Introduction

Over the last decades, increasing attention has been given to the notion of genre and its applications in language teaching and learning. In the academic context, genre is seen as a social action (Berkenkotter & Huckin, 1995; Miller, 1984), as a cultural and social construct or conventions which respond to “a recurring type of rhetorical situation” (Coe & Freedman, 1998: 137), such as purpose, audience and style. From a genre-analytic approach, a genre is primarily defined on the basis of its communicative purpose which gives it an internal structure constituted by conventionalised communicative events embedded within disciplinary practices (Bhatia, 1993; Swales, 1990, 2004). Academic texts are thus characterised by a set of functional and linguistic structures (rhetorical moves and steps) produced in accordance with the expectations and shared assumptions of particular discourse communities. As academic genres are not static but dynamic social constructs (Freedman & Medway, 1994; Hyland & Jiang, 2021) which reflect the values and demands of the particular discourse communities that shape them (Bazerman, 1988; Berkenkotter & Huckin, 1995; Hyland, 2002), we can assume that the prevalent practices of specific disciplinary communities can change over time and that these will be displayed in the academic genres that their members construct.

In today’s competitive socio-cultural context in which the ‘publish or perish’ syndrome pervades, it is not surprising to find that scholars from the various branches of knowledge increasingly make use of rhetorical strategies to persuade their peers of the value of their research. This growing need for enhancing research is thus expected to be reflected on the rhetorical practices which the members of a particular disciplinary community use nowadays when constructing an academic genre, and that may differ from the ones that were prevalent in different social and cultural contexts of previous periods of time. As is well known, most institutions of higher education worldwide require their teaching staff to publish their research in high-impact international journals, mainly English-medium journals, if they wish to acquire job promotion, financial rewards and international recognition (Hyland & Jiang, 2021, 2023; Lillis & Curry, 2010; Moreno, 2021). Over the last few years, this pressure has been put not only on those scholars who hold an academic position but also on postgraduate students who seek to complete their doctoral programmes (Flowerdew, 2022; Li & Flowerdew, 2020) and who are not completely familiar with the discourse practices favoured by their disciplinary communities since they have received little (if any) formal instruction on English for research publication purposes (ERPP). We should also mention the role of the national assessment agencies worldwide which attach a higher value to publications in international English-medium journals than to publications in national journals. This especially affects inexperienced users of English as an additional language (EAL) who not only have to strive to survive in this competitive world of academia but who also have to deal with the disadvantage that represents to write in a language which is not their L1 (Ammon, 2012; Ferguson, 2007; Flowerdew, 2019). To alleviate this situation, and considering that most researchers today are EAL scholars, there is a growing number of critical voices that argue for the acceptance of a type of varietal model based on intelligibility and flexibility rather than conformity to native English norms (e.g. Flowerdew, 2022; Seidlhofer, 2011) and which may reflect culturally determined rhetorical features such as the use of modality and evaluation in academic discourse (Pérez-Llantada, 2014). Other authors, such as Salager-Meyer (2014), have further argued for a modification of evaluation policies which should give more credit to publications in national journals or that should foster the proliferation of multilingual journals.

In an attempt to remedy the described situation, the critical view has been combined with a more pragmatic stance in the form of pedagogical initiatives and contributions to genre-based research. In a bid to offer resources and training to early career scholars and multilingual researchers, a growing number of ERPP courses and workshops, although still scarce, have been designed and are being implemented in various higher education institutions (e.g. Burgess & Cargill, 2013; Cargill et al., 2018; Corcoran & Englander, 2016; Douglas, 2015). With the aim of providing research-informed input to the pedagogy of ERPP, the theory of genre analysis (Swales, 1990, 2004) has been applied to various sections of the RA in English from different (sub)disciplines, i.e. the Introduction (e.g. Nwogu, 1997; Samraj, 2002), the Methods (e.g. Bruce, 2008; Peacock, 2011), the Results (e.g. Basturkmen, 2009; Brett, 1994) and the Discussion and/or Conclusion (e.g. Holmes, 1997; Moreno & Swales, 2018). This genre-analytical approach, from an intercultural perspective, has been extended to other studies which compare the rhetorical structure of English RAs with that of other languages such as Malay (e.g. Ahmad, 1997), Brazilian Portuguese (e.g. Hirano, 2009), Chinese (e.g. Loi, 2010; Taylor & Chen, 1991) or Spanish (e.g. Burgess, 2002; Mur-Dueñas, 2010). Following this line of research, this study focuses on a move/step analysis of the RA abstract, an academic genre of great relevance due to the important function it fulfils.

The main function of RA abstracts is informative as they provide a summary of the content of the associated paper, thus indicating to readers whether the full text merits their further attention. This is why they are considered “time-saving devices” (Salager-Meyer, 1990: 367) since this is the only part of the article that busy researchers typically read. An effective abstract accurately reflects, in a condensed form, the macrostructural components of the accompanying paper, mainly the IMRD pattern (Swales, 2004). RA abstracts also have a persuasive function as they attempt to attract the reader’s attention in order to increase the possibilities of citations. After the paper’s title, they constitute the first encounter with the text, and therefore, it is here that writers have to convince their peers (especially editors and reviewers) that they master the rhetorical conventions which are expected by the members of their disciplinary community to qualify for membership in the group. The rhetorical structure of RA abstracts has been examined extensively revealing the prevalent and optional moves and their lower-level constituents (sub-moves of steps) across a large variety of (sub)disciplines such as applied linguistics (dos Santos, 1996; Pho, 2008), literature (Tankó, 2017), medicine (Anderson & Maclean, 1997; Salager-Meyer, 1990), psychology (Hartley, 2003) or sociology (Lorés, 2016). Some of these studies have also compared the discourse features and rhetorical moves of RA abstracts written in English with those of other languages, such as Arabic (Alharbi & Swales, 2011), Chinese (Hu & Cao, 2011), French (Van Bonn & Swales, 2007), Persian (Ghasempour & Farnia, 2017) and Spanish (Martín, 2003), largely revealing a tendency among users of EAL to simplify the abstracts by reducing the number of moves and steps as opposed to their Anglophone counterparts. Conversely, the studies that have taken a diachronic approach are limited. Vinkers et al. (2015) examined the frequency of lexical items with a positive connotation (e.g. ‘robust’, ‘novel’, ‘innovative’, ‘unprecedented’) in a corpus of PubMed abstracts between 1974 and 2014. More recently, Méndez Alcaraz and Alcaraz Ariza (2020) analysed the RA abstracts published from 1943 to 2018 in an astrophysics journal, revealing that abstracts in this discipline have become longer, more informative and more precise, with an overall increase in the number of authors, active and modal verbs, self-mentions and compound groups. However, to our knowledge, no previous diachronic study on RA abstracts has taken a genre-based approach to the analysis of this type of text.

This study explores the RA abstract in the specific field of experimental medical biology and how the rhetorical practices of this disciplinary community have evolved over time. To this end, we have analysed the moves and steps of 180 RA abstracts written in English and published in the Journal of Experimental Medicine over the 1940–2022 time span. We hope that this study contributes to shed some more light on the preferred rhetorical practices of a particular discourse community with an underlying pedagogical purpose: to facilitate novice writers’ and EAL scholars’ acceptance in their international disciplinary communities.

Methods

The Journal of Experimental Medicine (JEM)

The JEM is a high-impact open-access journal with a long trajectory published by the Rockefeller University Press. Since its inception in 1986, it has been regularly publishing studies in experimental medical biology integrating disciplines within the field of pathogenesis. According to the Journal Citation Reports, the journal received in 2021 an impact factor of 17.579, ranking it fourth out of 133 journals in the category “Medicine, Research & Experimental”. An online archive of articles back to 1896 is available in text and PDF formats. Material over 6 months old is freely accessible, and access to all papers is also provided free of charge to developing countries.

In the early years, most articles in JEM were produced with an IMR(D)C content structure although the division of sections, on many occasions, was not made explicit by means of headings, and the articles did not include an abstract. In the 1920s and 1930s, a common practice was to include a final section labelled as ‘Summary’ but with the function of a conclusion which basically involved a summary of main findings and which cannot be considered a self-standing genre since it was dependent on the preceding information. It was only from the 1940s onwards when an explicit IMRD pattern was clearly signalled by headings, and the inclusion of a final summary section, with the function of an abstract, became an established practice. It is worth mentioning that in the early 1990s, this summary section became more prominent by appearing as the first element after the title, and it was in the early 2000s when the heading ‘Summary’ changed to be labelled as ‘Abstract’, including the use of keywords. We should also point out that in the current 2020s decade, with the purpose of reaching a broad readership, it is frequent to find that many papers include, together with the abstract, three new promotional genres: the summary (a 40-word statement for the online JEM table of contents and alerts, intended for a more general readership), the graphical abstract (a concise and self-explanatory visual summary of the main findings that should capture the content of the article at a single glance, available online and used in social media promotions) and the short video summary (a no longer than 3-min video intended for non-specialist scientific audience and shared on social media).

Data collection and procedures

The corpus consists of 180 RA abstracts randomly selected from JEM over the period 1940–2022. In order to have a representative selection of the nine decades, 20 texts were drawn from each decade. The decision to compile the corpus from a single journal was to avoid possible rhetorical variation across the papers published in journals of the same subdiscipline due to different policies of editorial boards of specific journals, such as explicit indications for structuring the submitted abstracts in the guidelines for authors or other factors such as the broad/narrow scope of the journals, the specialised/generalised nature of the journal’s readership and/or certain specificities of long-established versus emergent journals, as reported in previous research (see, Martín & León Pérez, 2017).

Regarding the procedure that we followed for the analysis of the sample texts, in the initial stage of the study, we examined the corpus of abstracts in terms of the possible rhetorical strategies that writers have at their disposal to fulfil specific communicative functions. We have adopted a genre-analytical approach following the work by Swales (1990, 2004), in which he defines the concept of move as a “rhetorical unit that performs a coherent communicative function in a written or spoken discourse” (Swales, 2004, p. 228). In his analytical framework, the rhetorical moves manifest themselves as text units that occur in typical sequences, and these can be realised by means of lower-level constituents (sub-moves or steps) which, depending on the frequency of occurrence, can be considered obligatory or optional. Our preliminary move/step analysis revealed the scheme which is displayed in Fig. 1.

Fig. 1
figure 1

Framework proposal for the analysis of RA abstracts in medical biology

Full size image

In the second stage of the study, we undertook a quantitative genre analysis of the types and frequency of occurrence of moves and steps in the corpus analysed applying the framework in Fig. 1, mainly focusing on those moves/steps that convey a promotional function, i.e. M1-S1 and M5-S3. Each co-analyst (the two authors of this paper) segmented independently the texts and annotated manually the specific discourse functions at the level of move and step according to the rhetorical context in which they appeared. In order to validate the findings, we then contrasted the results obtained. Since the moves and steps are functional units, in this study, the criteria for move boundary identification and the assignment of discourse values to the various steps have been semantic rather than grammatical (i.e. sentences or clauses) although, in the cases of ambiguity, linguistic realisations were particularly useful as a means of confirming the analyses. In the instances in which two or more communicative functions occurred in the same meaningful fragment, each function has been classified as an independent move or step. A high level of inter-coder agreement on move/step structure was initially achieved (97% of inter-rater reliability for moves and 91% for steps), and complete agreement was reached after discussion.

Results and discussion

The results of the quantitative diachronic analysis are displayed in Table 1. As seen in this table, there has been a marked shift in the number of moves/steps from the decades preceding the 1990s until the present decade, making the abstract more rhetorically complex. The basic pattern M3 (methods) + M4 (results) + M5-S1 (interpretation of results) of the early decades is included in the 1990s M1-S2 (background information) and M2 (purpose), arguably with the aim of reaching a broader audience. We should also note that in the early 1990s, the abstract acquires more relevance by appearing in a more prominent position, after the title, before the introduction of the research articles. Table 1 also shows that, over the last three decades, this pattern has evolved to a more complex one that prevails today in which the practice of using three additional promotional steps has become frequent: M1-S1 (claiming importance), M1-S3 (gap creation) and M5-S3 (significance of findings).

Table 1 Distribution and frequency of occurrence of moves/steps in the corpus analysed
Full size table

A more fine-grained analysis of the moves and steps revealed that M1 (introducing the study within its field), in which writers situate their work in their specific research area, is achieved by means of three steps. In S1 (claiming importance of the research topic), writers highlight the interest in the topic of their study with the main purpose of attracting readers’ attention. As illustrated in examples 1 and 2, typical linguistic exponents are the use of positive adjectives (e.g. interesting, crucial, critical, major, central, important, pivotal, key, essential) that contribute to fulfil this promotional function.

  1. (1)

    The identity of allogenic peptide/major histocompatibility complex (MHC) complexes that elicit vigorous T cell responses has remained an interesting problem for both practical and theoretical reasons. (1995a)

  2. (2)

    White adipose tissues (WAT) play crucial roles in maintaining whole-body energy homeostasis, and their dysfunction can contribute to hepatic insulin resistance and type 2 diabetes mellitus. (2021a)

As shown in Table 1, it is only from the 1990s onwards when researchers started to use this step in the introductions of their abstracts. In this decade, we found that 25% of the texts included this communicative function, and over the last decades, it has become a prevalent step since in more than half of the abstracts, the writers made use of this promotional strategy. These findings are consistent with the results obtained by Wang and Yang (2015) who found a frequent occurrence of claims that boost the significance of the topic or the research area in applied linguistics RA introductions. Likewise, Fraser and Martin (2009), in their study of biased language which attributes significance to the claims, found an increasing incidence of adjectives expressing subjective judgments in high-impact biomedical and clinical journals over a 20-year time period (1985, 1995 and 2005). These authors argue that, as editorial acceptance of articles is primarily based on the potential interest to the readership, writers tend to exaggerate the importance of the topic to increase the chances of getting published in high-ranked journals.

Within M1, the most frequent step that authors use to introduce their research topic is S2 (providing background information). This can be achieved by reporting what is known about their research topic without citations (example 3) or reinforced by making reference to the findings in previous work (example 4), which also demonstrates the authors’ credentials as qualified scholars who are acquainted with the relevant literature.

  1. (3)

    The in vivo administration of antibodies specific for gene products of the I-A subregion represents an immunologically specific approach to the manipulation of Ly-1+ T cell responses to antigen. (1982a)

  2. (4)

    A number of recent studies have demonstrated that cellular responses to tumor necrosis factor (TNF) mediated by the p55 and the p75 TNF receptors are distinct. (1994b)

This rhetorical practice of describing the context began to be used in the 1970s (5%) and in the 1980s (30%), probably in an effort to reach a broader audience, but it was from the 1990s onwards when it became a well-established communicative function.

S3 (indicating a gap in existing knowledge), within M1, has the function of justifying research. By means of this step, the authors point out the possible topics or areas that still need research in relation to previous work. This typically occurs in a sentence coalescing with a previous S1 (example 5) or S2 (example 6), although we found a few instances in which, after establishing the importance of the research topic, it occurs as an independent unit preceded by an adversative conjunction, such as ‘however’, which introduces the gap (example 7) or as the first sentence in the abstract (example 8).

  1. (5)

    The T cell costimulatory molecule CD28 is important for T cell survival, yet both the signalling pathways downstream of CD28 and the apoptotic pathways they antagonize remain poorly understood. (2002a)

  2. (6)

    Aberrant cytokine expression has been proposed as an underlying cause of psoriasis, although it is unclear which cytokines play critical roles. (2006a)

  3. (7)

    In addition to their well characterized role in allergic inflammation, recent data confirm that mast cells play a more extensive role in a variety of immune responses. However, their contribution to autoimmune and neurologic disease processes has not been investigated. (2000b)

  4. (8)

    The migration pathways for dendritic cells (DC) from the blood are not yet completely resolved. (1997b)

In Table 1, we can see that this practice of justifying one’s research was initiated in the 1990s (25% of the texts), and it has become a more prevalent function over the last three decades. This growing need to justify research can be explained by the competitiveness that exists in the world of academia in recent years due to the increasing number of scholars who seek to publish their research internationally. We should also recall that, apart from becoming more rhetorically complex, the abstracts in the 1990s began to appear in a more prominent position as the first element of the article after the title. This also shows how abstracts have evolved over time to become a more promotional genre.

In M2, authors announce the aim of their research by indicating the main purpose of their study or by describing its main features. This may occur as an independent unit (example 9) although, arguably, due to constraints of space, in most cases, this move merges in an initial single sentence with M3, in which the methodology is also described (example 10). It is worth mentioning that in the last decades, we found two instances in which M2 becomes more promotional by coalescing with M1-S3 (indicating a gap in existing knowledge) and with M4 (reporting key results), as illustrated in examples 11 and 12.

  1. (9)

    The present study sought to determine which subsets of thymocytes were infected in the SCID-hu mouse model and to evaluate HIV-related alterations in the thymic microenvironment. (1993b)

  2. (10)

    The auto-allergic lesion in guinea pigs inoculated with homologous testis plus the Freund adjuvant was investigated histologically. (1959a)

  3. (11)

    The present study was designed to investigate the previously unexplored role of GM-CSF in autoimmune-mediated demyelination. (2001b)

  4. (12)

    Here, we report that tissue-resident group 2 innate lymphoid cells (ILC2) accumulate in the choroid plexus of aged brains. (2020b)

Table 1 shows that M2 has progressively been used more frequently since the 1940s to the 1990s, in which it became an obligatory move (95% of the cases); however, in the last three decades, its frequency of use has decreased, as rhetorical space has been occupied by other more promotional functions, i.e. M5-S3.

M3 (describing the research methodology) has remained constant throughout the whole period analysed as an obligatory move, as seen in Table 1, by means of which writers briefly describe the most relevant details of materials, participants, analytical procedures, study design or instruments for data collection. Although this move tends to appear as an independent unit in the first decades (example 13), in the last decades, we reported a high level of incidence in which it merges in the same sentence with M4 (example 14) where the main findings are announced.

  1. (13)

    Rabbit antibodies were prepared against purified mouse macrophages, erythrocytes, and liver lysosomes. (1972b)

  2. (14)

    Using a myelin oligodendrocyte glycoprotein (MOG)-induced model of acute EAE, we show that mast cell–deficient W/Wv mice exhibit significantly reduced disease incidence, delayed disease onset, and decreased mean clinical scores when compared with their wild-type congenic littermates. (2000b)

M4 (reporting key results) is the other obligatory move which appears in most abstracts analysed since the 1940s until the present decade. It is by means of this communicative function that authors make new knowledge claims by reporting on the main findings obtained in the study. It was interesting to find a clear-cut turning point from the prevalent practice of reporting results objectively in the early decades to the frequent practice, since the 1990s onwards, of highlighting findings by means of boosters (examples 15, 16 and 17), that is, pragmatic devices that encode subjective attitude with the clear function of strengthening the certainty of claims (Hyland, 2002).

  1. (15)

    Importantly, the predicted conformational characteristics of this disruption differs for class I and II molecules. (1990b)

  2. (16)

    Interestingly, localization of T blasts in the thymus was 10-fold higher in irradiated hosts than normal hosts. (1991b)

  3. (17)

    Remarkably, many of these ECM proteins are specifically increased in human CRC versus healthy colon. (2016a)

In the last M5 (discussing main findings), authors make their final claims about the importance of their research mainly by means of S1 (interpreting or explaining results), as illustrated in example 18. This communicative function, as shown in Table 1, was found in more than half of all the texts analysed throughout the nine decades. We could also appreciate the frequent practice, in the early decades, of using passive or impersonal constructions (example 19) and hedging devices (example 20) when interpreting results, that is, cautious language which downplays the degree of certainty or strength of the assertions in order to avoid critical audience responses. Conversely, in the last decades, it has become a more frequent practice to discuss the results in a more assertive manner, including the use of boosters (example 21) and hyperbolic language (example 22) which explicitly enhances the novel contribution of the study.

  1. (18)

    These findings explain respiratory failure in infants with congenital pulmonary lymphangiectasia, and suggest that inadequate late gestation lymphatic function may also contribute to respiratory failure in premature infants. (2014b)

  2. (19)

    These data indicate that SF antigen is a "differentiation antigen" restricted to certain cells of mesenchymal origin and character (1975b)

  3. (20)

    These effects of LPS may account for the triggering or the exacerbation of ante-DNA antibodies during infections with gram-negative bacteria. (1974b)

  4. (21)

    These results strongly suggest that HSCs and EMPs are generated from distinct cohorts of hemogenic endothelium. (2019a)

  5. (22)

    These findings demonstrate a novel mechanism by which IFN-I regulate immunological memory (2020b)

Within M5, another step which has been used throughout the nine decades, although with a lower degree of incidence, is S2 (drawing conclusions from results). This communicative function represents a final concluding or summarising statement as derived from the previously reported results. It is frequently signposted with discourse markers such as “We (therefore) conclude…” (example 23) or it appears as a final sentence in which the authors make a hypothesis about the results obtained (example 24). As seen in Table 1, the highest peak of incidence occurs in the 1990s (45% of the cases); however, in the last three decades, this step has experienced a decline in the frequency of use, and it seems that the rhetorical space of this function in the abstract is being replaced by a more promotional step, i.e. M5-S3.

  1. (23)

    We conclude that selective p55 TNF receptor activation is associated with early hemodynamic changes and the autocrine release of endogenous TNF alpha. Significant systemic toxicity results from p55 TNF receptor activation, but the role of the p75 TNF receptor in systemic TNF toxicity requires further study. (1994b)

  2. (24)

    Since no exonuclease activity could be demonstrated in plasma in vitro, it was postulated that breakdown of ssDNA by exonucleases occurs on the surface of hepatocytes of Kupffer cells. (1978b)

In the final M5-S3 (stating the significance or contribution of the research findings), writers highlight the novelty/uniqueness of their results (examples 25 and 26) and the implications of their research for the disciplinary area (examples 27 and 28). Due to the current competitive context of research production, it was not striking to find that this practice has increased progressively over the last few decades. As seen in Table 1, this step was inexistent in the early decades, and we only reported one instance in the last year of the 1980s (example 29) and three instances in the 1990s. However, in the 2010s and the early years of the 2020s respectively, in more than half (55%) and 70% of the texts examined, the authors chose to highlight the significance of their research as the last step to finalise the abstracts, indicating that this communicative function has become an obligatory step in the subdiscipline analysed.

  1. (25)

    These results illustrate a novel strategy for identifying T cell-stimulating antigens in general and directly show that alloreactive T cells can respond to rather rare peptide/MHC complexes. (1995a)

  2. (26)

    These data reveal a hitherto unappreciated role for an IL-10–STAT3–Blimp-1 circuit as an initiator of an inflammatory Th2 response in the lung to allergens.(2020a)

  3. (27)

    These findings further demonstrate a novel role for IL-15 (distinct from any of IL-2) in regulating microvascular endothelial cell adhesive function, help to understand the role of IL-15R expression on endothelium, and further support a central position for this cytokine in orchestrating multiple sequential aspects of T cell effector function and therefore chronic inflammatory processes. (1999a)

  4. (28)

    These results demonstrate a novel mechanism by which IFN-I regulate immunological memory and provide insights for rational vaccine design. (2020e)

  5. (29)

    While these data are the first to describe an animal model with both the humoral and cellular characteristics of PBC, they also raise an interesting question regarding the preferential localization of lymphoid cells to the biliary system. (1989a)

As seen in Fig. 2, the data analysed in this study show an increasing tendency over the last three decades to use persuasive communicative functions (M1-S1, M1-S3 and M5-S3) with the aim of promoting one’s research.

Fig. 2
figure 2

Frequency of occurrence of promotional moves/steps across time

Full size image

This current practice has been reported in previous cross-disciplinary studies. Martín and León Pérez (2014) found that authors in both health sciences and political sciences frequently enhance the novel contribution of their work. Vinkers et al. (2015) also reported an increase in the number of positive words such as ‘novel’ and ‘innovative’ in the titles and abstracts of medical research published from 1974 to 2014. This growing trend has been referred to as ‘hype’ which Millar et al. (2019) define as “hyperbolic and/or subjective language to glamorize, promote and/or exaggerate aspects of research” (p. 140). More recently, Hyland and Jiang (2021) explored the use of hypes in research papers in four disciplines (applied linguistics, sociology, electrical engineering and biology) over the last 50 years, and they also found a massive increase in these items over the last years, particularly in the hard sciences. This shift in rhetorical practices thus shows how authors have changed their behaviours in response to the current pressure that they face to publish in high-impact journals and which contrasts with the more neutral position towards research publication prevalent in the decades preceding the 1990s.

Conclusion

This study has contributed to show that academic genres are not static, but they are dynamic entities that evolve over time in response to changes in particular communicative needs. Our findings have revealed that, in the subdiscipline of experimental medical biology, the simple prevalent pattern of the abstracts in the early decades (M3 + M4 + M5-S1) has become more complex in the last decades by including promotional rhetorical moves and steps (M1-S1, M1-S2, M1-S3, M2, M5-S3) which seek to reach a wider audience and highlight the relevance and the implications of research. At this point, we should acknowledge the limitations of our study as this could have been complemented with an ethnographic approach which investigates the intrinsic and extrinsic motivations of authors for using promotional communicative functions and the effects that this produces on readers.

As regards the possible reasons for this growing trend, we have to point out that the JEM submission guidelines of the most recent years state that abstracts should describe “the relevant background, key results, and conceptual significance of findings in a way that is accessible to a broad audience”. This journal policy therefore seems to be one of the key factors that explain the increasing number of persuasive communicative functions found in the sample analysed over the last three decades and which reflects the prevalent values of the current disciplinary community. The fierce competition for publishing in highly ranked journals is another crucial factor which drives the increasing use of rhetorical promotion (e.g. Flowerdew, 2022; Hyland & Jiang, 2021, 2023). As shown in this study, it has become an essential practice for scholars of medical biology to foreground the relevance of their research, and it seems that this publication practice will continue growing over the next years. Some authors, however, have expressed their concern about the increasing use of hyperbole as this undermines what traditionally has been established as objective scientific narration. Millar et al. (2019, 2020), for example, point out the existence of evidence that suggests that hype may bias readers’ evaluation of new knowledge and their interpretation of true scientific facts. These authors thus suggest a critical pedagogical approach to research writing. Fraser and Martin (2009) further question whether language that exaggerates the importance of findings could cause a counterproductive effect as it may “fuel scepticism and alienate the reader” (p. 5). They call journal editors to encourage authors to make weaker claims as a way to return to a commitment to the standard of objectivity.

Vinkers et al. (2015) also criticise the growing research marketability, as it appears that the scholar who is best able to ‘sell’ their research might be the most successful, and claim for a new academic culture that rewards quality over quantity and fosters objectivity.

We should also note that the current JEM submission guidelines encourage authors of accepted papers to include together with the manuscript and the abstract some of the three mentioned above new genres (the summary, the graphical abstract and the short video summary). These recent genres clearly have a main promotional purpose as they contribute to enhance further the research and to reach a wider audience. It would be worth exploring in further studies the rhetorical strategies used by scholars in RA abstracts from different (sub)disciplines and in these new genres and how these rhetorical practices are evolving over the next few years because of the increasing need to promote one’s research.

Previous intercultural studies on this genre (Martín, 2003; Çandarlh, 2012; Ghasempour & Farnia, 2017) have revealed that RA abstracts written in languages other than English (e.g. Spanish, Turkish, Persian) are less promotional inasmuch as they include fewer moves and steps that have a persuasive function, such as claiming importance of the research topic, indicating a gap in existing knowledge or pointing the implications of the results obtained. It would therefore be interesting to explore in future intercultural research if the growing practice of stating the significance or contribution of research is also less frequent in other languages than in English. Considering that scholars, when seeking to publish their research, have to meet the expectations of gatekeepers (journal editors and peer reviewers), who will accept their work if they prove that they are familiar with the rhetorical conventions that prevail in their specific discourse communities, this study can have pedagogical implications for ERPP practitioners, early career researchers and for those users of EAL who aim at accommodating to the rhetorical expectations of their national and international communities.