Reporting Verbs in Hard and Soft Sciences of Journal Article Abstracts: An Insight from Indonesia

Abstract Reporting verbs (RVs) express varying degrees of alignment and evaluate the credibility and merit of the reported claims in academic writing. This study aims to explore RVs’ tenses, frequencies, and functions in two major disciplines: hard science and soft science. Consequently, two specialised corpora were built from research article abstracts published in 2021–2022. The hard science corpus comprised 19,512 words; meanwhile, the soft science corpus contained 20,222 words. The Constituent Likelihood Automatic Word-tagging System (CLAWS) web tagger was utilised to identify RVs, and Antconc software was employed to retrieve RVs. The study findings revealed that RVs in hard science and soft science were employed in four tenses forms, i.e. past participle, present participle, infinitive, and past tense. Further, the study indicated that based on the frequencies, soft science employed RVs more frequently than hard science. However, based on functions, hard science utilised RVs more comprehensively than soft science. Hard science and soft science have distinctive ways of implementing RVs in the Indonesian journal article abstracts. Thus, designing an RVs list specialised for non-native English (NNE) students, authors, teachers, and instructors is recommended to strengthen each discipline’s academic writing proficiency.


Introduction
English is the lingua franca that acts as the primary hub to accommodate academic publications (Flowerdew & Li, 2009;Gnutzmann & Rabe, 2014;Mauranen et al., 2010).While this might cause a loss of domain in other languages (Baker, 2016;Phillipson, 2008), the preference for publishing research in English is affected by multiple factors, i.e., providing the author with a sense of professionalism, reaching broader target audiences, and the legitimate language status that has become academic norms and practices in research communities (Curry & Lillis, 2004;Hamid, 2006;Petersen & Shaw, 2002;Schluer, 2014).Consequently, as English domination in research publications is inevitable, it could affect writing styles, teaching, instructional practices, and attitudes among non-native English (NNE) and novice authors (Liardét, 2018;Rubin et al., 2005;Zhang et al., 2015).
The existing literature reveals that writing academic writing is different from other genres; for instance, the uptight and formal nature of academic writing makes the authors remain cautious and meticulous in composing academic writing (Bennett, 2009;Hundt & Mair, 1999;Hyland et al., 2017).Therefore, although English has become a requirement, writing academic writing in English could be challenging, especially for NNE and novice authors.Moreover, to succeed in academic writing, the authors should consider English systematic structure and grammar patterns (Ma & Qian, 2020;Su & Zhang, 2020;Su et al., 2021).Consequently, the present study explores reporting verbs (RVs), one of the crucial aspects of writing in English for academic purposes.
RVs are defined as verbs put in a citation (Thompson & Yiyun, 1991), i.e., argue, believe, claim, denote, find, suggest, think, that enable authors to contextualise the proposition by expressing varying degrees of alignment and evaluating the credibility and merit of the reported claim (Liardét & Black, 2019).Further, in academic writing, RVs serve four essential functions: (1) constructing argumentation, (2) presenting logical thinking processes, (3) indicating research steps, and (4) determining research findings (Charles, 2006;Cumming et al., 2016;Liu et al., 2016).The existing literature regarding RVs in academic writing found that it has been scrutiny among NNE authors (Chen et al., 2022;Kwon et al., 2018;Liardét & Black, 2019;Marti et al., 2019;Vass, 2017).However, when RVs are investigated in many NNE contexts and backgrounds, the study on RVs written and published by NNE authors in Indonesia remains lagging.Therefore, the present study urges exploring RVs in Indonesian journal article abstracts to bridge the existing research gap.
As the nation's official language, Indonesian is primarily used to teach, deliver instructions, and conduct research writing in higher education.Therefore, only English-majored students must comprehensively conduct their study and research writing in English.Conversely, non-English majors can still conduct research writing in the Indonesian language.However, since Indonesian higher education has begun to pursue world-class university recognition, it has become compulsory for students, scholars and academics to write publications in English, regardless of the major.Furthermore, the Indonesian journal publishers that allow the authors to submit in the Indonesian language still require the abstracts to be written bilingually, i.e., in Indonesian and English.Accordingly, exploring how NNE students and authors apply the RVs in composing journal article abstracts is a crucial topic to discuss in the present study.

Research questions
This study formulated two research questions: (1) how are RVs tenses employed in hard and soft science in Indonesian journal article abstracts?and (2) what are the frequencies and functions of RVs in hard and soft science in Indonesian journal article abstracts?

Literature review
Verbs in academic writing across disciplines have been a fascinating topic urgently needed to assist the development of English for academic purposes, i.e., enhancing understanding of verb meanings and how to put them in sentences (Deng et al., 2022).RVs in academic writing draw the attention of linguists and scholars, considering that the authors employ these verbs to put in a citation that establishes the author's stance, i.e., acceptance, neutrality, and rejection (Thompson & Yiyun, 1991).In academic writing, citations are the implementation of knowledge advancement and demonstrate the credit indicating the study and the authors are connected in more outstanding social networks (Hyland & Jiang, 2019).Thus, the authors can convey alignment and assess the credibility and merit of any legitimate claim (Liardét & Black, 2019).
Citation styles and subject areas of academic writing affect the verb tenses used (Winkler & Metherell, 2010).Further, it is explained that the Modern Language Association (MLA) and The Chicago Manual of Style (CMS) employ the present tense, American Psychological Association (APA) utilises past or present perfect tense.Humanities papers commonly use the present tense.Social science, particularly life science, also employs past or present perfect tense.Reports utilise past tense.Generalisations use the present tense.Reports that are still valid also employ the present perfect tense.Furthermore, A study further discovers that present simple, present progressive, present perfect, past simple, and past perfect are verb tenses in journal article abstracts published by Elsevier (Kwary et al., 2017).RVs serve four primary functions: (1) constructing argumentation, (2) presenting logical thinking processes, (3) indicating research steps, and (4) determining research findings (Charles, 2006;Cumming et al., 2016;Liu et al., 2016).
Several empirical studies have been devoted to RVs in academic writing.For example, the study on RVs in medical article journals explains that these verbs are used to report statements, citations, and discourse roles (Thomas & Hawes, 1994).This study further denotes that RVs are divided into three categories: (1) verbs referring to the real world or experimental activities, (2) discourse verbs, and (3) cognition verbs.Further, there is a study of RVs among NNE first-year students from China, India, Korea, and other backgrounds in writing programs in the US (Kwon et al., 2018).It is reported that these novice NNE students frequently produced the think and find category.Further, based on functions, the previous study reveals that RVs fall into self-reference, i.e.I think, and uncited generalisation, i.e. some people think.The study further explains that these novice NNE students' awareness of RVs choice rises when given appropriate instructions.Besides, a study investigated RVs as projections in summary writing by first, second, and third-year university students EFL students in China (Chen et al., 2022).This study points out a qualitative development in the use of RVs across the three years of university study but no significant quantitative development.The study then concludes that the higher the level of the students, the more accurate and appropriate the selection of RVs to project their writing.
Furthermore, a study compares native students and advanced authors using RVs (Liardét & Black, 2019).This study discovers that advanced authors prefer dialogic contracting RVs, i.e., show and find to endorse preposition; meanwhile, native students tend to utilise expanding RVs, i.e., suggest which treats evidence as an option for consideration, and RVs, i.e., state to quote outside experts.Nevertheless, the previous study shows that native and advanced authors tend to utilise neutral RVs.Consequently, this comparison provides a roadmap that helps novice authors adapt advanced authors' RVs conventions.A study also explores the clause following RVs, i.e. the research suggests from four groups of authors: native advanced authors, NNE advanced authors, native novice authors, and NNE novice authors in applied linguistics (Marti et al., 2019).The findings indicate that expertise levels matter more than being native or NNE in academic writing.
In Chinese-English academic writing across disciplines, RVs fall into the Identification category, specifically unnamed citations.The usage includes the passive form of RVs, zero-subject sentences, quotations, and RVs with an adverb to present reported ideas as a general view rather than the authors' opinion.These unnamed citations increase the cited propositions' debatability (Wang & Hu, 2022).Furthermore, a study compares RVs in Arab and English-based citation academic writing.The study discovers that RVs in the Arabic-based citation style demonstrate equivalence with some established academic norms to showcase the author's role in the academic community.However, the authors contrast these norms, which signifies the vital role of culture undermining English's dominance as the lingua franca in the academic realm (Alramadan, 2023).
Further, a previous study reports that verbs in the law journal article, the Supreme Court majority opinion, and the Supreme Court dissenting opinion can be used as RVs (Vass, 2017).The study further mentions that these verbs can be speculative or quotative category, whereas speculative if the verb is used with a first-person pronoun, i.e., I suggest; meanwhile, it is quotative when RVs is partnered with a third-person pronoun, i.e.Hyland suggests.Thus, third-person pronouns and RVs make an impersonal stance and distance authors from inserting personal contributions to the claims.Although these previous studies have examined RVs in varying contexts, the previous studies lack discussion of RVs in academic writings in Indonesia.Investigating RVs is crucial, considering the number of Indonesian authors who published journal articles are striving.Thus, the findings from this study are expected to contribute significantly to English for academic purposes.

Methodology
The present study's corpora were derived from the Indonesian national journal indexation system called the Science and Technology Index (Sinta).Sinta is an extensive web-based research information system that comprehensively measures the performance of Indonesian scientists and scholars.Based on the statistics, Sinta indexed significant numbers of 246,784 authors (SINTA -Science and Technology Index, n.d.-b) from 5,297 universities (SINTA -Science and Technology Index, n.d.-a), 7,408 journal publishers (SINTA -Science and Technology Index, n.d.-c).Consequently, Sinta is the most extensive Indonesian academic community database providing information regarding journal articles published in various subject areas.In order to accomplish the objectives of the present study, the selected subject areas are limited to (1) science, (2) social, and (3) humanities.Journal articles obtained from science were referred to as hard science; meanwhile, journal articles from social and humanities were referred to as soft science.
Three criteria were set in selecting the journal articles as the corpora.First, the journal articles must be published by legitimate university publishers that received the highest national indexation rank indicated by Sinta 1 (S1) accreditation and Scopus indexed (see Appendix 1.) Thus, it must be ensured that the selected article abstracts were published when the journals were still indexed when this study was conducted.Second, the present study explored only the abstract section of the journal articles.An abstract is the vital section of a journal article that briefly provides the ideas, steps, results and discussion, and conclusion of a research.Presenting research in compact and to-the-point manners must require suitable RVs.Third, only the latest publications from 2021-2022 were employed in this study.The present study was not a diachronic study, which means the study did not aim to compare or analyse RVs changes from different years.Thus, the latest issues were selected to represent the recent RVs application in academic writing.Eventually, two corpora were built; a hard science corpus comprised 19,512 words, and a corpus of soft science contained 20,222 words.
First, The Constituent Likelihood Automatic Word-tagging System (CLAWS) Tagger was employed to identify RVs tenses.The tagger with 96.97% accuracy and an error rate of only 1.5% with 3.3% ambiguities (UCREL Lancaster University, n.d.-b) was employed to determine RVs tenses.UCREL CLAWS7 Tagset (C7) were utilised to annotate the text (UCREL Lancaster University, n.d.-c) in this study.CLAWS tagger offered several tagset options; however, the C7 tagset was selected because it is the current standard of data tagset of the previous tagsets.Second, the output texts from Free CLAWS Tagger were converted to plain text (txt) format and then inputted into a software called AntConc, (2022).The conversion was required because Antconc could only process files in the txt format.
Further, using the basic guideline manual (UCREL Lancaster University, n.d.-a), RVs were retrieved from Antconc.The analysis was conducted in three steps.First, RVs were categorised based on the respective tenses from the data retrieved.The existing literature revealed that only several verb tenses commonly appeared in academic writing.Second, the frequencies and functions of RVs from hard science and soft science were presented.The frequencies were required to identify how often certain RVs appeared in the journal article abstracts.The functions were necessary to explore how RVs were employed within the journal article abstracts.Third, examples of RVs were presented in this study.Thus, the context and implementation of RVs in the Indonesian journal article abstracts can be observed thoroughly.

Results
Drawing from the method explained in the previous section; the present study discovered four verb tenses that appeared from the 19,512 words corpus of hard science and the 20,222 words corpus of soft science, respectively.These RVs tenses were past participle verb, present participle verb, infinitive verb, and past tense verb.From these findings, only future tense verbs did not appear in the journal article abstracts.Further, the distribution of RVs in each tense was normalised per 1,000 words, presented in Table 1.
Table 1 showed that the highest hint in the hard science corpus was past participle verbs reaching 34,29 hints per 1,000 words.Then, present participle verbs reached 15,84 hints per 1,000 words.It was followed by infinitive verbs, which reached 15,63 hints per 1,000 words.Finally, past tense verbs reached 9,07 hints per 1,000 words.Likewise, in the soft science corpus, the highest hint was past participle reaching 27,89 hints per 1,000 words.Then, it was followed by the present participle verb, which reached 22,55 hints per 1,000 words.Following that were infinitive verbs, which reached 18,35 hints per 1,000 words.Ultimately, past tense verbs reached 9,20 hints per 1,000 words.Further, the present study also revealed that the overall frequency of RVs in hard science was 74,83 hints per 1,000 words.Meanwhile, the overall frequency of RVs in soft science was 77,98 hints per 1,000 words.The findings indicated that in the Indonesian journal article abstracts, RVs were employed more frequently in soft science than hard science.A possible explanation is that it could be due to the nature of soft science articles, which tends to be more descriptive than hard science articles, which use more numbers and statistics.The existing literature mentioned that RVs have four main functions.This study found that in hard science, RVs implemented those four functions in the Indonesian journal article abstracts, as presented in Table 2.
Moreover, this study presented examples of how RVs were implemented in journal article abstracts.Example 1 until example 11 were the implementations of RVs in past participle verb forms in hard science.e.g. 1, "In this study, 30 training images and 30 testing images from Kotabaru Oncology Clinic in Yogyakarta were used."e.g. 2, "In this work, we are interested in finding new catalysts for catecholase, whose principle is based on the oxidation reaction of catechol to o-quinone."e.g. 3, "Then, we study the integration of IoT services based on application placement strategies on the fog cloud compared to the traditional centralized cloud strategy."e.g. 4, "Data associated with risk factor values were obtained from an LPG storage tank in a gas distributor company."e.g. 5, "First of all, a traceability system architecture was developed."e.g. 6, "This study can be considered a valuable reference for in-depth works on current related issues."e.g. 7, "Moreover, helical Darrieus was found to produce lesser noise and suitable for conventional areas."e.g. 8, "The design analysis of RESO-based ADRC has been presented and a computer simulation has been conducted to verify the effectiveness of the proposed controller."e.g. 9, "In this study, acid-treated corn leaves combustion was performed to produce high purity silica (SiO 2 )." e.g. 10, "Therefore, a bibliometric analysis was carried out to evaluate the performance and emissions of a diesel engine with the B100 being tested on a multi-cylinder diesel engine for cars."e.g.11, "The diffraction pattern shows that high crystalline zeolite Na A-X was produced at temperatures of 100°C and 8 h hydrothermal time." The present study revealed that in past participle verb forms, RVs that served the function of constructing argumentation were example 3 and example 6.Then, RVs that presented logical thinking processes were example 2. Further, RVs that served as indicating research steps were example 1, example 4, example 5, example 8, example 9, example 10, and example 11.Ultimately, RV that served as determining research findings was example 7.
Example 12 and example 13 were the implementations of RVs in present participle verb forms in hard science.e.g. 12, "The developed bamboo stick simply operates using ultrasound sensors for sensing the impediments before contact and a water detection sensor for water detection on the pathway."e.g. 13, "The result showed that the activity of the electrode and energy consumption were increased with values of 29 and 23%, respectively, by adding 3 g of the media." It was suggested that in present participle verb forms, RV that indicated research steps was example 12.Then, RV that served as determining research findings was example 13.
Further, example 14 and example 15 were the implementations of RVs in infinitive verb forms in hard science.e.g.14, "Spin-assisted layer-by-layer assembly (SA-LbL), one of the LbL variants, was selected for the fabrication of a nanofiltration membrane due to its versatility to produce an ultra-thin film with highly controlled film properties within an incredibly short time."e.g. 15, "This study aims to determine the length-weight relationship and the condition factor (K) of Trachurus trachurus." It was denoted that in infinitive verb forms, RV indicating research steps was example Then, RV that served as constructing argumentation was example 15.
In addition, example 16 was the implementation of RV in the past tense verb form in hard science.e.g. 16, "The novelty of this study showed surface of composite based of hydroxyapatite has the ability to highlight the surface designed for efficient removal of Cu 2+ and Zn 2+ ions, from aqueous solutions more than other studies."Ultimately, the study signified that RV functioned as determining research findings in past tense verb forms.Furthermore, instead of serving four functions according to the existing literature, in the current study, RVs in soft science implemented only two functions in Indonesian journal article abstracts, as presented in Table 3.
Table 3 illustrated that RVs in past participle tense forms were used with 2,16 hints, based with 1,90 hints, conducted with 0,89 hints, and collected with 0,76 hints per 1,000 words, respectively.Meanwhile, RV in the present participle forms were using with 4,95 hints per 1,000 words.Further, RVs in infinitive verb forms were analyze with 0,82 and learn with 0,63 hints per 1,000 words, respectively.Lastly, RV in past tense verb forms were showed with 1,14 hints per 1,000 words.The present study discovered that in past participle verb forms, RVs that indicated research steps were example 17, example 19, example 20, and example 21.Eventually, RV that served as determining research findings was example 18.
Example 22 was the implementation of RV in the present participle verb forms in soft science.e.g.22, "The data were collected from the targeted population using a questionnaire and interview whether gender and study duration influence students' speaking strategies use or not." It was implied that in the present participle verb forms, RV only indicated research steps.
Further, example 23 and example 24 were the implementations of RVs in infinitive verb forms in soft science.e.g. 23, "The research method used to analyze social phenomena is a qualitative approach, using direct interviews and observations as the main data collection instruments."e.g. 24, "32 EFL learners were exposed to learn vocabulary by the use of WhatsApp application for a period of 6 weeks." It was signified that RVs only indicated research steps in infinitive verb forms.
Then, example 25 was the implementation of RV in past tense verb forms in soft science.
e.g. 25, "The expert validation results showed that the developed web-based teaching materials was feasible with a very good predicate." Eventually, it was denoted that in past tense forms, RV only served as determining research findings.

Discussion
In the academic community globally, English is established as a legitimate language for publications (Flowerdew & Li, 2009;Gnutzmann & Rabe, 2014;Mauranen et al., 2010), communications, and connections with the targeted and larger audiences (Curry & Lillis, 2004;Hamid, 2006;Petersen & Shaw, 2002;Schluer, 2014) is indisputable.The present study explored academic English use in nationally indexed journal article abstracts.Supporting the statements from the existing literature, the research article abstracts indexed by Sinta published in Indonesian consistently used appropriate formal English.Moreover, the journal selected in this study were Scopus indexed, which demanded the use of language that could establish a connection not only to Indonesian audiences but also audiences and the academic community worldwide.Consequently, there was a greater demand for novice authors, students, and scholars, mainly in NNE countries (Liardét, 2018;Rubin et al., 2005;Zhang et al., 2015), to emphasise using proper English to warrant successful academic writing.
Moreover, with rigid and impersonal characteristics of academic writing (Bennett, 2009;Hundt & Mair, 1999;Hyland et al., 2017), understanding the correct language structure and grammar would be beneficial to achieve success in academic writing (Ma & Qian, 2020;Su & Zhang, 2020;Su et al., 2021).Building upon previous research findings, the present study carefully selected research articles, focusing on academic articles from esteemed national journals with prominent indexation, i.e., Sinta and Scopus.This meticulous selection process aims to ensure the utmost English language proficiency and professionalism exhibited by the authors in their research article.Consequently, the data gathered from these studies could reflect the English grammar and language structure in reputable Indonesian national journal research articles.Understanding the use of verbs is vital in academic writing (Deng et al., 2022) for a myriad of reasons, i.e., RVs were employed in citing and expressing the authors' stance, such as agreeing, rejecting, or evaluating the credibility of an expert opinion, making claims, and composing reports (Thompson & Yiyun, 1991).
The present study's first objective was to explore RVs tenses employed in hard and soft science in Indonesian journal article abstracts.The existing literature mentioned that citation style, subject areas, interpretation, generalisation, and the report's validity could affect the verb tense employed in academic writing (Winkler & Metherell, 2010).However, since the data of this study was obtained from the journal article abstracts, the citation style used was hardly determined.Consequently, the interpretation of the tenses used in this study was the subject areas, i.e., hard and soft science.Thus, the present study followed the existing literature suggestion for conducting a study on RVs tenses based on subject areas.Further, the finding of this study confirmed the existing literature that RVs from different subject areas implemented different RVs tenses.
The literature found that humanities and social papers commonly used the present, past, and present perfect tenses; meanwhile, life science mainly employed past or present perfect tense.This study resonated with the literature given the present findings that revealed RVs in tenses in hard and soft science comprised all four primary verb tenses: (1) past participle verbs, (2) present participle verbs, (3) infinitive verbs, and (4) past tense verbs.It indicated that the RVs tenses produced by the journal article abstracts indexed by Sinta aligned with RVs tenses in journal article abstracts published by Elsevier (Kwary et al., 2017).Further, the present study and the existing literature also confirmed that the future verbs were hardly employed and appeared in the journal article abstracts.
The present study's second objective attempted to investigate the frequencies and functions of RVs in hard and soft science in Indonesian journal article abstracts.The frequencies were intended to give insight into the differences and similarities of RVs and hints between hard science and soft science.The study discovered that based on frequencies, RVs appearance in hard science and soft science were almost identical, with soft science slightly higher with 77,98 hints than in hard science with 74,83 hints, which signified that in terms of frequencies, soft science in the Indonesian journal article abstracts were more significance than hard science.The present study revealed that past participle verbs were the highest RVs form in hard and soft science, followed by past participle verbs, present participle verbs, infinitive verbs, and past tense verbs.It contradicted the existing literature finding that the present simple and past simple verb tenses were the most frequently used in journal article abstracts published by Elsevier, followed by present progressive and past perfect verb tenses (Kwary et al., 2017).The study's second objective was also to scrutinise the functions of RVs in the Indonesian journal article abstracts.The literature explained that there were four primary functions of RVs in academic writing, which included: (1) constructing argumentation, (2) presenting logical thinking processes, (3) indicating research steps, and (4) determining research findings (Charles, 2006;Cumming et al., 2016;Liu et al., 2016).However, when examining the functions of RVs in this study, it was found that RVs in hard science entailed all of these four functions; meanwhile, RVs in soft science only served two functions, i.e., denoting steps taken in the research and establishing the research findings.A possible explanation was that it could be due to the respective discipline's differences in nature and demand.Thus, the present study suggested that functions served by RVs in hard science tended to be more comprehensive than in soft science.Although, it did not necessarily mean to generalise that soft science lacked argumentation or logical thinking process in writing journal article abstracts.Therefore, the present study confirmed the existing literature that RVs in hard science comprised all of the proposed functions.Moreover, this study finding regarding RVs in soft science slightly differed from the existing literature because only two out of four functions appeared.
The present study contributed to bridging the gap from previous studies.The study explored RVs implemented in the Indonesian journal article abstracts.Meanwhile, although the existing literature provided evidence from various NNE authors, students' backgrounds, and countries, they lacked empirical evidence of RVs from Indonesia.However, this study supported the previous study's finding, which observed RVs' production by undergraduate and graduate students with NNE backgrounds in the US from the semantic category (Kwon et al., 2018).Further, it was revealed that from the semantic category, RVs in hard and soft science of journal article abstracts from Indonesia and NNE students in the US employed verbs from argue, show, find, and think categorisation.However, it must be addressed that the present study focused on other aspects, i.e., tenses and frequencies.Thus the differences between this study and the previous study regarding rhetorical category might be overlooked.Besides, there were previous studies that focused on particular subject areas, i.e., medical science (Thomas & Hawes, 1994), law (Vass, 2017), and applied linguistics (Marti et al., 2019).Meanwhile, the present study used general subject areas classification, i.e., hard and soft science journal article abstracts.The present study validated some of the findings from medical science, where it was found that they also implemented RVs for stating specific results and findings (Thomas & Hawes, 1994).However, the present study did not yet attempt to explore the rhetorical functions of RVs as conducted in the previous study.Further, the present study supported the previous study on RVs in law journal articles and the Supreme Court context, in which first-person pronouns followed by RVs would become quotative verbs (Vass, 2017).Quotative verbs serve to justify propositions by referring to the report of others.Thus, the present study strengthened the significance of RV functions to align and evaluate the credibility and merit of the reported claims.Moreover, the present study conformed to the existing literature on RVs in applied linguistics.Hard and soft science journal article abstracts from Indonesian journals and applied linguistics papers of advanced and novice NE and NNE authors utilised RVs from argue, show, find, and think categorisation.Besides, both studies considered the significance of nativeness and expertise regarding RVs choices.
Furthermore, the present study conducted a different path in obtaining the data for the study by analysing NNE advanced authors in Indonesia; meanwhile, the previous studies attempted to compare novice and advanced authors' RVs implementation (Liardét & Black, 2019) and first, second, and third-year students RVs writing (Chen et al., 2022).The previous study on novice and advanced authors showed that the RVs they employed were identical in frequency but distinctive in the RVs forms made (Liardét & Black, 2019).Thus, the present study strengthened these findings, showing nearly identical frequency between hard science and soft science.However, in the present study, all authors were considered advanced because they had already made publications in nationally accredited journals.Additionally, the existing literature denoted a decent progression from the first, second, and third-year, where RVs became comprehensive (Chen et al., 2022).However, although the present study did not explore RVs in students writing, the present study's findings resonated with the literature.The possible explanation that could be offered was that with more experience and knowledge in writing, the RVs usage was getting appropriate.With advanced authors' journal article abstracts as the data, it could be assumed that these authors also learned and earned their ability to select suitable RVs for their academic writing process.
Previous studies further attempted to compare RVs between two languages, i.e., RVs between English and Chinese (Wang & Hu, 2022) and RVs between English and Arabic (Alramadan, 2023).The literature described that RVs, i.e., argue and think existed in the article written in English and Chinese in their passive forms (Wang & Hu, 2022).The present study also confirmed the previous study's findings because, in the Indonesian journal article abstracts, similar forms and functions of RVs appeared, and these RVs also appeared in passive forms.Ultimately, in English and Arabic contexts, the existing literature revealed that RVs in academic writing in both languages running functioned to acknowledge but less endorse, distance, and contest other sources in their citation.The present study also supported the previous study's finding because RVs in the Indonesian journal article abstracts also functioned to acknowledge, endorse, keep distance, and contest external sources.However, the present study took a different approach from these two previous studies because it did not attempt yet to compare RVs in English and Indonesian.

Pedagogical implications
The present study extends the insights from previous empirical studies that reported on RVs usage among experts, novices, natives and NNEs authors in varying subject areas and authors' backgrounds and expertise outside of Indonesia-proving that similar studies were rarely done in the Indonesian context.Theoretically, this study gives insights into RVs are employed, distributed by frequencies, and functions in hard science and soft science of the Indonesian journal article abstracts.Pedagogically, the findings could be instructional materials for academic writing subjects to enhance comprehension of RVs to NNE authors in Indonesia and a valuable source of information for students and teachers in other NNE countries.Furthermore, academic writing teachers and instructors could develop lessons or instructions that enhance novice authors' and students' comprehension regarding RVs.

Conclusions
RVs hold a crucial aspect in academic writing.The present study already explores how this essential aspect is employed in hard science and soft science of Indonesian journal article abstracts.The findings show similarities and differences in the RVs employed by both disciplines.The similarities are the verb's tense forms, whereas hard science and soft science employed: (1) past participle verbs, (2) present participle verbs, (3) infinitive verbs, and (4) past tense verbs.However, each RV in the verb tenses in hard science and soft science were varied.Further, following the frequencies findings, the study indicates that soft science employed more RVs than hard science.Additionally, according to the findings regarding the functions, the hard science signified a more comprehensive RVs functions that entailed: (1) constructing argumentation, (2) presenting logical thinking processes, (3) indicating research steps, and (4) determining research findings; meanwhile, soft science only comprised demonstrating research steps and establishing research findings.
Moreover, the findings of this study also signify a crucial point that although Indonesia is one of the NNE countries, the RVs produced and employed are aligned to the suggested RVs by the existing literature and empirical studies that utilised Elsevier as the data.It indicates that advanced Indonesian authors, who are members of a vast academic community, can adapt to the demand of using English in their academic writing.Further, regardless of the contributions, this study has a limitation.The present study used two corpora built from Indonesian journal article abstracts.Thus, the study's findings could not be generalised to other NNE countries.Therefore, it is suggested that future studies must be conducted using a larger corpus from NE countries and other NNE countries other than Indonesia and or comparing RVs in English and Indonesian.

Funding
The author received no direct funding for this research.

Table 3 . Reporting verbs frequencies and functions in soft science
Additionally, examples of RVs implementations in soft science were presented in this study.Example 17 until example 21 were the implementations of RVs in past participle verb forms in soft science.
e.g. 20, "The data acquired through the closed-ended items of the questionnaire was analyzed by a one-sample t-test while the data obtained through an open-ended items of the questionnaire; interviews, and informal conversation were analyzed qualitatively through verbal descriptions."e.g.21, "Data was collected through vignettes, participant observations, and students' written texts."