INTRODUCTION

Mastery of academic writing in English is essential for the academic success of graduate students, as this proficiency is crucial for thesis composition and scholarly publication (Dehghan, 2021; Li et al., 2023). A key aspect of scholarly works—such as articles, dissertations, seminar papers, and essays—is the practice of citing contributions from other authors (Dontcheva-Navratilova, 2016; Shooshtari Gooniband et al., 2017). As noted by Swales (2014), citation serves as the most explicit and readily apparent marker that a text qualifies as academic in nature. Authors utilize citations not just to endorse their findings but to enhance their credibility and situate their research within the existing body of knowledge (Mansourizadeh & Ahmad, 2011). Citations also enable academic writers to introduce and discuss the work of peers, aligning themselves with specific disciplinary communities (Donner, 2021; Hu, 2023). Furthermore, they validate the originality of research while establishing the authors’ scholarly authority and acknowledging those who have influenced their work (Bruton et al., 2024). In this manner, citations play a significant role in the ongoing dialogue within academic discourse, delineating how current contributions build upon, contest, or corroborate prior research.

Competence in citation is an integral intertextual skill that clarifies and articulates the writer's perspective and fulfills various rhetorical functions in academic writing. Equipped with this skill, academic writers possess the essential capabilities to excel in numerous areas, including contextualizing their research within a specific discipline, illustrating the importance of their study, formulating their academic assertions, and enhancing both the credibility and novelty of their work (Mansourizadeh & Ahmad, 2011; Wu, 2013). The evaluation of citation competence, particularly within applied linguistics, often involves scrutinizing writers' citation practices. Citations function as a crucial mechanism through which academic authors exhibit their familiarity with their specific fields of study, affirm their membership within distinct discourse communities, and align their work with particular theoretical frameworks or methodologies. By doing so, they situate their research within the broader academic context, facilitating engagement with peer contributions and promoting a sense of belonging and intellectual continuity within their disciplines (Bruton et al., 2024; Hu, 2023; Nesi, 2021; Thompson & Tribble, 2001; Zhang, 2022).

In recent decades, citation practices have garnered considerable scholarly attention, resulting in a wealth of research exploring various facets of this phenomenon. Numerous studies have concentrated on specific sections of academic writing, such as the forms of citation employed in introductions (Arizavi & Choubsaz, 2021; Jalilifar, 2012), the literature review (Badenhorst, 2017), and discussions (Dobakhti & Zohrabi, 2018; Samraj, 2013). Additionally, comparative analyses of citation patterns between non-native and published works (Ahn & Oh, 2024) and across different proficiency levels (Mansourizadeh & Ahmad, 2011; Petrić, 2007; Samraj, 2013) have been conducted. Furthermore, comparative studies examining citation practices in relation to different writing tasks, such as undergraduate papers, master’s theses, and PhD dissertations, highlight the varying approaches to citation across educational levels (Badenhorst, 2019; Lee et al., 2018).

In light of the considerable research conducted thus far, the need for a systematic investigation comparing the citation practices of native and non-native English writers is evident. A crucial aspect of this comparison lies in the potential relationship between a writer's linguistic and rhetorical background and their citation choices. Specifically, the study investigates whether nativeness correlates with a more sophisticated command of citation for persuasive and critical engagement, while non-nativeness might be associated with a more explicit, formulaic use focused on fulfilling academic requirements. This comparison is necessary because it moves beyond a simple deficit model to illuminate the distinct rhetorical challenges faced by non-native writers, such as integrating sources to construct a persuasive argument rather than merely attributing ideas. Marti et al. (2019) indicate that non-native expert writers still exhibit differences in constructing stance through reporting clauses compared to native expert writers, suggesting that linguistic and rhetorical proficiency, not just disciplinary expertise, shapes citation behavior. Presently, there is a conspicuous absence of cohesive studies that explore the forms and functions of citation practices among these groups within the field of applied linguistics. The act of citation is critically significant, as the competencies necessary for effective, precise, and context-appropriate citation are notoriously complex for learners to acquire (Nesi, 2021). By addressing this identified gap and building on prior findings, the present study aspires to furnish illuminating insights into the pedagogical ramifications of citation practices, thereby enhancing the academic writing proficiency of non-native writers.  

LITERATURE REVIEW

Intertextuality, a concept initially articulated by Kristeva (1980), posits that all texts are interconnected and that both authors and readers leverage these connections when creating and interpreting texts. This framework suggests that texts are not isolated entities but rather dynamic constructs, evolving through their relationships with other texts. Barthes (1986) reinforces this perspective by asserting that a text comprises a multidimensional space filled with various writings, implying that meaning is not singular or fixed but instead is derived from an interplay of influences. Prior (2004) argues that each new text is informed by existing texts that mediate the construction of meaning, whether consciously or unconsciously. Bazerman (2004) further elaborates on this by emphasizing that intertextuality involves not only the selection of texts but also the manner in which they are used and the positioning of the writer in relation to these texts to achieve particular social goals.

Kristeva's framework emphasizes the dialogic character of intertextuality, particularly through the process of quotation, wherein she asserts that every text is comprised of a mosaic of quotations and represents the assimilation and transformation of preceding texts. This assertion underscores the interconnectedness of texts, inviting a critical engagement with how they inform one another. Bhatia (2010) expands on this notion by framing intertextuality as a dynamic process that reconfigures past texts in the present, emphasizing the reciprocal influence that texts exert on each other within literary discourse. His framework aligns with Kristeva's classification of intertextuality into horizontal and vertical types. Horizontal intertextuality pertains to the connections among specific texts, whereas vertical intertextuality examines how a text relates to its broader textual and cultural context, thereby situating it within both immediate and distant influences. Together, these perspectives illustrate the multifaceted nature of intertextuality, reinforcing the idea that texts are part of a complex network of meaning-making.

In the context of citation practices, Swales (1990) introduced a foundational framework that categorizes citations into integral and non-integral forms, a categorization that has influenced subsequent studies in academic writing (Ahn & Oh, 2024; Arizavi & Choubsaz, 2021; Gao et el., 2021; Jalilifar, 2012; Mansourizadeh & Ahmad, 2011; Samraj, 2013; Thompson, 2005; Thompson & Tribble, 2001). This binary distinction not only categorizes citations based on their syntactical function within a text but also reflects how writers construct their academic voice.

Investigations into the linguistic characteristics of citations are frequently associated with their rhetorical functions or the intentions of authors. According to Swales (1990), one of the primary rhetorical moves in scholarly writing involves the engagement with prior literature to situate one’s own research within a wider academic context. To explore this dimension of citation practices, Thompson (2001) conducted a functional analysis that differentiates between integral and non-integral citations. He identified that integral citations serve to narrate research processes, reflect cognitive processes, or reference specific scholars or works. In contrast, non-integral citations are utilized functionally to identify original concepts or to direct readers toward additional clarifications.

Thompson and Tribble (2001) expanded upon Swales' typology, investigating citation functions in doctoral theses across different fields. They distinguished between integral citations, which are further divided into subcategories such as verb-controlling, naming, and non-citation, and non-integral citations, categorized functionally into source, identification, reference, and origin. Their findings indicated that different disciplines favor distinct citation forms, revealing underlying norms and expectations within specific academic contexts. Additionally, subsequent modifications to these frameworks have included creating subcategories such as “agentive” and “non-agentive” integral citations (Campbell, 1990) and further refinements by Hyland (1999), who delineated integral citations into subject, non-subject (passive), and phrase-level adjunct structures. These evolving frameworks highlight the complexity inherent in citation practices and the dynamic interplay of voices within academic discourse.

Marti et al. (2019) explored the influence of nativeness and expertise on reporting practices in academic writing through a corpus-based analysis of various research articles in applied linguistics. They found minimal variation in reporting practices between expert native and non-native writers, while significant differences emerged between novice writers across these backgrounds, highlighting the discursive hybrid nature of non-native writing. This developmental aspect is also evident in the citation practices of students; Khamkhien (2025) found that Thai university students utilized a narrow range of reporting structures primarily to demonstrate topic understanding, often reflecting a non-committal stance compared to professional writers. Similarly, Mansourizadeh and Ahmad (2011) analyzed research papers by non-native English-speaking writers in chemical engineering and discovered that expert writers had a higher citation density and favored non-integral citation forms, emphasizing a balanced use of integral citations. Ahn and Oh (2024) further emphasized this developmental trajectory in citation practices by comparing citation forms used in Korean master's theses and research articles, revealing that expert writers preferred non-integral citations, while students leaned towards integral citations. Examining the citation practices in Chinese graduate writing, Li and Zhang (2021) compared literature reviews and discussion sections from 30 master’s theses and 30 doctoral dissertations. Results indicated that doctoral candidates more effectively employed citations to apply and affirm previous models, thereby constructing an academic identity.

Arizavi and Choubsaz (2021) examined citation representation within theoretical and applied linguistics articles, finding significant disparities in the use of integral and non-integral citations based on the research methodology. In a cross-linguistic and cross-disciplinary context, Shooshtari Gooniband et al., (2017) reported no significant differences in citation practices between Persian and English research articles. This aspect was further explored by Esfandiari and Saleh (2024), who analyzed citation practices in Economics and Industrial and Manufacturing Engineering, revealing that Economics writers preferred integral citations, while their Engineering counterparts gravitated towards non-integral citations.

Farnia et al. (2018) compared citation intricacies in articles from Iranian and international journals, noting higher citation frequency in Iranian publications and a preference for non-integral citations. Jalilifar et al., (2018) highlighted inappropriate citation strategies among Iranian English journal articles, indicating challenges in effective content borrowing. In a similar vein, Ebrahimi and Weisi (2019) investigated rhetorical moves in research article introductions in Computer Sciences, finding significant variation in the use of specific steps between Iranian and international journals, which underscores the influence of discourse community expectations. Additionally, Lee et al. (2018) reported lower citation density among ESL students, suggesting the pedagogical influence on citation employment. Overall, the existing literature establishes that citation practices are significantly influenced by expertise, nativeness, and disciplinary context, yet further research is warranted to understand the nuanced impact of nativeness in academic writing. McKinley and Rose (2018) highlight that nativeness continues to be favored in academic publishing, emphasizing the importance of considering language background, cultural context, and local discourse conventions in future investigations (Hyland, 2012).

PURPOSE OF THE STUDY

The current study aimed to examine and compare citation practices among English, Iranian, and Afghan scholars in the field of applied linguistics, utilizing the framework established by Thompson and Tribble (2001). This framework was selected not only for its enduring relevance and widespread application in citation analysis (e.g., Ahn & Oh, 2024; Arizavi & Choubsaz, 2021; Farnia et al., 2018; Jalilifar, 2012) but, more critically, for its comprehensive and functionally-oriented structure. Unlike models that focus solely on surface forms, Thompson and Tribble's (2001) framework integrates both citation forms and functions. This dual focus is essential for a meaningful cross-linguistic comparison, as it allows the analysis to move beyond mere frequency counts to investigate how and why different scholarly communities employ citations to construct knowledge, position their work, and persuade their audience. By analyzing 90 research articles across these three scholarly communities, this study aimed to uncover patterns and trends in how citations are integrated into academic writing.

CORPUS AND METHODOLOGY

This research constitutes both a qualitative and quantitative investigation based on a corpus of citation forms and functions within the domain of applied linguistics. The study involved the collection of three distinct corpora aimed at analyzing variations in citation practices influenced by author nativeness, specifically differentiating between L1 English, L1 Persian, and L1 Dari (for Afghan authors). The data gathered are representative samples of academic writing sourced from the field of applied linguistics, which was selected as the focal discipline due to the researchers' extensive knowledge and expertise, thereby positioning them to draw more reliable conclusions than they might in other disciplines.

To determine relevant journals, a comprehensive list from the 2019 Scimago Journal and Country Rank, produced by Scimago Lab and based on Scopus data, was assembled. Attention was directed toward the most relevant subject category, incorporating only Q1 and Q2 journals as ranked by their SJR (SCImago Journal Rank) indicator. The dataset comprised research articles and papers conforming to a specific structural format, including an abstract, introduction, literature review, methodology, results, and conclusion, all of which were authored or published within the timeframe of 2020 to 2024.

In total, the three corpora encompassed 90 research articles within applied linguistics, with each group contributing 30 articles. Specifically, the corpora included: (1) a collection of 30 published research articles by English authors, (2) a corpus of 30 published research articles by Iranian authors, and (3) a set of 30 published research papers by Afghan authors. The corpus was deliberately limited to 90 articles through a stringent selection process to ensure a valid, comparative analysis of citation practices. This controlled approach guarantees that observed differences are attributed to authorial background rather than other variables. A key factor in determining this number was the inherently limited pool of eligible articles published by the emerging Afghan scholarly community itself, which naturally constrained the corpus size.

The study operationalized native English writer by applying a composite of three criteria, requiring that authors satisfy at least one of the following: Wood’s (2001) strict criterion, which stipulates that the author must possess an English name, (2) the criterion proposed by Rowley-Jolivet and Carter Thomas (2014), which requires that authors be affiliated with an institution in an English-speaking country, and (3) the researchers' own criterion, which asserts that the authors were either born or educated in an English-speaking nation.

Researchers manually searched the corpus for specific tokens, such as acknowledged citation markers, including dates in brackets, authors' names, personal pronouns, and institutional references, which were cross-referenced against bibliographies. Following Hyland’s (1999) model, self-citations and proper nouns denoting schools of thought, such as the Chomskyan approach or references to groups with minimal relevance to the research article, like advocates of Positivism, were excluded. This exclusion was based on the premise that frequent self-citations could distort frequency counts and yield biased results, while generic references to groups lacking significant roles were deemed irrelevant.

To ascertain the total number of citations, every instance of brackets containing either single or multiple authors was systematically identified and tallied. Each citation that includes multiple, non-integral references is recorded as a singular citation (Bloch & Chi, 1995). Consequently, the citation counts reported in this study reflect the frequency of citation occurrences, irrespective of whether they are single or multiple reference entries (Thompson, 2000). All instances of citations were meticulously counted and subsequently classified into various forms and functions in accordance with the citation typology outlined in the following section.

The selected papers underwent meticulous manual examination, guided by frameworks established by Thompson and Tribble (2001) to analyze citation forms and functions. Forms of citation are identified based on whether they are integral or non-integral (Swales, 1990) which is predicated on the syntactic positioning of the cited author’s name. Thompson and Tribble (2001) sub-classification of integral and non-integral citations was utilized to identify their functions. Non-integral citations are divided into four primary functions: Source, which attributes a statement or proposition to another author, covering assertions such as factual accounts of other research findings or the attribution of ideas; Identification, which identifies an agent within the corresponding sentence; Reference, which is indicated by the phrase “see,” directing readers to further sources; and Origin, which designates the original creator of a concept or product. In contrast, integral citations are categorized into three types: Verb controlling, in which the citation acts as the agent governing a verb in either active or passive constructions; Naming, which refers to a noun phrase or part of one; and Non-citation, where an author is mentioned by name without including a year reference, often used to avoid redundancy when the reference has already been provided earlier in the text. The data was subsequently sorted to identify and classify citation types and functions and their frequency was calculated to discern potential differences among different disciplines and timeframes.

To ensure consistency in the coding process, detailed guidelines were created. These included types of functions, examples, and specifications of the coding scheme. Initially, the researchers independently coded about 10 percent of the dataset. They achieved an acceptable level of agreement for this subset through discussion and assessment of differences. This process led to high inter-coder reliability, shown by a Cohen’s Kappa Coefficient of .86. Throughout the analysis, the researchers held ongoing discussions to refine the coding guidelines and improve the accuracy of function assignments. After establishing this high reliability, the primary researcher applied the finalized coding scheme to the rest of the dataset.  

RESULTS

A comprehensive examination of the corpus revealed a total of 5,515 citations, of which 2,414 (43.77%) were categorized as integral, while 3,101 (56.23%) were classified as non-integral. The analysis indicated that 1,830 (33.18%) of the citations originated from articles authored by English researchers, whereas 2,144 (38.87%) were contributed by Iranian scholars, and 1,541 (27.95%) were attributed to Afghan authors. Table 1 presents a detailed analysis of the frequency of integral citations utilized by three groups of writers: English, Iranian, and Afghan.

Table 1: Frequency of Integral Citation by English, Iranian, and Afghan Writers

As Table 1 indicates, a total of 642 integral citations were identified in articles authored by English writers, with 316 citations categorized as verb controlling, 321 as naming, and only 5 instances of non-citation. In contrast, the corpus of articles written by Iranian authors yielded a total of 854 integral citations, which comprised 581 instances of verb controlling, 261 of naming, and 12 occurrences of non-citation. Similarly, Afghan writers demonstrated a total of 918 integral citations, with verb controlling used 604 times, naming 312 times, and non-citation appearing just twice. The analysis reveals a distinctive trend: Afghan writers displayed a marked tendency to employ integral citations over non-integral citations, while English and Iranian writers exhibited a preference for non-integral citation forms. Furthermore, the findings underscore that verb controlling emerged as the predominant form of integral citation among both Iranian and Afghan writers. In contrast, naming was the most frequently employed type of integral citation within English articles. Notably, non-citation was consistently the least utilized form across all groups examined.

Examples of Integral Citation

Verb-controlling: In this function, the citation acts as the subject or agent that controls the main verb of the sentence. It signifies that the cited authors are actively engaged in the action described, thereby attributing the argument or finding directly to them. This type of citation adds authority to the statement and integrates the authors’ contributions within the analysis, indicating their central role in the research context.

• Kormos (1999) focused on simulating conversations in oral-proficiency assessment of role plays and non-scripted interviews in language exams. (Ren & Seedhouse, 2024, English writer)

• Zhou, Dewaele, et al. (2023) highlighted the need for further studies on various aspects of FLPoM in different EFL contexts. (Derakhshan, & Noughabi, 2024, Iranian writer)

• Khosiyono et al. (2019) studied EFL teachers' attitudes towards the use of Maritime English textbooks (Orfan, Noori, & Akramy, 2021, Afghan Writer).

Naming: This function occurs when the citation appears within a noun phrase, essentially serving as a key element within the grammatical structure of the sentence. It highlights the authors' contribution by incorporating their name into the discussion, which emphasizes their work and fosters a connection between their study and the current argument. Naming citations often reflect the significance of the referenced work in relation to the topic being addressed.

• Early studies of second language writing were mainly inspired by Flower and Hayes (1981) influential cognitive theory of writing. (Hosseinpur, & Kazemi, 2022, Iranian writer)

• In Walsh’s (2010, 2019) view, a critical intercultural approach identifies relations of sociopolitical power and dominance that prevent egalitarian intercultural relations. (Ortega & Oxford, 2023, English writer)

• It is inconsistent with the findings of the study by Vanha (2017) and Guerrettaz and Johnston (2013) whose participants stated that the textbooks had a restricting role. (Orfan, Noori, & Akramy, 2021, Afghan Writer).

Non-citation: Non-citation references occur when an author's name is mentioned without including publication details such as the year. This is often used when the work was previously cited earlier in the text or when the focus is primarily on the content rather than the specific citation details. This approach helps maintain the flow of the narrative while acknowledging prior scholarship, allowing the writer to discuss ideas without redundancy.

• According to Kavaliauskiene, L1 use causes students to believe that there are L1 equivalents for second language (L2) structures and vocabulary. (Orfan, 2020, Afghan Writer).

• Mu’s taxonomy of second language writing strategies enjoys more important features that distinguishes it from other available taxonomies and enables it to contribute to both practical and theoretical investigation of second language writing strategies. (Hosseinpur, & Kazemi, 2022, Iranian writer)

• Policy discourse, then, suggests that the Scottish Government differs from the UK Government by taking an emancipatory, inclusive approach towards ESOL and, more broadly, the immigration settlement process (Brown, S. 2021, English writer).

To further evaluate the differences in citation practices among these groups, Chi-Square analysis was conducted. The results indicated significant disparities in the utilization of integral, verb-controlling, naming, and non-citation methods among the three writing groups, with the following Chi-Square values: X² = 51.87 for integral citations, 102.39 for verb-controlling citations, 7.02 for naming citations, and 8.31 for non-citation (p < .05). These findings reveal not only the varying preferences of citation practices across different cultural contexts but also highlight the significance of understanding these differences in academic writing. Table 2 provides an overview of the frequency of non-integral citations across three distinct groups of writers: English, Iranian, and Afghan.

Table 2: Frequency of Non-integral Citation by English, Iranian, and Afghan Writers

As Table 2 shows, English authors incorporated a total of 1,188 non-integral citations within their articles. Of these citations, 1012 were classified as source, 85 as identification, 38 as reference, and 53 as origin. In contrast, Iranian writers produced a greater volume of non-integral citations, totaling 1,290. This corpus contained 1123 citations categorized as source, 74 as identification, 21 as reference, and 72 as origin. Afghan writers, while generating a comparatively lower total of 623 non-integral citations, demonstrated particular tendencies in their citation practices, with 554 instances of source, 52 instances of identification, 3 instances of reference, and 14 instances of origin.

Examples of Non-integral Citation

Source: This type of citation refers to a specific theory, concept, or framework introduced or elaborated upon by specific authors. It serves to ground the discussion in established research and provides a basis for the claims being made. The authors are mentioned in a way that highlights their contribution but does not inherently form the grammatical subject of the sentence. Instead, the focus remains on the theory or concept itself.

• Specifically, the resources and appropriation theory of digital divides (van Dijk, 2020) posits that resources (material, temporal, mental, social, and cultural) that are useful for accessing technology are influenced by personal characteristics. (Cox, Chen, & Okatch, 2023, English writer)

• Vocabulary makes up the building block of language, and specific class time should be allocated to teaching vocabulary to EFL learners (Brown, 2000). (Orfan, 2020, Afghan writer)

• L2 enjoyment has been found to be associated with ideal L2 selves (Teimouri, 2017), the eager use of the target language, and L2 achievement (e.g., Papi & Khajavy, 2021). (Tahmouresi, & Papi, 2021, Iranian writer)

Identification: This function indicates a trend or focus within a broader area of research by naming studies conducted on a particular population or topic. It acknowledges the work of other scholars without explicitly attributing a claim to them as the primary subject of the sentence. Identification citations point to a body of research, emphasizing the importance of previous findings and situating the current study within that wider context.

• Concurrent with the flourishing of positive psychology in the arena of foreign language education, researchers attempted to identify the role of positive emotions in English as a foreign language (EFL) learners’ psychological growth and linguistic performance (Budzi´nska & Majchrzak, 2021; Derakhshan, 2022; Wang et al., 2021). (Derakhshan, & Noughabi, 2024, Iranian writer)

• To date, research looking into teachers’ professional life phases has focused largely on pre-service (e.g., Birchinall, et al., 2019; Hong, 2010; Vaisanen et al., 2018) and early-career teachers (e.g., Clandinin et al., 2015; Peters & Pearce, 2012; Schaefer, 2013). (Babic, Mairitsch, Mercer, Sulis, Jin, King, Lanvers, & Shin, 2022, English writer)

• Many of these studies have shown that gender bias is widely prevalent in textbooks (e.g., Zakka and Zanzali 2015; Wu and Liu 2015; Gharbavi and Mousavi 2012; Bahman and Rahimi 2010; Hamdan 2010; Stockdale 2006). (Orfan, 2023, Afghan writer)

Reference: Reference citations typically provide a means to direct readers to additional sources for further information regarding a specific aspect of the discussion. This type of citation acts as a pointer to more detailed discussions or findings elsewhere, encouraging readers to delve deeper into the existing literature. The authors are mentioned in a way that suggests a recommendation for supplementary reading rather than asserting their work in support of the current argument.

• More in-depth research on the reasons why AR can initiate development may be particularly useful in contributing to the study of language teacher agency (see Miller and Gkonou 2018) (Edwards, 2020, English writer)

• The task-based analysis led to the formulation of analytic syllabi through non-linguistic units of analysis, reflecting more dynamic rather than static qualities of target situation discourse (see Johnson, 2009) (Nateghian, 2023, Iranian writer)

• Considering the multilingual nature of language learning in Afghanistan, the low competency of teachers and learners in English, and the ideologically laden burden of sociocultural bearings on language learning of Afghan teachers and learners (for a complete discussion, see Coleman, 2019; Khawary & Ali, 2015). (Nazari, Miri, & Golzar, 2021, Afghan writer)

Origin: This function cites authors to indicate the foundational theories or concepts that inform the current research definitions or frameworks. It highlights the theoretical or contextual influences that underpin the author's approach to a certain issue or topic. In this way, origin citations serve to connect the current study to the lineage of ideas and research that shaped its development, illustrating the academic conversation that precedes it.

• Developmental Model of Intercultural Sensitivity (DMIS) (Bennett, 1986) is one of the classic models of intercultural competence. (Gholami Pasand, Amerian, Dowlatabadi, & Mohammadi, 2021, Iranian writer)

• Our definition was guided by theoretical (e.g., Miller, 2009; Pennington & Richards, 2016; Yazan, 2018) and contextual (e.g., Khawary & Ali, 2015) stances in addressing challenges that these three aspects could pose to Afghan L2 teachers. (Nazari, Miri, & Golzar, 2021, Afghan writer)

• When it comes to L2 learning and teaching, task complexity has been framed within the Limited Attentional Capacity Model (Skehan, 1998, 2003) and the Cognition Hypothesis (Robinson, 2001, 2003, 2005). (Zhang & Zhang, 2023, English writer)

The statistical analysis employing Chi-Square tests revealed significant differences in the citation practices of these three groups regarding non integral citations, specifically in the categories of source, identification, reference, and origin. The Chi-Square values obtained from this analysis were as follows: X² = 24.76 for non-integral citations, 202.99 for source citations, 8.02 identification citations, 29.64 for reference citations, and 37.74 for origin citations, with all results demonstrating significance at p < .05. These findings underscore the distinct citation strategies employed by writers from different cultural and linguistic backgrounds. The greater reliance on identification citations among Iranian authors, as well as the varied use of source, reference, and origin citations across the groups, illustrates the complexity and nuance of academic citation practices. This analysis not only highlights the behavioral variations in citation use among different writing communities but also emphasizes the importance of contextual awareness when engaging in academic writing and referencing.

DISCUSSION

Our investigation revealed distinct patterns in citation usage among English, Iranian, and Afghan writers. Notably, both English and Iranian authors showed a marked preference for non-integral citations, a trend corroborated by earlier corpus studies, such as those conducted by Samraj (2008) and Swales (2014), who explored thesis work by master's and undergraduate students, respectively. Conversely, Afghan authors exhibited a pronounced tendency towards integral citations, favoring them over non-integral alternatives. The decision to employ either integral or non-integral citations is influenced by several intricate variables, including linguistic background and an understanding of the distinct functions served by each citation type, which may contribute to the differing levels of engagement with these citation forms. From a sociocultural view (Vygotsky, 1978), these contrasting practices could be viewed as different kinds of apprenticeship and socialization into academic discourse communities. Integral citations by the Afghan writers may relate to their educational background that values the master-apprentice model of knowledge transmission through authoritative figures. In addition, such citations foreground authors as knowledge agents in a manner that shows respect and is linked to a more hierarchical understanding of scholarly authority. English writers appear adept at navigating the conventions established by their target academic communities, often generalizing concepts that necessitate a balanced application of both integral and non-integral citation strategies. Consequently, international authors tend to create a discursive framework involving both forms of citations, facilitating publication opportunities.

The preference for non-integral citations in Iranian and English authors is a natural inclination toward emphasizing the conceptual elements of arguments over the individuals contributing to them. This is consistent with the objectivity and impersonality often associated with scientific writing (Hyland, 1999; Thompson, 2005). This practice aligns with a sociocultural environment that valorizes the depersonalized presentation of knowledge, where the focus is on the truth of the claim rather than the authority of the individual making it. This is indicative of a cultural model of science reinforced through participation in a global, predominantly Anglophone, academic discourse. Additionally, non-integral citations contribute to a more coherent argumentative flow, avoiding interruptions in the narrative caused by frequent authorial attributions (Hewings et al., 2010). The inclination of Afghan writers toward integral citations indicates an emphasis on authorship, potentially mirroring cultural values that prioritize acknowledgment of scholarly contributions. This pattern resonates with findings from Khamkhien (2025), who noted that L2 writers often use citations to demonstrate topic understanding and acknowledge sources, which can manifest as a more author-centric, integral approach.

Further analysis uncovered that verb controlling citations constituted the most prevalent form of integral citation among both Iranian and Afghan writers, corroborating findings by Jalilifar (2012) and Shooshtari et al., (2017), who noted a considerable emphasis on verb-controlling citations in the writing of international research article authors. In contrast, English writers predominantly utilized naming citations, with noteworthy similarity in the frequency of integral-verb controlling citations and naming citations within English language articles. On the other hand, Iranian and Afghan writings demonstrated an evident preference for verb-controlling citations, which were employed almost twice as frequently as naming citations. This inclination underscores a desire to highlight authorship through verb-controlling citations, whereas naming citations, while integrated within the narrative, do not exert the same control over the verb (Jalilifar & Dabbi, 2012; Swales, 1990), potentially limiting their effectiveness in articulating authors' positions. This distinction can be seen as a difference in the cultural tools (Vygotsky, 1978) employed. The verb-controlling citation is a tool that explicitly stages the cited author's action, making the act of research highly visible. The naming citation, conversely, is a tool for incorporating the author's work as a conceptual object into the writer's own argument, a more subtle form of engagement that is privileged in many Western academic contexts. Iranian and Afghan writers frequently exhibit a reporting style that prioritizes factual presentation through the use of factive or non-factive verbs, leading to a diminished focus on the rhetorical and discourse aspects of citations. This observation aligns with Thompson and Ye’s (1991) assertion that an exclusive concentration on surface-level information may result in a failure to grasp the underlying intent.

These disparities may arise from challenges faced by non-native writers in constructing nominalizations and complex noun phrases, competencies that are frequently less developed in these groups. This observation is consistent with findings by Tambul ElMalik and Nesi (2008) and Esfandiari and Saleh (2024), indicating that nominalization occurs less frequently among non-native writers relative to their native peers. Ahn and Oh (2024) found that students, as developing writers, preferred integral citations, while expert writers favored non-integral forms. This suggests that Afghan writers' preference may indicate a stage in learning academic writing conventions. Sociocultural theory views this not only as a developmental delay but as a different path of socialization. The challenge is not just grammatical; it includes understanding a new set of cultural norms for using language to build credibility and find a place within a specific community of practice (Lave & Wenger, 1991). The adept use of both naming and verb-controlling citations by native writers reflects their comprehensive understanding of the varying functions entailed in citation strategies, allowing them to convey substantial information concisely (Petric, 2007). Moreover, our investigation established that non-citation constituted the least utilized citation type across all groups. The Chi-Square analysis delineated significant differences in the employment patterns of integral, verb-controlling, naming, and non-citation methods among the three writer groups.

In terms of non-integral citations, source emerged as the predominant function used by all three groups, recognized as the default non-integral citation type (Petric, 2007), while reference was the least utilized. The results indicated that both reference and origin citations were notably infrequent, aligning with the findings of Esfandiari and Saleh (2024) and Ebadi et al. (2021). This observation is justified by Thompson's (2005) findings that categorizations related to origin are typically reserved for methodology sections to describe materials and methods, rather than being employed in introductory contexts. The significant differences highlighted through Chi-Square analysis regarding the usage of non-integral citation types-including source, identification, reference, and origin-further elucidate the distinctive citation strategies employed by English, Iranian, and Afghan writers. Overall, this study provides invaluable insights into the citation practices of varied writing communities, emphasizing the impact of linguistic background on academic writing. Understanding these differences is essential for enhancing effective communication within scholarly discourse and refining the writing competencies of non-native authors in an increasingly interconnected global academic landscape.

CONCLUSION AND IMPLICATIONS

It is evident that referencing previous findings transcends a mere stylistic option; it is an essential requirement for substantiating arguments within scholarly articles, thereby bolstering the writer's credibility and the robustness of their research. Our inquiry revealed that citation practices are notably influenced by the writer's native language. The array of citation methodologies employed by diverse linguistic groups, as evidenced by the differing inclinations towards integral and non-integral citations, underscores the intricacy and subtleties inherent in these practices. Viewed through a sociocultural lens, these practices are not arbitrary but are deeply embedded in the specific academic and cultural contexts in which the writers have been socialized.

The propensity for integral citation among Afghan writers highlights their comfort with formal citation features—such as placing an author in a subject position to afford the author an explicit grammatical role—while concomitantly revealing a lack of awareness regarding the functional aspects of citation. This lack of awareness indicates insufficient familiarity with the operational dimensions of citation, which typically do not receive explicit instruction. In contrast, Iranian and Afghan writers frequently utilize verb-controlling citations, whereas their English counterparts often demonstrate adeptness in deploying naming and verb-controlling citations. This complexity poses significant challenges for non-native writers, who may encounter difficulty in mastering these conventions and seamlessly integrating citations within their discourse.

Writers proficient in English are more likely to exhibit an extensive knowledge of citation types in alignment with the standards set forth by their target discourse community. They typically generalize their concepts, necessitating a utilization of diverse citation conventions—both integral and non-integral—to optimally leverage the advantages of each category. Consequently, international writers establish a discursive framework utilizing both citation styles to create an avenue for publication. Conversely, Iranian and Afghan writers often fail to critique the reported text but instead reproduce it using factive or non-factive verbs, resulting in an oversight of the rhetorical and discourse aspects of citations. This conclusion aligns with Thompson and Ye’s (1991) assertion that focusing solely on the provided information may frequently lead to misunderstanding or misinterpreting the intended purpose.

As indicated in the literature, it is anticipated that non-native writers will cultivate these essential competencies through engagement with academic texts and participation in writing courses (Fazilatfar et al., 2018; Li et al., 2023). However, the reality oftentimes does not meet these expectations, revealing a necessity for more comprehensive instructional methods and practical opportunities. A sociocultural approach to pedagogy suggests that this instruction must go beyond mechanistic exercises. It should involve guided participation and scaffolding that makes the implicit cultural norms of the target discourse community explicit. In light of the intricate nature of citation practices and their pivotal role in academic discourse, educators in academic writing should enhance students’ citation proficiencies by implementing effective strategies that promote the development of their academic reading skills. To develop this skill, instructors can use teaching tasks like citation analysis. In these tasks, students break down texts to identify different citation forms and assess their rhetorical purposes. Additionally, citation transformation exercises can be very effective for building flexibility. These exercises ask students to change citations into different grammatical structures, helping them see how each choice changes emphasis and fits into the narrative.

The literature suggests a strong advocacy for the creation of targeted pedagogical approaches and practical strategies aimed at improving the citation capabilities of emerging scholars (Friedman, 2019; Jalilifar, 2012; Lee et al., 2018). Such strategies may encompass modeling citation practices, providing specific feedback on citation usage, and elucidating the rhetorical purposes underlying citations—such as establishing authority, contextualizing one's research within existing scholarship, and demonstrating engagement with the scholarly community. To deepen rhetorical awareness, instructors can employ a structured peer-review checklist focused exclusively on citation usage. This tool moves beyond checking formatting conventions to evaluate strategic choices, guiding reviewers to analyze, for instance, the diversity of citation types used and the potential for revising citations to improve grammatical integration and argumentative clarity. Furthermore, educational institutions and writing programs should contemplate the integration of explicit instruction on citation practices within their curricula. This should include workshops dedicated to the practical application of various citation types, the analysis of pertinent examples, and discussions regarding the ethical dimensions surrounding citation practices.

The workshops may involve active learning such as collaborative writing tasks when small groups are given a research finding and they have to collaboratively write a short paragraph of a literature review, synthesizing their information in the form of at least two different citations and justifying their reasoning for their choices. Another productive exercise is citation mapping when students literally map out the citations in the introduction section to see how authors are creating a research space when looking at the citations used to establish territory, establish a gap, or occupy a niche. Ultimately, the goal is to apprentice writers into the new discourse community by providing them with the tools and meta-awareness to navigate its unspoken rules. By equipping non-native writers with a comprehensive understanding of citations as both a scholarly obligation and a fundamental component of academic communication, we can enhance their capacity to engage meaningfully with the discourse community. As academic environments continue to evolve, the imperative to facilitate the mastery of these practices among emerging scholars becomes increasingly vital. Prioritizing thorough instruction in citation practices will undoubtedly improve the efficacy of academic writing and contribute to the ongoing evolution of a more inclusive and proficient scholarly community.

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID