Generic Overlap Between Publication Genres: The Case of Biology Research Articles’ and Research Letters’ Introductions in the Journal Nature

The present paper explores aspects of similarity and difference between the generic structure of research letters’ abstracts (henceforth RLsA) and research articles’ abstracts (henceforth RAsA). It aims at investigating and documenting the different rhetorical patterns of 19 RLsA and 19 RAsA in order to identify if there is any unique shared way to write them, determine the most publishable way of writing this genre, and detect any possibility of generic overlap between the two genres. Melliti (2016, 2017) CARL model has been adopted to identify the kind, frequency, and overlap of moves in RLsA and RAsA of the Journal Nature. The results indicate that although the RAs are longer than the RLs, the number of sentences in the RLsA is more than the RAsA. Results show also that there are fundamental as well as expendable sets of keys in both genres. The study succeeded also in identifying the number of sentences required to write a publishable research letter abstract and research article abstract in the field of biology. These findings have interesting implication on teaching academic writing and teaching English for publication purposes.

"functional components" as essential elements of genre. According to him, genres are "communicative events" that are characterized both by their communicative purposes and by a variety of patterns of structure, style, content and intended audience. Swales ' (1990, 2004) model of genre analysis, move structure analysis, classifies segments of texts according to their prototypical communicative purpose for a particular genre. In this respect, his model has been used as a framework in ESP research that focus on the analysis and teaching of the spoken and written language required of non-native speakers in academic and professional settings (Hyon, 1996).
RLs are small reports of pioneering research focused on extraordinary results. The most important advantage that RLs offer is the rapid review and publication. As briefly put by Rutkowski and Ehrenfest (2012), "the research letter is an autonomous format for the rapid publication of data (…) [as the] concept of this format is to publish quickly good preliminary data (or data obtained with a simple protocol) in a short format" (p. 102). Its autonomy means that there is a strong possibility for being different from other genres and formats in terms of generic organization, which is, eventually, the issue that this research article seeks to explore.
Throughout history, researchers have used several ways to report science. Among these methods, one could list books, magazines, newspapers, journals, editorials, research articles, reviews, letters to the editor, and the RL. In an attempt to investigate the generic layout of texts reporting scientific research, researchers used various frameworks, approaches, and tools.
This work makes use of genre analysis to explore the way RLsA and RAsA in biology are organized in terms of their common rhetorical structure. This means investigating and documenting the different rhetorical patterns of 19 RLsA and 19 RAsA in order to identify the unique shared way to write them, to determine the most publishable way of writing this genre, and detect any possibility of overlap between the two genres.
The notion of genre has been used to investigate several domains (Del Arbol, 2005) such as linguistic anthropology, folklore studies, conversational analysis, the ethnography of communication, the sociology of language, applied linguistics, and literary theory and rhetoric.
For Swales (1990): A genre comprises a class of communicative events, the members of which share some set of communicative purposes. These purposes are recognized by the expert members of the parent discourse community, and thereby constitute the rationale for the genre (p. 58).
This means that for a genre to be recognized it necessitates primarily encompassing a set of common features. These features create (or may be are created by) a particular group of community members and experts. Although genre remains a blurry concept, it could be defined as "the study of situated linguistic behavior in institutionalized academic or professional settings" (Swales, 1990, p. 629). This fuzziness may be the result of the variability in the ways of addressing genre. For Berkenkotter and Huckin (1995), genre is considered typification of rhetorical action. However, for Martin, Christy and Rothery (1987) and Martin (1993) it is regularities of staged goal oriented social processes. Besides, for Swales (1990) and Bhatia (1993), genre is looked at in terms of consistency of communicative purposes.
In spite of these differences, these orientations share three main common features, which are "emphasis on conventional knowledge, (…) versatility of generic description, (…) and propensity for innovation" (Swales, 1990, p. 630).
The present work attempts to explore the rhetorical conventional rules governing the construction of RLsA being an emergent genre (Maci, 2009;Gotti, 2007) in scientific writing. It aims also at comparing the generic structure of RLsA to that of RAsA. In an attempt to structure the RL genre, some journals suggest different forms of constructing letters. Such difference is what necessitates the unification of the rhetorical structures of this genre. The ultimate aim is to make it more comprehensible to readers and easy to write for scientists all around the globe, and especially non-Anglophone scientists who are striving to be published authors. An example of the RL's layout that journals provide is the one suggested by Nature in the following diagram. Chinese and Anglo-American scientists concerning the science paper introductions.
In her analysis of Lte, Del Arbol (2005, p. 156) found that they are composed of 3 moves and 11 steps. Comparing editorials to Lte, Del Arbol (ibid) found that the structure of the latter is more rigid than the former and she returned this to the fact that "each genre may have unique linguistic patterns which are not shared with the rest" (p. 157). However, the study conducted by Del Arbol (2005) could be considered limited for it focused only on 25 Lte. Swales (1990) developed a model for research articles introductions based on studying numerous corpora. He found that academic articles are basically composed of three moves: establishing a territory, establishing a niche, and occupying the niche. Based on this model, various research studies have been conducted in order to assess CR claims that written discourse structure differs in relation to the linguistic background of the writer (Taylor & Chen, 1991).

C Conclusions
A modified version of this model has been used by Melliti (2016) to identify the generic structure of 37 RLs dealing with different disciplines. Suggesting what he called Create A Research Letter Introduction Model (CARL Introduction Model), he found that RLs' abstracts are composed 12 sentences of which 7 are obligatory and 5 are optional. The model has also been used by Melliti (2017) to identify what he called Create A Research Letter Body Model (CARL Body Model). He found that the body section of RLs is composed of 58 sentences where 49 are obligatory and 9 are optional. These findings are important as they assist researchers in scientific disciplines in their attempt to write publishable research letters and help EAP teachers teach writing in a structured and standardized way.
It is important to investigate the rhetorical structure of different genres and sub-genre such as RL as this helps scientists become aware of the patterns of this mode of scientific contribution in knowledge production. It is crucial also for it assists ESP teachers in identifying the organizational structure of RLs and teach them effectively to learners as according to Henry and Roseberry (2001) "ESP practitioners need to be aware of not just this range of features, but where they are used and for what purpose" (p. 167).

Methods
The model of Swales (2004) has been adopted to be compared with the model of research letters identified by the researcher. In order to identify the generic structure potential, Swales (2004) suggested the following Moves and Steps Create a Research Space (CARS) model. Establishing a territory

Move 2 Establishing a niche
Step 1A: Indicating a gap or Step 1B: Adding to what is known ijel.ccsenet.org International Journal of English Linguistics Vol. 11, No. 1;2021 Step 2: Presenting positive justification (optional)

Move 3 Presenting the present work
Step 1: Announcing present research descriptively AND/OR purposively (obligatory) Step 2: Presenting research questions or hypotheses* (optional) Step 3: Definitional clarifications* (optional) Step 4: Summarizing methods* (optional) Step 5: Announcing principle outcomes (optional)** Step 6: Stating the value of the present research (optional)** Step 7: Outlining the structure of the paper (optional)** The researcher resorted to Melliti (2016Melliti ( , 2017 CARL model in order to extract the rhetorical patterns of 19 Biology RLsA and 19 Biology RAsA randomly downloaded from Nature journal.
RLsA and RAsA were selected and analyzed sentence by sentence. Each sentence was allocated a particular structural element or key according to its function. The occurrence and frequency of these functions in the introduction of each RL were counted in order to identify the shared rhetorical patterns between the 19 RLsA and 19 RAsA randomly selected from the journal Nature. The aim is to contribute to the effort of identifying the hidden structure of this new and under researched genre.

Types and Distribution of Keys in RAsA and RLsA
The The above-mentioned keys are the different moves constructing the abstracts of RLs. They are resorted to by Biology researchers in their 'abstracting' process to give the readers an idea about the paper, its context, results, and implications. Needless to mention that the frequency of the moves is not the same in the structure of RLsA. Some moves are mentioned more frequently than others. For example, the BI move is mentioned 20 times in the 19 Biology RLs while FR is mentioned only once. This inequality in moves distribution in RLsA shows that some moves are more valued than others in the generic structure of this genre. It could be explained by the fact that the researcher does not necessarily need to expect future studies with regard to the issue tackled in the RL at the abstract phase. It is so as the aim of the abstract is primarily to contextualize the study and expose its results.
On the other hand, the results indicate that the RAsA are composed of only 8 kinds of moves: Background Information (BI)

Identification of Gap (IG)
Background Information/Identification of Gap (BI/IG) Study-related (22), and Background Information-related (20) sentences. Together, they form 137 out of the 169 mentioned sentences in the RLsA analyzed. This set could be then the main part in the structure of the RLsA in biology as it represents 81% of the total sentences, as illustrated in the table below. In particular, the findings of the present study show that the RAsA is composed of 6.84 sentences distributed as follows: The table above identifies the number of keys that are mentioned in the analyzed biology RAsA. It clarifies the dominance of results and conclusion sentences.

Suggested Number of Sentences
In order to suggest a reasonable number of sentences that can be used by researchers when writing RAsA in biology, there is a need to provide decimal numbers. In order to suggest the Suggested Number of Sentences (SNS) the researcher considered all numbers that are equal or above 0.4 as 1 and all numbers under it as 0. This aims at suggesting a logical number of sentences by the end as it is not possible to write 0.5 or 0.3 of a sentence. Applying this rule, the researcher ended up having the below distribution of sentences: The table shows that the publishable RAsA needs to contain five types of keys forming seven sentences: • One Background Information sentence • One Identification of Gap sentence • One Purpose of Study sentence • Three Results sentences, and • One Conclusion sentence. This finding means that in order to write a publishable biology research article, researchers need first to write one background information sentence in which they situate and contextualize their study. Second, they need to write one identification of gap sentence in which they mention the gap they intend to fill in research in their study. Third, they need to write one purpose of study sentence where they clearly announce the aim behind their study. Fourth, the need to write three main results they reached in their study before finally drawing one conclusion out of them.
In their turn, each RL abstract is shown to be composed of 8.89 different move. The following table shows the distribution of the moves per research letter.  Vol. 11, No. 1;2021 The table above shows the distribution of keys across every RL analyzed. It clearly reveals the dominance of the results section. Applying the same methodology and in order to suggest a reasonable number of sentences that can be used by researchers when writing RLsA in biology, there is a need to provide decimal numbers. In order to suggest the Suggested Number of Sentences (SNS) the researcher considered all numbers that are equal or above 0.4 as 1 and all numbers under it as 0. Applying this rule, the researcher ended up having the below distribution of sentences: Seven types of sentences are found to be making up the biology RL genre. These types are the following: • Background Information • Identification of Gap • Purpose of Study

• Previous Research
• Previous Research melted with Background Information • Results • Conclusion It seems that at the abstract level, the RL is richer than the RA as it outnumbers it in terms of type and quantity of sentences. This could be attributed to the need to summarize the most appealing points in the RL at the abstract level to get more attention. It seems that this genre is securing its niche in publications sphere by being practical and up to the point through focusing on what researchers expect from publications in a world characterized by speed. Giving readers what they need in a direct and thorough way is what is probably, interalia, raising attention to this genre.
The table above identifies the required kind and number of sentences to write a publishable RL. It shows that researchers need to start with a Background Information sentence followed by a Previous Research sentence followed by one Previous Research melted with Background Information sentence. Researchers need to resort then to writing an Identification of Gap sentence followed by a Purpose of Study one. After that, they need to write a, then Results sentences, before ending with one Conclusion sentence. This confirms the findings of Melliti (2017) that RLs abstracts are composed of three main phases • The Introductory Phase (IP) • The Contextualization Phase (CP) • The Findings Phase (FP) In the present study, and based of the phases identified by Melliti (2017), three types of sentences fall into the IP, which are Background Information, Previous Research, and Previous Research melted with Background Information. The CP phase is exemplified by Identification of Gap and Purpose of Study moves while the FP is represented by the Results and the Conclusion keys.
Needless to mention that the suggested number of sentences is identified through resorting to the data collected, but the order of these sentences is based on the researcher knowledge of the genre as well as the logical layout of research papers in general.
It could be noted that there are two kinds of sentences that are mentioned in the RL genre but not the RA one are:

• Previous Research
• Previous Research melted with Background Information Regarding the PR and PR/BI case, their absence in the RAsA could be interpreted by the fact that this genre reserves a whole section in the body to review previous literature and provide background information about the topic in an extensive way compared to the RL genre (Melliti, 2017). The presence of these moves in RLsA could be attributed to the fact that it is hardly dealt with in the body sections (Melliti, 2017).
The present study shows that there are differences between RLsA and RAsA in terms of generic structure. ijel.ccsenet.org International Journal of English Linguistics Vol. 11, No. 1;2021 However, it shows also that there is overlap as they both invest a similar set of keys or functions. These differences and overlaps make it imperative to teach English for publication purposed to graduate and postgraduate students. It would equip them with the necessary tools to write publishable research letters and research articles not only at the abstract level but also the other sections. Developing a course that exposes the internal structure of these genres and gives them practice on how to write them can assist Anglophone and non-Anglophone researchers in their endeavor to publish their research studies (Melliti, 2019).

Conclusion
The present study aimed at identifying the generic structure of research letters and research articles' abstracts through using Swales (2004) CARS model and Melliti (2017) CARL model. The study found that RLsA are richer than RAsA in terms of keys. It managed also to identify the exact number and kind of sentences required to write both genres and have ones' research paper published. Research in this area of investigation has several pedagogical and research related implications that are mainly related to genre teaching. Further research may need to focus on other disciplines, other than biology, in order to uncover their hidden generic structure and compare it to the present study.