Unfocused Written Corrective Feedback for Academic Discourse: The Sociomaterial Potential for Writing Development and Socialization in Higher Education

Background. There is a prevailing belief that unfocused written corrective feedback may not be suitable to promote students’ academic writing development. Purpose. This perspective piece demonstrates how unfocused written corrective feedback reflects the principles of sociomateriality, which views learning as dynamic. Perspectives. Unfocused written corrective feedback has the potential to support university students’ academic discourse socialization. This perspective is based on the observation that actual written corrective feedback in a classroom setting is varied and contextual, and not focused on any particular grammar form or writing feature. Conclusion. Unfocused written corrective feedback represents an optimal approach to support university students’ awareness and engagement with variables found in their learning ecology. These variables can support students’ academic writing development.


INTRODUCTION
Unfocused written corrective feedback (WCF) in the context of academic writing development of students who learned English as a second language is believed to be less effective, especially when compared to its counterpart -focused WCF (see meta-analysis by Kang & Han, 2015). The general opinion of its lack of efficacy mainly stems from studies that report positive outcomes from employing focused WCF, in terms of the improved or accurate use of particular grammatical forms or writing features among university students. For instance, Ellis et al. (2008) reported improvements in Japanese university students' use of definite articles; Frear and Chiu (2015), who examined the effects of WCF feedback on Taiwanese university students' use of the past form (verb), also reported delayed post-test improvements; and in a more recent study, Reynolds and Kao (2021) found that the effects of focused WCF with other forms of intervention had a positive impact on Taiwanese university students in their use of English articles in academic writing.
Nonetheless, recent synthesis studies and critical voices have pointed out that pedagogical research in English as a second language environments, including studies on WCF, are typically conceived as a 'laboratory', where pedagogical interventions are deliberately planned and language forms or writing features for which feedback is given are pre-selected. These studies often take on a quasi-experimental or experimental setup, where distinct variables are examined, leading to rather contextualized results, which, to a large extent, may only be applicable to the setting of these studies (see discussion by Mao & Lee, 2020;Kang & Han, 2015). These studies are problematic not only because they present a narrow ac-| Opinion Articles count of how feedback can be delivered, but they also fail to account for the role that English teachers or writing instructors play, as well as other sociomaterial entities that may shape students' writing development (see Mao & Lee, 2020;McKinley, 2019;Rose, 2019).
To advocate for an alternate (and more realistic) view of WCF, this paper contends that unfocused WCF provides a greater learning potential, especially for university students. To this end, this paper discusses unfocused WCF as a sociomaterial prospect that supports academic discourse socialization. Viewing unfocused WCF as having sociomaterial potential positions students as having the agency to interact with social and material entities encountered in the writing process (Nieminen et al., 2021). It also recognizes that students' feedback interactions are not simply confined to their teachers; rather, students are viewed as social actors with the capacity to disrupt or even reject educational notions preferred or promoted by their teachers or the institutions (Zukas & Malcolm, 2019).

OVERVIEW: FOCUSED AND UNFOCUSED FEEDBACK
Before examining how unfocused WCF offers a sociomaterial prospect, this section will first present a brief overview of focused and unfocused WCF, drawn from research published over the last two decades. As stated earlier, focused WCF has been regarded more positively, as it has been shown to reduce the number of errors in the use of targeted language forms. Focused WCF typically aims to address only one or two error types that may be pre-selected by the writing instructor or researcher (Lee, 2020;Lee, Luo, & Mak, 2021). Some errors focused on by studies were indefinite article use (Ellis et al., 2008) and the past forms of verbs (Frear & Chiu, 2015). One of the primary reasons for using focused WCF is the belief that it helps students notice issues found in their writing (Rahimi, 2019). This belief may be a crucial contextual variable for studies on focused WCF, given that noticing may only be possible with older students, or those studying a particular program or with high English proficiency. For example, Frear and Chiu's (2015) study had participants who were English majors; Sheen, Wright, and Moldawa's (2009) study, on the other hand, was conducted with graduate students in the USA. There are some studies on focused WCF set in other school level settings, such as that by Lee, Luo, and Mak (2021); nonetheless, their participants were reported to have strong academic abilities and English competency. From these studies, it may be the case that focused WCF was effective given the nature of the participants' study program (e.g., English majors or graduate degrees) and their ability to monitor their own performance (e.g., graduate students), or having a high English proficiency. Focused WCF has also been purported as having a more lasting impact (e.g., Sheen, Wright, & Moldawa, 2009;Frear & Chiu, 2015); nonetheless, at least one study reported that its effect was not statistically different than that of unfocused WCF (Ellis et al., 2008).
On the other hand, unfocused WCF aims to highlight an array of errors or issues. Unfocused WCF is also referred to as comprehensive feedback (Lee, 2020;Rahimi, 2019). For instance, in the study by Sheen, Wright, and Moldawa (2009), unfocused WCF given to several language forms (articles, copula 'be', regular past tense, irregular past tense and preposition) were found to have no significant impact even in subsequent revisions. Because unfocused WCF covers an array of error types, it is assumed that this WCF approach may hamper students' capacity to notice and to correctly revise errors (Frear & Chiu, 2015). Furthermore, it has been claimed that dealing with various errors does not create a supportive environment for writing development (Sheen, Wright, & Moldawa, 2009). In settings where unfocused WCF did show an extent of significant effect, it was provided with other feedback or relevant tasks, such as that demonstrated by Brudermann, Grosbois, and Sarré (2021), where online unfocused indirect WCF combined with micro-tasks was given to written tasks in an online EFL course taken by first-year STEM university students. This was also reported by Nicolás-Conesa, Manchon, and Cerezo (2019), where unfocused WCF led to greater accuracy in subsequent revisions when used alongside languaging; that is, the identification and explanation or discussion of an error with peers or with the writing instructor. Moreover, Nicolás-Conesa, Manchon, and Cerezo (2019) found that unfocused direct WCF had a greater uptake and retention when compared to indirect WCF. It should be noted, however, that their study comprised participants who were studying English as a major, similar to the study of Frear and Chiu (2015).
From these studies, it becomes apparent that both forms of WCF yield different, and at times, contradictory results, mainly due to varying contextual factors, including the profile of the participants and the setting of the studies (Mao & Lee, 2020). The ambiguity of the efficacy of distinct WCF types is actually the constant of what is known in literature. In the case of unfocused WCF, the question whether it can be a catalyst to maintain or improve overall writing accuracy in subsequent writing tasks also remain elusive (Frear & Chiu, 2015). The use of unfocused WCF, however, is more likely to be a typified approach for feedback provision in different language learning settings. Lee (2020) states that, "[i]n real-world contexts, teachers are likely to vary WCF strategies based on error types and students' abilities and needs, whether they respond to errors comprehensively or selectively." (p. 2). In other words, real WCF, whether used in an English for academic purposes (EAP) writing class or in the supervision of a research paper being written, may be decided based upon the nature of the writing task, or the assessment tool used for the writing tasks, or the students' language/writing capabilities. Hence, paying attention to | Opinion Articles only one error type, such as giving focused WCF, may not be a truthful reflection of classroom practice.

WRITING AS ACADEMIC DISCOURSE SOCIALIZATION IN HIGHER EDUCATION
Situating feedback as a part of academic discourse socialization is integral as it illustrates the processes of writing at different university levels. At these levels -undergraduate, master's, and doctoral -various studies that take a broader approach in assessing students' writing development show students' engagement with feedback as a process for academic discourse socialization. According to Morita (2009), academic discourse socialization can be defined as "learning how to participate in a competent and appropriate manner in the discursive practices of a given academic community" (p. 444). Anderson (2017) further explains that the process of academic discourse socialization may be informed by what is observed in external forces, such as expectations of supervisors or the communication conventions of the immediate community. These external forces then inform internal socialization, which consists of a student's own self-regulation of the learning experiences. In Anderson's study, the participants -all Chinese PhD students at a university in Canada -expressed concern over the quality of their academic discourse, particularly their writing, which they said would be scrutinized by their PhD supervisors, other academics, and potential employers. This issue was similarly raised by Kobayashi, Zappa-Hollman, and Duff (2017), but from the perspective of instructors who oversee processes pertinent to students' academic discourse socialization. Specifically, how are these socialization processes evaluated, or even established and made known to the students? One aspect in students' academic discourse socialization is feedback provided to their writing. When feedback is viewed as a variable for academic discourse socialization, it diminishes the view that processes of academic writing are confined only in a particular context; these processes are actually shaped by other confounding variables found within the writing task, the course, and even the students' learning ecology and the wider university environment. The next section illustrates this latter point through research findings regarding academic writing expectations and feedback provision across different levels of university.

Undergraduate Level
For many undergraduate students, the transition into university may be rather unnerving. In particular, undergraduate students might find themselves in uncharted territories of university-level academic literacy expectations. Their writing instructors, on the other hand, may take it for granted that students are familiar with academic writing expectations and conventions (Elliott et al., 2019). This concern is also observed among subject-content instructors, who ex-pect students to know specific writing features found in their disciplinary areas (e.g., Dang, Bonar, & Yao, 2021). However, while these instructors may be experts of their subject-content, they may not have a clear understanding of the function of assignments or assessments, especially in terms of how these tasks might impact students' academic discourse socialization at a broader level (Kobayashi, Zappa-Hollman, & Duff, 2017). On the other hand, it may also be the case that university policies may overlook students' academic writing needs (see Pineteh, 2014). While there have been attempts to create an inclusive support system to address students' writing needs, universities may still find these efforts challenging, especially when there are students from different cultural backgrounds and English language experiences, with different academic writing needs for their university courses (e.g., Dimova, 2021). From these observations, it may be surmised that several factors, such as time, the feedback practices of the writing instructor, as well as the objectives of a writing assignment or course determine WCF.

Master's and Doctoral Levels
Some studies on academic writing at the master's level have reported a lack of understanding of WCF. For instance, Nguyen and Buckingham's (2019) study of Vietnamese master's students at the University of Auckland reported occasions where WCF was misunderstood. Feedback such as 'limited reading' was misread as not providing sufficient sources, instead of the students' lack of understanding in the sources cited, which was the intended meaning. This led to students citing sources that were recent and perceived as more prestigious. Furthermore, at the master's level, WCF may not necessarily come from instructors. At this stage, many graduate students find themselves commenting on the work of their peers. For some, it may be challenging to offer critical feedback to peers as it may not be culturally appropriate, or it may be that students have not been socialized into the practice of offering feedback to classmates. In such situations, students may offer basic WCF focused on language use. This was observed in Zhang, Yu, and Yuan's (2020) study, which found that master's students in a Chinese university who engaged in peer feedback focused mostly on language issues and hardly on content. The main reason for giving only language-based WCF was that these students did not think of them as the authority to offer comments, or constructive criticism, on their peers' work.
At the doctoral level, feedback is viewed as an integral process for knowledge validation. For example, Anderson (2021) reported that the doctoral students in his study did not view feedback as negative; instead, they viewed feedback positively and as a necessary element in being legitimized as (future) PhD holders. Anderson (2021) further demonstrates that at the doctoral level, feedback is offered through different modes and junctures as an effort to maximize students' academic discourse socialization. This is also possible due to the regular engagement that doctoral students have | Opinion Articles with their supervisors over the period of a few years. At this point, it may also be useful to consider WCF provided at the workplace. From the study of Yusuf, Yunus, and Embi (2018), as well as the report by Knoch et al. (2016), we can see how former students find themselves learning about writing conventions at the workplace, with feedback given by their superiors. This feedback generally concerns the alignment of content with its genre (e.g., information presented through an email should be direct and concise) and the ability to comply with established text templates. Aside from having utility for the immediate work task, feedback was also deemed necessary by the graduates as a means of validating their positions within a company and discerning their professional progress.

SOCIALIZATION AS A SOCIOMATERIAL PROSPECT
So far, this paper has shown how a myriad of reasons can shape the provision of WCF. Research findings, particularly those that investigate academic writing as a whole, do not view WCF as having the singular goal of error reduction; it has instead viewed feedback as serving the purpose of preparing students to participate in knowledge production or to fulfil assessment requirements expected at the university level. In fact, in a recent large-scale study done in China, university English majors reported that WCF was actually the least employed feedback. Feedback that these students received was expressive, in the form of suggestions on how their work can be improved, or a hedged evaluation if they did not do well (Yu, Jiang, & Zhou, 2020). These observations also point towards the relevance of viewing students' writing development as a form of academic discourse socialization, as it provides information about which aspects of the work had been well-written and what may require further improvement. From this perspective, developing discourse literacy in academic English, including writing, should be seen in light of various factors (e.g., Loo et al., 2018). This constitutes a sociomaterial approach in conceiving the process of learning as it takes a keen interest in 'everyday work practices'. Moreover, the sociomaterial approach recognizes how "work is assembled and reassembled and academic learning is enacted but also how they are interrupted, resisted and rejected" (Zukas & Malcolm, 2019, p. 274). With regards to feedback, this calls into question the utility of a linear or causal effect of feedback provision. As Gravett (2020) argues, "feedback literacy may, likewise, be more appropriately conceptualised as a complex breadth of dynamic, nuanced, situated feedback literacies, with the employment of the plural here indicating a wider understanding of the concept of feedback literacy than has traditionally been adopted to date." (p. 11). Hence, pedagogy that is oriented towards a sociomaterial perspective will never assume teaching and learning as definite or an event that can be prescribed (Fenwick & Landri, 2012;Gourlay, 2017). For WCF shaped by a sociomaterial per-spective, the goal is then to create opportunities for students to engage with various unpredictable social and material prospects from within the WCF event, along with the wider learning ecology (Guerrettaz, Engman, & Matsumoto, 2021).

CONCLUSION AND IMPLICATIONS: THE PROMISE OF UNFOCUSED WCF
Viewing academic writing as an academic discourse socialization process highlights the disconnect between studies that take a narrow view of WCF with studies that examine academic writing in broad terms. The disconnect is perhaps due partially to the positivist aim of identifying and justifying certain WCF that works and to the onus of a student's writing development being on the writing instructor (see discussion by Winstone & Carless, 2021). For researchers of WCF, it may also be due to the appeal of 'packaging' the provision of focused feedback, or a constrained form of WCF, as a neat classroom phenomenon for the ease of research investigation and subsequent publication (McKinley, 2019). Nonetheless, the perspective being argued for here is that WCF should not consist of just one type, as made evident earlier through the overview of studies on academic writing at different university levels. Thus, the main implication is that if the aim of WCF is for students to engage in a myriad of sense-making pursuits, then unfocused WCF would be suitable. Through unfocused WCF, students will need to work through language and writing concerns that affect their writing or academic discourse (see Anderson, 2017). Furthermore, the outcomes of the provision of unfocused WCF should not be viewed as positivist evidence to form feedback principles that are generalizable. Feedback outcomes should instead be seen as efforts that support learning from an ecological perspective (Lee, Luo, & Mak, 2021;Loo, 2020;). Yet, with a broader view of WCF, there will be a more truthful representation of what really goes on in the classroom. In other words, "they can also see that TESOL research is inherently messy, leading to gains in confidence to conduct their own research without the pressure of producing something methodologically 'perfect.'" (McKinley, 2019, p. 882). This will lend support to the authentic representation of WCF provision and avoid oversimplifying learning processes as linear transactions between instructors and students.