackwell, 95-113. I VANIČ , Roz (1998): Writing and Identity. The Discoursal Construction of Identity in Academic Writing. Amsterdam: Benjamins. K RUSE , Otto (2006): “The origins of writing in the disciplines: Traditions of seminar writing and the Humboldtian ideal of the research university”. In: Written Communication 23, 331-352. L EA , Mary R. / S TREET , Brian V. (1998): “Student writing in higher education: An academic literacies approach”. In: Studies in Higher Education 23, 157-172. L IMBURG , Anika (2014): “Imitationslernen in den Wirtschaftswissenschaften: Unterschiede zwischen Examensarbeiten und Forschungsartikeln”. In: Zeitschrift Schreiben. http: / / www.zeitschrift-schreiben.eu (15 October 2014). ESP writing and disciplinary identity 95 44 (2015) • Heft 1 N ESI , Hilary / G ARDNER , Sheena (2012): Genres Across the Disciplines. Student Writing in Higher Education. Cambridge: Cambridge University Press. P AQUOT , Magali (2010): Academic Vocabulary in Learner Writing. From Extraction to Analysis. London: Continuum. P ETERS , Pam / S MITH , Adam / M IDDLEDORP , Jenny / K ARPIN , Anne / S IN , Samantha / K ILGORE , Alan (2014): “Learning essential terms and concepts in Statistics and Accounting”. In: Higher Education Research & Development 33, 742-756. P OHL , Thorsten (2014): “Entwicklung der Schreibkompetenzen”. In: F EILKE , Helmuth / P OHL , Thorsten (eds.): Schriftlicher Sprachgebrauch - Texte verfassen. Baltmannsweiler: Schneider Hohengehren, 101-140. P OHL , Thorsten (2009): Die studentische Hausarbeit. Rekonstruktion ihrer ideen- und institutionsgeschichtlichen Entstehung. Heidelberg: Synchron. R AYSON , Paul (2003): Matrix: A Statistical Method and Software Tool for Linguistic Analysis Through Corpus Comparison. Lancaster. S FARD , Anna (1998): “On two metaphors for learning and the dangers of choosing just one”. In: Educational Researcher 27, 4-13. S HELDON , Elena (2009): “From one I to another: Discursive construction of self-representation in English and Castilian Spanish research articles”. In: English for Specific Purposes 28, 251-265. S IEPMANN , Dirk / G ALLAGHER , John D. / H ANNAY , Mike / M ACKENZIE , Lachlan J. (2008): Writing in English. A Guide for Advanced Learners. Tübingen: A. Francke Verlag. S ING , Christine S. (2013): “Shell noun patterns in student writing in English for specific academic purposes (ESAP)”. In: G RANGER , Sylviane / G ILQUIN , Gaëtanelle / M EUNIER , Fanny (eds.): Twenty Years of Learner Corpus Research. Looking Back, Moving Ahead. Louvain-la-Neuve: Presses universitaires de Louvain, 411-422. S TARFIELD , Sue / R AVELLI , Louise J. (2006): “‘The writing of this thesis was a process that I could not explore with the positivistic detachment of the classical sociologist’: Self and structure in New Humanities research theses“. In: Journal of English for Academic Purposes 5, 222-243. S TEINHOFF , Torsten (2007): “Zum ich-Gebrauch in Wissenschaftstexten”. In: Zeitschrift für Germanistische Linguistik 35, 1-26. T ANG , Ramona / J OHN , Suganthi (1999): “The ‘I’ in identity: Exploring writer identity in student academic writing through the first person pronoun”. In: English for Specific Purposes 18, 23-39. T ARDY , Christine M. (2012): “Writing and language for specific purposes”. In: C HAPELLE , Carol A. (ed.): The Encyclopedia of Applied Linguistics. Oxford: Wiley-Blackwell, 6266-6274. T URNER , Joan (2011): Language in the Academy. Cultural Reflexivity and Intercultural Dynamics. Bristol, UK: Multilingual Matters.