ELF in the making? simplification and hybridity in abstract writing

1 Introduction

The overall use of English as the international language of scientific communication involves, among many other sociological, political, economic, and ideological implications, an increasing pressure on academics to disseminate their research internationally in English. Social sciences, together with the natural or hard sciences, seem to be the most productive macroareas as far as publication in English is concerned: more than 95 % of indexed natural science journals and 90 % of social science journals use all or some English (data provided by Thomson Reuters 2008, in Lillis and Curry 2010: 9). Furthermore, it seems to be the case that the proportion of papers authored by non-native academics is on the increase (Ferguson et al. 2011: 43). To publish in English is, according to Hyland (2015: 45), “more than a choice; it has come to designate research of a high academic quality deemed worthy of a place in globally accessible peer-reviewed journals.”

It is undeniable, then, that the academic community speaks English, but not necessarily as native speakers. The fact is that all writing in international journals is, by definition, for an international rather than an ENL (English as a native language) audience. In this scenario, “academic brokers” (Lillis and Curry 2010) influence opportunities for gaining access to English-medium journal publication and significantly contribute to the shaping of textual knowledge. These academic brokers (translators, proofreaders, copyeditors) sometimes act as “gatekeepers” that exert control on what can or cannot be published. A recent study by Hyland (2015) analyzing the affiliations of editorial boards in several disciplines showed, for instance, that in the discipline of Sociology, the focus of the present study, Anglophone editorial members amount to 67.3 % of those belonging to the five most prestigious journals in the field (highest five-year impact factor) and only 32.7 % are non-Anglophone. That this fact may have some kind of impact on what is finally published and what is not is undeniable. This situation of non-equality raises issues such as the disadvantageous position of non-Anglophone writers having (or choosing) to publish their papers in English (see, e. g., Canagarajah 1996, Canagarajah 2002; De Swaan 2001; Ferguson 2007; Flowerdew 2007, Flowerdew 2008; Tardy 2004), which also has the effect of depriving fields of enquiry of insights and perspectives that could contribute to the advancement of the discipline (Seidlhofer 2012; Bennett 2014). However, this is not a one-way dynamic and, as Seidlhofer (2012: 395) points out: “the compliance in the non-Anglo world with Anglo norms and Anglo ways of doing things is akin to ‘pre-emptive obedience’.”

As a natural consequence of this inescapable need for non-Anglophone scholars to disseminate their research in English, the study of academic written genres from an intercultural perspective has attracted a lot of scholarly attention, the ultimate aim being to help researchers from varied cultural contexts present their research in English for international publication. A great deal of research has focused on the study of the genre par excellence in the academia, the research article (RA), in a wide range of languages and contexts of publication, very prominently in English and Spanish, e. g.: Carciu (2009), Lorés-Sanz (2011a), Lorés-Sanz (2011b), Mur-Dueñas (2007), Mur-Dueñas (2011), Murillo (2012), Pérez-Llantada (2010), Salager-Meyer et al. (2003), Sheldon (2009), and Sheldon (2011). Research has also been carried out on the RA abstract contrasting English with another language, e. g., English and Spanish (Martín-Martín 2002, Martín-Martín 2005; Divasson and León Pérez 2006; Lorés-Sanz 2006, Lorés-Sanz 2009; Burgess and Martín-Martín 2010; Alonso-Almeida 2014); English and Italian (Diani 2014); English and French (Van Bonn and Swales 2007); and English and Arabic (Alharbi 1997; Alharbi and Swales 2011). All this text-based research, which follows the tradition initiated by Purves’s (1986) analysis of the writing styles from twelve different countries, has been extremely relevant to highlight rhetorical differences across language and disciplines, and the results obtained have informed English for Academic Purposes (EAP) materials and instruction.

These studies are deeply rooted within the tradition of Contrastive Rhetoric, stemming from the publication of Robert Kaplan’s 1966 groundbreaking article “Cultural thought patterns in intercultural education.” Contrastive Rhetoric, defined as a field that “examines differences and similarities in writing across cultures” (Connor 2001: 75), has as its ultimate objective to test the hypothesis that different languages have different rhetorical systems which manifest themselves in different ways of organizing ideas. Kaplan’s work was, by no means, without limitations, the most relevant here being the assumption of “negative” L1 interference in L2, on the understanding that non-Anglophone writers have to use L1 rhetorical conventions in their English texts.

Connor (2004a, 2004b) problematized the concept of Contrastive Rhetoric by pointing out the negative connotation that the term had gained over the years. In her view (Connor 2004a: 272), “contrastive rhetoric is often characterized as static, and is linked to contrastive analysis, a movement associated with structural linguistics and behaviourism.” Thus, to distinguish between this “static” model and the new advances that had been made, Connor proposed the use of the term “Intercultural Rhetoric” to refer to the “current dynamic models of cross-cultural research” (Connor 2004a: 272), which includes the majority of text and genre analysis studies and allows for the study of interactions in intercultural settings.

Whereas Contrastive Rhetoric is a traditional perspective applied to EAP, exploring the processes of writing and publishing in English as a Lingua Franca (ELF) is quite a recent focus of study. According to Jenkins et al. (2011), apart from a few occasional studies between the late 1980s and the late 1990s, ELF remained a minority interest until the beginning of the third millennium. From that moment on, research on ELF has produced some of the most inspiring and thought-provoking studies within the broad field of applied linguistics. ELF, defined as a shared common language between users with different language backgrounds, has been mainly explored in spoken interaction, more saliently at the phonological (i. e., Jenkins 2000), lexicogrammatical (i. e., Cogo and Dewey 2012), and pragmatic level (i. e., House 2009; Mauranen 2009). ELF, House (2012: 173) states, “is a new type of English, a hybrid language, a kind of ‘pluralized English’ that accommodates diverse speakers’ needs, norms and values.”

The written academic context has recently become fertile ground for ELF studies (e. g., Carey 2013; Mur-Dueñas 2013). The compilation of the WrELFA corpus (Written English as a Lingua Franca in Academic settings), ^[1] collected at the University of Helsinki (Finland), is proof of the interest that the academic written environment is generating for ELF research.

The present study should be understood as a contribution to this more recent line of research within ELF studies. As some scholars (Mauranen 2012; Hartse and Kubota 2014) claim, so far there is little empirical knowledge about how variations are tolerated in academic writing publications. Here I intend to explore the written rhetorical forms that may emerge when English is used as an international language of academic communication, in which speakers from different similects are involved. Similects, L1 based lects, arise in parallel and not in mutual interaction, and originate in cross-linguistic influence, therefore comprising “a renewable resource for the mix that ELF is made of” (Mauranen 2012: 29). English used as a lingua franca is, in fact, the result of contact and evolution processes and situations. Language contact does not necessarily involve face-to-face interaction among groups of speakers using different languages. It may also involve non-personal contact of persons with texts available in the written medium (Braunmüller and House 2009), as is the case in contexts of academic publications at an international level. Thus, although research on language contact has traditionally assumed the existence of a well-defined geographical scenario where contact takes place, more recent research allows for a wider conceptualization of the notion of “space,” incorporating “larger areas of the world” (Muysken 2010: 268). It can be inferred, therefore, that written instances of English used as an international language of academic communication may also be opened to research from this perspective. The use of ELF creates a sort of “virtual scenario” where processes of language contact, variation, and change take place, taking into account, as Seidlhofer (2011: xi) points out, that English is unlike other languages in the sense that “most of its users do not mainly, or even at all, engage with native speakers.”

Rhetorical structure may well be one of those aspects open to variation and change, as a result of contact among ELF users. According to Mauranen et al. (2010), most studies in contrastive rhetoric reveal that the strongest influence of different academic writing cultures is shown in the textual organization and textual preferences displayed. However, “language editors saw as their main task the correction of lexicogrammatical errors, but they abstained from tackling the textual organization and pragmatic aspects of the texts they were revising” (Mauranen et al. 2010: 638). Hartse and Kubota (2014), reflecting on their attempts as gatekeepers to “pluralize English” in L2 writing, also focus on lexis and grammar, and state: “in high-stakes writing, lexicogrammatical variation from standard English is being noticed by native and non-native English speakers alike, and it can quickly lead to sentences, texts, or authors being perceived as deficient” (Hartse and Kubota 2014: 73). Therefore, if the focus for “language brokers” (Lillis and Curry 2010) lies mainly at word and sentence level, rhetorical aspects may show more patently the variations contributed by non-Anglophone writers, as these variations will be less “noticeable”. Even if, as Mauranen (2012: 3) states, “some strata or systems of language are most susceptible to change than others, and lexis has usually found to be the site of rapid spread of influences, in contrast to structure, which is more resistant,” does that mean that rhetorical structure is closed to variation and change?

Another perspective which may contribute to the present discussion is the perception of genres as dynamic constructs. I here adhere to Hyland’s (2004) view that genres are, to a certain extent, open to choice, and although certain regularities or conventions are expected by readers as regards grammar, vocabulary, content, organization, etc., a certain degree of variation is also allowed: “Genre knowledge comprises an awareness of variation – an understanding that deviations are acceptable to the extent that they do not cancel out function or appropriateness” (Hyland 2004: 64).

The academic genre I will focus my research on is the research article abstract written by non-Anglophone users of English. As Swales and Feak (2009: 1) state, an “information explosion” has taken place in the academic world, which accounts for a massive number of peer-reviewed journals contained in major databases and thousands of research papers being published every year. This amount of information has become almost unmanageable. Thus, the research article abstract is now playing a major role as a time-saving and information-managing device.

Having conceived the present study as a preliminary approach to the, to my knowledge, rather unexplored path of variation in academic written genres in ELF situations, I will focus my analysis on a single disciplinary field, Sociology, traditionally dominated by Anglophone writers, with their most prestigious journals hosted either in the United States or in Great Britain. Here I explore the evolution of one single journal, Social Science Research, and the way this prestigious journal has evolved with respect to the contributions by non-Anglophone speakers.

My contention is that the fact that in recent years an increasing number of non-Anglophone academics are contributing to this field of study in high impact English-medium publications (see Hyland 2015) is presumably having an influence on the rhetorical re(shaping) of one of the more common academic genres, the abstract. Sociology abstracts published in this prestigious site, as written by users of English as an additional language (EAL), may not be conforming to the rhetorical Anglophone conventions and may be displaying new, hybrid rhetorical structures which are the result of the interaction and contact between English and their L1.

2 Sociology abstracts: the case of the journal Social Science Research

Social Science Research (http://www.journals.elsevier.com/social-science-research) is an American-based journal by Elsevier which publishes papers devoted to quantitative social science research and methodology. It was first launched in 1972 and the members of its Editorial Board and the Board of Advisory Editors are all affiliated to American institutions.

Together with the American Journal of Sociology and the British Journal of Sociology, it is one of the most prestigious sites of publication for academics in the field. At the moment it publishes six volumes per year and its journal metrics are the following: ^[2]

1. Source Normalized Impact per Paper (SNIP): 1.266

2. SCImago Journal Rank (SJR): 1.047

3. Impact Factor: 1.295

4. 5-Year Impact Factor: 1.912

Please write your text in good English (American or British usage is accepted, but not a mixture of these). Authors who require information about language editing and copyediting services pre- and post-submission please visit http://www.elsevier.com/languageediting or our customer support site at http://epsupport.elsevier.com for more information.

The Elsevier language editing service the document refers to is advertised as follows: ^[4]

Language editing and translation services

In today’s competitive environment, it is essential that the English language used in your paper is of a high quality. Your research may be critical, but poor English is a common cause of delays and initial rejections.

Do you need English-language support? We can edit your manuscript within 7 business days. Our long history of publishing peer-reviewed scientific journals has equipped us to ensure your English is free of grammatical and spelling errors. Learn more about Elsevier’s English Language editing service and what you can expect from us.

We offer:

1. Native English speakers from top universities; expert input from PhDs or PhD candidates matched to your field of study

2. An edited manuscript containing correct scientific English (US or UK)

3. A self-service website with easy article upload and retrieval

4. Clear editing rates

It is interesting to notice that the “high quality” and “good English” mentioned in the instructions is immediately associated to “native English speakers” and “correct” scientific English.

This is the type of gatekeeping function that Seidlhofer (2004: 222) refers to when talking about the practice of subjecting non-native academic writing to correction in order for it to conform to native conventions: “[Journal editors] exert a gatekeeping function based not on academic expertise but purely on linguistic criteria whose relevance for international intelligibility has not actually been demonstrated.”

Instructions about abstract writing are also provided in the journal webpage: ^[5]

Abstract

A concise and factual abstract is required. The abstract should state briefly the purpose of the research, the principal results and major conclusions. An abstract is often presented separately from the article, so it must be able to stand alone. For this reason, References should be avoided, but if essential, then cite the author(s) and year(s). Also, non-standard or uncommon abbreviations should be avoided, but if essential they must be defined at their first mention in the abstract itself.

Abstracts should not exceed 100–150 words.

That is, as happens in this type of instructional information about abstract writing, there is an inherent recommendation as to the rhetorical pattern which should be used, and which should include aims, results, and conclusions.

At the time of its launch in 1972, Social Science Research published one volume per year with four issues per volume. This changed in 2010, when six issues started to be published per volume and it resulted in a significant increase in the number of RAs published annually. This was a turning point, probably in response to the high number of proposals the editors received and which was a consequence, at the same time, of the globalization of scientific knowledge dissemination through the Internet and the spread of major databases.

Thus, whereas in 1972 (vol. 1) 26 papers were published, in 2014, 92 saw the light in Social Science Research (vols. 43–48). There was a steady flow of articles published since the beginning until about 2003, when the number of papers rose sharply up to 2012.

Contributions by non-Anglophone speakers have increased accordingly and have undergone an exponential growth. Whereas in the early 1970s, only an average percentage of 4.67 % of the papers had been written by non-Anglophone speakers, in 2013 the proportion was 20.6 %, and in the year 2014, 23.9 %. Thus, since the launch of the journal in 1972, there has been a steady increase in the number of contributions by non-native speakers, with some ups and downs. A decisive factor which explains the significant rise in the number of RAs by non-Anglophone writers is, without doubt, the editorial decision in 2010 to increase the number of published articles and, accordingly, the number of issues per volume.

These data are very much in line with the information reported in Hyland (2015) about the provenance of first-named authors for articles in the top five journals by impact factor in Sociology (among other disciplines) in the years 2000 (79.0 % NES and 21.0 % EAL ^[6]) and 2011 (69.8 % NES and 30.2 % EAL).

It is interesting to notice that Social Science Research editors take pride in the varied linguacultural origin of their contributors: ^[7]

Authors

Our authors submit research articles from all over the world. Authors collaborate frequently with peers at different institutions and in different countries. This graph indicates the number of primary corresponding authors at the country level in the last five years.

In a way, therefore, they are somehow equating the international character of their contributors to prestige. Thus, even a map indicating the origin of the contributors and the number of RAs that they have contributed within the last five years is offered on the journal webpage (see Figure 1).

Figure 1:

Indication of location of corresponding authors (http://journalinsights.elsevier.com/journals/0049-089X/authors accessed 24 July 2015).

In the same section a counting is offered of the affiliation country of each corresponding author in the last five years, with 81.59 % authors affiliated to Anglophone institutions (United States, United Kingdom, Canada) and 18.4 % to non-Anglophone, in countries such as the Netherlands, Germany, Israel, Belgium, Sweden, and Hong Kong, among others, both in Europe and Asia.

If contributions by non-Anglophone users of English are increasing (and amount to almost 20 % in the last five years), it may be the case that some rhetorical aspects of the abstracts being published do not fully adjust to Anglophone conventions of text construction and, although gatekeepers take American/British English as models, other rhetorical uses of the language may be permeating. It will then be my purpose in the present study to explore the rhetorical structure of abstracts contributed by non-Anglophone writers publishing in Social Science Research in the last five years in order to gain insights as to the variety of rhetorical organizations which may be permeating and, therefore, reshaping conventional abstract writing. In the end, my aim is to explore whether “emerging patterns” (Jenkins et al. 2011: 289) can be identified in terms of rhetorical structure as a result of the linguistic interaction and contact between the participants in this specific ELF context.

3 Corpus description and methodology

For the purposes stated above, I selected a corpus of 66 abstracts published in Social Science Research in the last three years, corresponding to the volumes 41 (2012), 42 (2013), and 43–48 (2014) and which include all the abstracts whose author/s are not affiliated to Anglophone institutions. ^[8] Identifying the authors’ native language is a conflictual point. Acknowledging the methodological weakness it implies, I have here nevertheless followed conventional practice in cross-cultural studies, in which authors’ native language is generally (assumed to be) identified by institutional affiliation and family name (i. e., Alharbi and Swales 2011; Salager-Meyer et al. 2003; Van Bonn and Swales 2007; Vassileva 2000).

The composition of the corpus is shown in Table 1.

As shown, not all the similects contribute to the same extent. However all the abstracts have been examined as it is understood that all of them illustrate valid ways of displaying the rhetorical structure of Sociology abstracts.

The analysis of this written corpus will then allow me to provide preliminary answers to the following research questions:

1. In general terms, do EAL writers in the context under study strictly adhere to the conventional rhetorical structure of Social Sciences abstracts described in the literature?

2. If/when not, are abstracts by authors from different similects finding ways of modifying conventional rhetorical patterns?

To explore rhetorical structure, the notion of “move” (Swales 1990) is important. Essentially, a move in a text is a functional unit, used for some identifiable rhetorical purpose: “a discoursal or rhetorical unit that performs a coherent communicative function in a written or spoken discourse” (Swales 2004: 228). It is a functional, not a formal unit. However, it may sometimes be characterized by grammatical or lexical indicators (i. e., negative elements to indicate a gap). For the purposes of the present study I will here introduce the rhetorical organization of research article abstracts for the Social Sciences described by Lewin (2010) in her book, Writing Readable Research: A Guide for Students of Social Science. Lewin’s proposal is advantageous for the present study in the sense that, to my knowledge, it is the only description of a rhetorical structure for abstracts in the discipline under study. Variations on IMRD (Introduction–Method–Results–Conclusion) structure have been proposed for abstracts in other disciplines such as Applied Linguistics (i. e., Lorés-Sanz 2004; Pho 2008; Santos 1996) and Medicine (Salager-Meyer 1992). Lewin establishes a six-move structure, based on Swales’s, which includes the following moves:

1. Relevance

2. Aim

3. Gap

4. Method

5. Results

6. Conclusions/Implications

As observed, the basic difference between Lewin’s proposal and the IMRD structure as described by Swales for the rhetorical study of RAs is that the inner three-move structure of the Introduction in Swales’s description is given the status of three separate moves (Relevance, Aim, and Gap) in Lewin’s proposal. ^[9] This structure was identified by Lewin in twelve empirical abstracts published in Demography, the journal of the Population Association of America, and was then checked again, to see if it was representative of the Social Sciences, in several journals from different disciplines within this area (e. g., Sociology, Psychiatry, Psychology, Anthropology, and Social Work). Although Lewin does not make any reference to whether the authors were native speakers or not, the journals she drew the texts from are all based in Anglophone countries, and the fact that Lewin is trying to define a template makes us think that she observed abstracts written by native speakers. As part of her results, Lewin (2010:161) observed that there was a three-move pattern which was prevailing in her corpus: the pattern “objectives–method–results.”

In a previous study (Lorés-Sanz 2014), Lewin’s template was applied to a corpus of thirty Sociology research article abstracts written by Anglophone authors and published between 2003 and 2010 in the journals Social Science Research, American Journal of Sociology, and British Journal of Sociology. The identification of the rhetorical structure in terms of moves (Swales 1990) was carried out following a top-down approach (Pho 2008), based on the function or content of the text. Based on Pho’s microanalysis for applied linguistics abstracts and on Swales’s general description of sections, moves, and steps in research articles, a framework for the rhetorical analysis of the present corpus was proposed (see Table 2).

Some of the results found in the study are summarized in Table 3, which shows the percentage of presence of each rhetorical move in the corpus analyzed.

The data gathered in this previous study confirm Lewin’s insights that there is a prevailing three-move pattern (Aim–Method–Results) in Social Science empirical abstracts. Above all, Aim and Results were the more recurrent moves. They seem to be almost compulsory moves in ENL abstracts. This was quite an expected finding, since these were the moves, together with Conclusions, that journal editors more frequently requested to be included in the guidelines for abstracts. However, the other moves are also well represented in the corpus under study (60 % of the abstracts present the Relevance and the Conclusion move and 50 % include the Gap).

4 The analysis of the EAL corpus: some results

In the volumes of Social Science Research corresponding to the years 2012, 2013, and 2014, the percentage of research papers by EAL writers was 12.5 %, 17.35 %, and 23.9 % respectively, indicating a steady rise in their contribution to the journal. The hypothesis that an increase in the number of EAL users publishing in a journal traditionally in the hands of ENL writers may be bringing along variations in the rhetorical structure of a widely spread academic genre such as the abstract will be explored and illustrated below.

To start with, I analyzed the frequency of inclusion of the different moves in the corpus under study. To do so all the EAL texts (66) included in the corpus were tagged for move structure, following the functional framework for abstract analysis displayed in Table 2.

The results obtained in this initial stage of the analysis are shown in Table 4.

When compared to the results obtained for the study in abstracts by ENL writers (Lorés-Sanz 2014), similar conclusions are drawn in the sense that the three-move structure “Aim–Method–Results” also prevails in abstracts by EAL users. However, a few differences are also observed, for instance in the higher presence of Method and also in the much lower inclusion of a Gap move (only 27.3 %) and of a contextualization for the research (Relevance).

A second stage in the analysis to find out the degree of complexity in the EAL corpus was to explore the combination of the different moves. Thus, the number of moves present in each of the abstracts was counted to see which patterns emerged. Table 5 shows different degrees of complexity in abstract structure in terms of inclusion of moves and the corresponding percentages.

When contrasted with the results yielded in the ENL corpus (Lorés-Sanz 2014), interesting insights are provided, as Figure 2 shows.

Figure 2:

Comparison of degrees of complexity of rhetorical structure in EAL and ENL abstracts.

Although the three-move pattern “Aim–Method–Results” seems to be pervasive across linguacultural contexts, the internal structure of abstracts shows its own peculiarities when we look into features such as the number of moves included in each text. Thus, whereas in abstracts by ENL writers the tendency is to make arrangements which include five moves (46.6 % of the ENL texts), there is no such clear trend in EAL abstracts, which favor, to almost equal degrees, 3-move and 4-move structures (33.3 % and 30.3 %, respectively), thus showing an inclination towards rhetorical simplification, while maintaining the convention of the discipline to include aims, methods, and results.

Moreover, the exploration of the present corpus has allowed me to identify possible variations of the conventional rhetorical structure described above. These variations may be manners in which EAL users are exploiting rhetorical conventions in ways accepted by gatekeepers in spite of their not fully complying with “American or British usage.” These modified rhetorical structures display a level of hybridity which may be the result of situations of language contact. As a result of a systematic analysis of all the abstracts in the corpus, three variants have been identified, which will be illustrated here with examples:

5 Conclusions

As mentioned above, the aim of the present study was to assess whether new, hybrid, ways of constructing the rhetorical structure of research article abstracts could be emerging (and then could be identified and described) as the result of an increasing number of contributions by non-Anglophone writers to academic communication in the field of Sociology. This hypothesis has been tested here by means of a case study: the American-based journal Social Science Research, which has lately witnessed an exponential growth in the number of contributions by researchers who are not affiliated to Anglophone institutions.

Several tentative conclusions can be drawn, some more general, some more specific, which make us think of a changing scenario with regard to the textual construction of research article abstracts in the field under study.

To start with, more and more contributions by non-Anglophone writers are seeing the light in this American-based publication. It is an expected outcome, therefore, that English, as a language of international communication, is given shape in this particular case by the co-existence of the rhetorical patterns described in the literature and less conventional ones.

This outcome is indeed apparent in the analysis presented in this paper, which provides further empirical illustration of “Academic English shaped by non-native speakers” (Mauranen 2012). More specifically, the analysis of the rhetorical structure of abstracts currently being written and published by non-Anglophone academics in Sociology has shown that, almost to equal degrees, they do not simply adopt the IMRD structure typical of the Anglophone academic world, but adapt it by giving it a new rhetorical shape. Thus, on the one hand they simplify the rhetorical structure by reducing the number of moves (see Figure 2) and, on the other, they problematize this simplified structure by creating new patterns which display a higher degree of textual complexity, showing different, hybrid ways of articulating moves in non-linear patterns (e. g., cascading, chain, and focalizing structures). In this way, they give expression to their own “hybrid voices” (Mauranen 2007).

Some insights have also been gathered as to the dynamicity of such processes of hybridation when the current picture is contrasted with the one we could get in the 1970s. Not only was the number of non-Anglophone contributors anecdotal at that time. The small numbers of abstracts written by non-native scholars differed from abstracts written by Anglophone academics (mostly American and British) mainly in the reduction and simplification of the canonical IMRD structure. It seems, then, that the process of rhetorical hybridization that we are currently witnessing emerges from an initial stage of textual simplicity, which has evolved into more complex (hybrid) rhetorical ways. The new, hybrid, patterns identified here are the result of processes of contact and evolution, reshaping old forms into new forms that model English conventional patterns in innovative and creative ways.

Scenarios such as the one we have explored here, where native speakers are still the majority of contributors, may be considered of marginal interest for investigating ELF. In my view, however, when the presence of non-Anglophone writers is seen to be growing, we are in fact exploring a moving target, a changing landscape that we might label “ELF in the making.” These research sites provide a snapshot of a dynamic process of evolution and contact, a process which is worth investigating. As Hyland (2015: 25) states: “[g]lobalization offers greater opportunities for increased scholarly dialogue by broadening the corpus of academic literature, by providing new avenues for collaboration, and by opening new channels for reporting location-specific research.” The process of globalization of knowledge cannot be stopped and, consequently, collaboration and exchange among scholars will also entail opening paths for cooperative modification of conventional and assumed pragmatic, discoursal, rhetorical, and linguistic patterns in English.