English and Arabic abstracts in medical research articles: A contrastive study

Abstract Abstracts in academic research serve important functions. An abstract is a summary of an entire article or thesis that defines its objectives, research problem, methodology, major findings, and implications. Abstracts help potential readers to decide whether the article is relevant to their work. Thus, it saves their time and effort. This study aims to answer the following questions: (1) To what extent are the move structures in English and Arabic medical abstracts similar or different, (2) How are metadiscourse features similar or different in English and Arabic medical abstracts? The sample of the study consisted of 120 abstracts (60 English and 60 Arabic) extracted from three major medical journals that require bilingual Arabic and English abstracts. The move structure and metadiscourse features in bilingual abstracts were analyzed using Hyland’s models of 2000 and 2005. The move stability was classified according to Kanoksilapatham’s (2005) model. Both Arabic and English abstracts showed similar move structure patterns, with the method and product moves being obligatory; the purpose and conclusion moves were conventional, and the introduction move was optional. English and Arabic abstracts employed interactive metadiscourse features more than interactional ones. The interactive metadiscourse features in Arabic displayed a higher frequency than those in English, but the interactional metadiscourse features had a lower frequency in the Arabic abstracts than those in English. Transitions were the most used form of interactive metadiscourse features, whereas self-mention and hedges were most common among the interactional metadiscourse features in both Arabic and English abstracts. Arabic and English medical abstracts showed a homogenous rhetorical move structure across both languages. However, their metadiscourse features showed a significant divergence. The findings of such a paper may have theoretical and practical implications. They may contribute to contrastive textology and genre studies since they highlight the similarities and differences between English and Arabic article abstracts. The findings can also be used in training researchers in writing research abstracts.

English and Arabic article abstracts.The findings can also be used in training researchers in writing research abstracts.

Introduction
The abstract summarizes the research article at its beginning.It was defined by the American National Standards Institute as "an abbreviated, accurate representation of the contents of a document, preferably prepared by its author(s) for publication with it" (Lorés, 2004, p. 281).It is considered the most crucial section of a research article because it is the most read part.It represents the doorway to disseminating knowledge and persuading the reader about the argument of the research article.Thus, the abstract is frequently the basis for the prospective reader's choice to read the full article or the journal.For conferences and seminars, it is notable that coordinators usually request only the abstract of the research article to be considered, which will be the sole base for their decision to accept or reject the article.Thus, the abstract should be a stand-alone section, as it can frequently be viewed and read independently of the research article.However, the abstract section is usually restricted by a specific word limit as per instructions to authors of most medical journals.Despite that restriction, the abstract is expected to have separate sections or statements executively covering the background of the research article topic, the study objective, the methodology used, the results, and the conclusion.This scheme or move structure is frequently laid down within the journal's instructions to authors.Based on this, the abstract constitutes an independent genre, allowing linguists to investigate it as a separate text unit for linguistic features.Bhatia (1994) proposed a move structure model for the abstract of research articles.The model adopted a four-move structure for the abstract, which included the study purpose, methodology, results, and conclusions.The model by Bhatia was further modified to the currently adopted Hyland's (2000) which consisted of a five-move structure including the introduction, the purpose, the methods, the product, and the conclusion (Behnam & Golpour, 2014).In Hyland's model, the introduction of an abstract establishes the context and motivates the research.The purpose is the study's goal, aim, or objective.The methods represent a brief account of the design, procedure, and analysis.The product is the study's results, whereas the conclusion provides a more expansive massage and a broader location or implications of the study (Behnam & Golpour, 2014).
On the other hand, the metadiscourse within the abstract can represent several rhetorical devices that are crucial for understanding the dynamic nature of the text.The metadiscourse was defined by Vande Kopple (1985) as "discourse that people use not to expand referential material but to help their readers connect, organize, interpret, evaluate, and develop attitudes toward that material."(p.83).Crismore et al. (1993) viewed metadiscourse as "non-propositional aspects of discourse which help to organize the prose as a coherent text and convey a writer's personality, credibility, reader sensitivity and relationship to the message" (Alejo González, 2005, p. 83).Hyland (2005) described a model for metadiscourse that is an extension of the work by Crismore et al. (1993) andVande Kopple (1985).This metadiscourse model consists of two major categories: the interactive and the interactional.The interactive categories include frame markers, transitions, evidentials, endophoric markers, and code glosses.They aim to provide a coherent text that guides the reader according to the writer's expectations.The interactional categories include boosters, hedges, self-mentions, attitude markers, and engagement markers.They aim to provide a lively text that connects to the reader, making an interpersonal connection with him, thus, conveying the reactions and context of the writer (Hyland, 2005).
Few studies were conducted to compare English and Arabic abstracts regarding move structure and metadiscourse features.Alharbi and Swales (2011) evaluated the move structure and person reference metadiscourse in Arabic versus English abstracts of research articles from the field of language sciences.Their results showed limited moves structure (background to findings) and very few instances of personal reference use in both Arabic and English abstracts.El-Dakhs (2020) used Hyland's model of move structure of research articles abstracts to examine bilingual abstracts of published research articles in the field of education.El-Dakhs (2020) noted that English abstracts followed the five moves cited by Hyland's model, whereas Arabic abstracts relied on a few moves, primary purpose, method, and findings.Bouziane and Metkal (2020) studied muti-lingual abstracts from a group of linguistics journals, including English, French, and Arabic.They noted that Arabic and French abstracts rely on a few structure moves (purpose, findings, product) as opposed to the English ones, which use the conventional five-move structure.Alotaibi (2015) used Hyland's model to examine the metadiscourse features in abstracts of research articles from the field of linguistics.He found an overuse of interactive metadiscourse markers compared to the interactional ones in both English and Arabic abstracts.Except for self-reference, English abstracts used more interactional markers than their Arabic counterparts.Remarkably, according to our literature search, no study has examined English-Arabic abstracts from medical sciences journals for moves structure and metadiscourse features.Given the distinct features of medical discourse (Abdollahpour & Gholami, 2018;Džuganová, 2019;Fryer, 2012), there is an apparent gap in research to warrant to the study of move structure and metadiscourse features in English-Arabic abstracts from medical sciences journals.Based on that, the research questions of this study were formulated as follows: (1) How are the move structures in English and Arabic medical abstracts similar or different?
(2) What are the differences and similarities between using the metadiscourse features in English and Arabic medical abstracts?

Literature review
The topic of this paper lies in the realm of research in contrastive rhetoric and applied linguistics as stated by Connor (1996, p. 5): Broadly considered, contrastive rhetoric examines differences and similarities in writing across cultures.The underlying premise of the field is that any given language is likely to have written texts that are constructed using identifiable discourse features, and these features may differ across Languages".(In Kroll, 2012:2018) Studies on move structure and metadiscourse features in medical abstracts are limited.Salager-Meyer (1992) analyzed move structure concerning verb tenses in abstracts written in medical journals.This researcher found that the simple past tense was associated with purpose, methods, and results moves.In contrast, the simple present tense was mainly used for the conclusion and recommendation moves.
Zhao and Tao (2013) also compared medical abstracts (n = 100) across English and Chinese for move structure.They found that Move 1 (research background) was nearly absent in the abstracts written by Chinese writers as opposed to those in English.Suntara et al. (2010) studied the interactional metadiscourse features in research article abstracts in the public health discipline.Using Hyland's (2005) model, they analyzed 60 English abstracts and found frequent use of attitude markers, self-mentions, hedges, and boosters.In contrast, the use of transition markers and consequential or contrastive connections was minimal.Abdollahpour and Gholami (2018) analyzed move structure in 1500 English medical abstracts extracted from high-impact factor medical journals.They found that English medical abstracts used a fixed pattern of five-move structure, with the introduction, aims or objectives, and conclusion considered conventional.In contrast, moves describing the methodology and the product were essentially obligatory.
Muresan & Kic-Drgas, in a couple of studies (Muresan & Kic-Drgas, 2018, 2019) compared the macrostructure and verb tenses in 40 abstracts in medical and economics journals.They found that 75% of abstracts in medical journals were structured, whereas only 10% in economics journals were structured.In economics abstracts the present simple tense was preferred, whereas, in medical sciences abstracts, the past tense was consistently used to describe the purpose, methods, and results.Ashofteh et al. (2020) examined the functions and distribution of move structure of 125 abstracts in applied linguistics.Their findings showed that authors in this field tend to employ more hedges in their abstracts.They also offered some recommendations for novice authors in applied linguistics.
In a more recent study, Saidi and Karami (2021) investigated the frequency of interactional metadiscourse markers in 19 applied linguistics reply.Their findings revealed that self-mentions were the most frequent among other discourse markers followed by hedges, boosters, engagement markers and attitude markers.The findings are in harmony with similar studies that support the specific nature of different genres.However, the sample of their study could have been larger in order for the results to be generalizable.
The previous studies examined either move structure in English medical or business abstracts or compared them with abstracts written in Chinese.This study investigates English and Arabic medical abstracts.
As far as Arabic medical abstracts are concerned, few studies were conducted to compare English and Arabic abstracts regarding move structure and metadiscourse features.Alharbi and Swales (2011), for example, discussed Arabic and English abstracts in language science publications.They examined move structure and person reference metadiscourse features in 28 Arabic and English-paired abstracts from three journals.This study demonstrated limited move structure (background to findings) and few instances of personal reference use in both Arabic and English abstracts.
Furthermore, Alotaibi (2015) used Hyland's (2005) approach to examine the metadiscourse features in 44 Arabic and English linguistics abstracts.The findings of his study revealed that interactive metadiscourse features were employed more than interactional ones in both sets of abstracts.Except for self-reference, English abstracts used more interactional markers than their Arabic counterparts.Alzarieni et al. (2019) examined interactional metadiscourse markers according to Hyland's (2005) approach in Arabic abstracts of 60 patents written by Arabic-native innovators.According to this study, attitude, hedge, and booster markers were the most frequently used, whereas the other features showed a low frequency of use.The evaluation also indicates that interactional metadiscourse features in Arabic abstracts of patents served various functions, including persuasion, avoiding commitment to precise numbers, and delivering data honestly.
El-Dakhs (2020) compared interactional metadiscourse and move structure between Arabic and English abstracts of research papers.To this goal, Hyland's (2005) interaction model and Hyland's (2000) five-move model were adopted to examine abstracts (n = 400) from famous publications in the education field.The abstracts were evenly split between Arabic and English.Results showed that English abstracts followed the conventional move structure per Hyland's 5-move structure.However, Arabic counterparts focus on Purpose, Method, and Findings move structures.This study revealed that Arabic abstracts had more extended Methods and Findings, whereas English abstracts had a longer Introductions and Conclusions.Regarding interactional metadiscourse features, Arabic had more boosters, whereas English utilized more hedges and self-mentions.Interestingly, engagement markers were rarely used in abstracts of either language.
Bouziane and Metkal (2020) conducted another study based on Hyland's (2000).This study examined English, Arabic, and French for adherence to the similar move structure within or between languages.Abstracts (n = 112) in applied linguistics were compared.The findings revealed that Arabic and French abstracts relied on a few structure moves (purpose, findings, product) as opposed to the English ones, which use the conventional five-move structure more frequently.It was also found that English abstracts written by native and non-native English speakers and Arabic abstracts written by Middle Eastern authors adhered to move structure conventions in English.Those from North Africa, whether written in Arabic or French, did not exhibit any distinctive patterns that might have been associated with the language used to write them.Furthermore, Prasetyanti et al. (2023) conducted a study to identify both the interactive and interactional metadiscourse markers in 100 introductions of English dissertations written by native speakers and Indonesians.Their findings revealed that the interactive metadiscourse features including transitions, frame markers, and evidentials were more frequent in linguistics than those in the education field.It was also found that there was no significant difference between the uses of attitude markers in both disciplines.This study examined the metadiscourse features in dissertation introductions of various disciplines.However, the linguistics authors tended to use interactive markers more than those in education, but education writers used interactional discourse markers more.
It has become clear that Arabic and English medical abstracts have not been accorded adequate attention in research.Up to our knowledge, no study has examined English-Arabic abstracts from medical sciences journals in terms of move structure and metadiscourse features.Therefore, the present study attempts to fill in a gap in this field.

Conceptual framework
It might be expedient at this point to introduce the basic terms used in this study since they are drastically needed for a better understanding of the topic.These terms include metadiscourse, interactive metadiscourse features and interactional metadiscourse features.
Metadiscourse was defined by Hyland (2005, p. 37) as "the self-reflective expressions used to negotiate interactional meanings in a text, assisting the writer (or speaker) to express a viewpoint and engage with readers as members of a particular community."In a more recent publication, Hyland (2017, p. 16) defined metadiscourse as "the ways in which writers and speakers interact through their use of language with readers and listeners . . . the commentary on a text made by its producer in the course of speaking or writing."Metadiscourse markers can are either interactive or interactional.

Interactive metadiscourse
Interactive metadiscourse markers provide the organization of texts and enhance readers' comprehension, while interactional metadiscourse shows the writer's attitude toward readers to construct the text (Hyland, 2005).Examples of both types in Arabic and English can be seen in the results section below.

Methodology
Data consisted of 120 abstracts (English and Arabic) of research articles from three medical journals published in the Middle East, the Saudi Medical Journal, the Jordan Medical Journal, and the Jordan Journal of Pharmaceutical Sciences.The selected journals are top-rated among middle Eastern medical authors, and they have a regular publication track record over the past decade, with a relatively high volumes/issue frequency.They also include publications from authors from most Arab countries, and they are among the few journals that request a mandatory Arabic abstract to accompany the English research articles.Included in the study were abstracts published within the years 2021 and 2022.Excluded were abstracts of methodology articles as they have a distinct abstract structure, which is different in move structure from empirical medical research articles.
The identification of move structure was based on Hyland's (2000) five rhetorical move model.These moves are Introduction, Purpose, Method, Product, and Conclusion.Both the frequency and percentage of each move structure were quantified.Moves were then classified using the conventional cutoff values for move stability (Kanoksilapatham, 2005).In this model, if a move occurred in the abstracts 100% of the time, it is considered obligatory, and if it occurred < 100% and ≥ 60% of the time, it is considered conventional.The move is considered optional if it occurred < 60% of the time.
The identification of metadiscourse features followed Hyland's (2005).The classification of metadiscourse features according to this model is detailed in Table 1.The metadiscourse features were then analyzed quantitatively by calculating the frequency and percentage of each metadiscourse feature.Finally, the qualitative contrastive analysis of each metadiscourse feature was carried out between Arabic and English abstract pairs.

Results
The findings revealed that English and Arabic abstracts of research articles in medical sciences showed equal frequency and percentage in terms of being structured versus non-structured.For both the English abstracts and their Arabic translation, 44 (73.3%) were structured, whereas 16 (26.7%)were non-structured (Figure 1).The move structure, which was examined based on Hyland's (2000) model, was closely comparable between Arabic and English abstracts (Table 2).The method and the product move structures occurred in all Arabic and English abstracts.In contrast, the introduction move structure occurred 41.7% of the time in both the English abstracts and their Arabic counterparts.The move structure frequency was used to determine move stability  (Kanoksilapatham, 2005).According to this model, the introduction move was optional.In contrast, the method and the product move structures were obligatory, and the purpose and conclusion move structures were conventional in both the English and Arabic abstracts.
The number of move structure in both the English and Arabic abstracts was either five or four.The five-move structure pattern had the sequence of introduction-purpose-method-product-conclusion (I-P-M-R-C) 38.33% of the time in English abstracts and 35.00% of the time in Arabic abstracts.The four-move structure pattern had three patterns.The most common one is purpose-method-product -conclusion (P-M-R-C) that occurred 58.3% of the time in both the English and Arabic abstracts.Other patterns are shown in Table 3.The possible association between language (English versus Arabic) and the number of move structure in the abstract was examined using Pearson's Chi-square test.However, no association was detected (X2 (1, N = 60) = 0.14, p = 0.70, Table 4).
The move length analysis was carried out and results showed that both Arabic and English abstracts showed the product move structure occupies the longest stretch of abstracts, followed by the method and conclusion move structures.This indicates the crucial role of those move structures among writers in the medical field (Table 5).Both English and Arabic abstracts employed interactive features more than interactional ones for the metadiscourse features.Additionally, the interactive metadiscourse features had a markedly higher frequency in the Arabic abstracts (n = 603) compared to English abstracts (n = 854) of medical journal articles.In contrast, the interactional metadiscourse features had a lower frequency in the Arabic abstracts (n = 52) than English abstracts (n = 44, Table The metadiscourse features were analyzed according to each move structure.Each of the purpose, method, product, and conclusion move structures showed a higher frequency of interactive metadiscourse features in Arabic abstracts compared to their English counterparts Figure 2(A).On the other hand, the introduction move showed comparable interactive metadiscourse features in Arabic versus English abstracts Figure 2(A).in other parts of the text.Since the abstracts are meant to be standalone and are frequently presented in separation from the manuscript, it is natural that no endophoric markers are employed in the abstracts of articles from medical sciences.
For the interactional metadiscourse features, the encountered in English abstracts were self-mention followed by hedges.In Arabic abstracts, the order is reversed, where the most common are hedges followed by self-mention.Interestingly, the engagement, attitude, and boosters metadiscourse features were absent from abstracts and the least encountered in both English and Arabic abstracts (Table 6).
Like the case with other interactional metadiscourse features, the widespread use of self-mention was more in English than in Arabic medical abstracts.The conclusion move had the most frequent occurrences among other moves, followed by the method and product move structures.The conclusion move of English abstracts had about double the self-mention occurrences of the conclusion move of the Arabic abstracts.In contrast, the method move had the opposite trend, with Arabic abstracts having double the self-mention occurrences of those of the English abstracts (Table 8).Interestingly, all the self-mention occurrences were in the plural form.Most of the abstracts were written by single authors and very few were co-authored.It seems that there is a tendency among Arab writers to use the first person plural pronoun)"we " ( ‫ﻧ‬ ‫ﺤ‬ ‫ﻦ‬ to refer to themselves rather than using the first person singular pronoun.In contrast, the abstracts in both languages were devoid of the use of the singular or the objective forms of self-mention.The markers used in English were 'we' or 'our,' whereas, in Arabic, they  ,our analysis" or the plural verb form " ‫ﻧ‬ ‫ﺠ‬ ‫ﺪ‬ .We find" Concerning hedges, they were slightly more employed in English as opposed to Arabic abstracts (Table 6).The hedges were concentrated in the conclusion, followed by the product move structures.The other three moves (introduction, purpose, and method) had a single or a couple of hedge occurrences in the examined medical articles' abstracts (Table 8).The most employed Arabic hedge was " ‫ﻗ‬ ‫ﺪ‬ " (qad: (may/might), which does not have a direct meaning in English, but was suited to "may closely related to the Arabic ‫ﺭ‬ ‫ﺑ‬ ‫ﻤ‬ ‫ﺎ‬ "(might be) and a single occasion to "might" and "probably."The hedge "may" was the most frequently used hedge in the English abstracts.Other hedges that appeared in English abstracts were "possible," "approximately," "more likely," "about," and "probably."

Conclusions and discussion
Arabic and English medical abstracts showed a homogenous rhetorical move structure across both languages.However, their metadiscourse features showed significant divergence.These results could have considerable implications for the assumption made by professional translators and writers in the medical field, who cannot assume homogeneous metadiscourse features to exist between English and Arabic translated abstracts.
The present study showed that both Arabic and English abstracts had similar move structure patterns, with the method and product moves being obligatory, the purpose and conclusion move being conventional, and the introduction move being optional.This is in harmony with a previous study, which reported that English medical abstracts used a fixed pattern of a five-move structure, with methodology and product moves as obligatory, and moves to describe aims or objectives and conclusion as conventional (Abdollahpour & Gholami, 2018).
The interactive metadiscourse feature prevailed over the interactional ones in both English and Arabic medical abstracts.This agrees with previous studies on English versus Arabic abstracts from the language science field (Alotaibi, 2015;Salek, 2014;Sultan, 2011).
The cross-language examination of interactive metadiscourse features showed a markedly higher frequency in the Arabic abstracts than in the English ones in medical journals.This agrees with Alotaibi, who showed a similar trend in English and Arabic abstracts from language sciences (Alotaibi, 2015).Transitions were the most used form of interactive metadiscourse features in both languages.Alotaibi (2015) showed that transition markers were most prevalent in Arabic abstracts in the field of language sciences, whereas frame markers were most prevalent in English ones.On the other hand, Sultan (2011) stated that transitions should be employed more frequently than frame markers in discussion sections of both English and Arabic Most transitions used in both Arabic and English medical abstracts were of the additive type, namely, ‫:ﻭ"‬ and" "and."Kaplan (1966) has argued that the high use of additive transition markers to serve parallelism in paragraph structure, such as the case of the Arabic language, among other Semitic languages, could point to the author's intention to build facts rather than argumentation.This could be an embedded property in the medical abstracts as opposed to social and language sciences.Such an argument could explain the high use of additive transition markers in English and Arabic abstracts from medical sciences.
The present study showed that interactional metadiscourse features had a lower frequency in the Arabic abstracts than in the English abstracts.Self-mention and hedges were most common among the interactional metadiscourse features in Arabic and English medical abstracts.The present study showed that the most used hedge marker in Arabic was " ‫ﻗ‬ ‫ﺪ‬ " (qad: may).In contrast, English abstracts employed more hedges markers such as "may," "possible," "approximately," "more likely," "about," and "probably." Self-mention was used more in English as opposed to Arabic medical abstracts.This was not the case with language sciences abstracts, where English and Arabic abstracts had similar usage frequency (Alotaibi, 2015).This highlights the expected differences in using metadiscourse markers among different science fields.Interestingly, in the present study, all the self-mention occurrences were in the plural form.In contrast, the abstracts in both languages were devoid of the use of the singular or the objective forms of self-mention.This again indicates the distinct nature of the medical field texts, as previous studies from other fields of science showed that the use of the self-mentioned was distributed among singular and plural pronoun forms and the objective form (Alharbi & Swales, 2011;Alotaibi, 2015;Gillaerts & Van de Velde, 2010).
Engagement markers were absent from the studied medical abstracts in both languages.This could point out the nature of medical abstracts, where they must be a standalone piece of text not dependent on other sections of the article.Boosters and attitude markers were also absent from the studied abstracts.We believe that due to the nature of medical texts, writers tend to be less expressive of their attitudes, thus not relying on attitude markers, and less dependent on certainty markers in the text as they mainly discuss direct factual evidence; therefore, not relying on boosters.
The current study has some limitations.It included mainly abstracts written by non-native English speakers, which could have implications for the authenticity of the English text.These abstracts could have been written by non-native speakers but edited by language specialists.To this end, studies that include a sample of English abstracts written by exclusively native-English authors are warranted.Yet, such studies from the medical field will not usually be published in journals that require Arabic abstracts.This is because such journals are exclusively based in Arab countries, which is not the typical publication venue for English-native medical authors.On another front, the only definitive manner to verify if the text writer is a native speaker of English is by directly asking them about that fact.In that regard, interviewing the text writers could be of added value, pointing to the text writer's perspective when discussing the implications of metadiscourse features, which could have pedagogical value to language instructors, translators, and designers of writing programs.
In conclusion, this study showed that there are similarities in rhetorical move structures; nevertheless, Arabic and English medical abstracts displayed a considerable degree of divergence in the use of metadiscourse features.These findings highlight the fact that languages offer their speakers a variety of rhetorical and stylistic options from which they can choose to express their thoughts, ideas, and attitudes in both modes of communication, speaking and writing.Consequently, these findings might have useful applications in the fields of translation, translation training, cross-cultural communication, and language pedagogy.

Figure 1 .
Figure 1.Distribution of Arabic and English abstracts as structured versus non-structured.

Figure
Figure 2. The frequency of (A) interactive and (B) interactional metadiscourse features according to move structure.

Table 4 . Comparison of the move patterns in Arabic versus English abstracts from medical journals
*Using Pearson's Chi-square test.

Table 5 . Move length analysis* Move structure
*included in the analysis were only abstracts with five moves in both English and Arabic.

Table 6 . Total frequency and percentage of each interactive and interactional metadiscourse feature in English and Arabic abstracts Metadiscourse features
were in the pronoun form such as ': we did ‫ﻗ‬