575 Studies in Second Language Learning and Teaching Department of English Studies, Faculty of Pedagogy and Fine Arts, Adam Mickiewicz University, Kalisz SSLLT 12 (4). 2022. 575-595 https://doi.org/10.14746/ssllt.2022.12.4.3 http://pressto.amu.edu.pl/index.php/ssllt Reexamining feedback on L2 digital writing Idoia Elola Texas Tech University, Lubbock, USA https://orcid.org/0000-0003-0722-0332 idoia.elola@ttu.edu Ana Oskoz University of Maryland, Baltimore County, USA https://orcid.org/0000-0003-2770-4844 aoskoz@umbc.edu Abstract The integration of digital multimodal composing (DMC) in the second language (L2) and heritage language (HL) classrooms has expanded our notion of writing, shifting from a focus on the written mode to include other modes of expression (e.g., visual, textual, or aural). Notwithstanding the increasing presence of L2 multimodal learning tasks, which combine different semiotic resources (e.g., language and visual compo- nents such as images or videos) as intrinsic elements used to generate meaning, in- structors have not yet modified the way in which they provide feedback. That is, de- spite the increasing integration of different modes in a multimodal task, instructors still focus exclusively on language development – replicating the feedback behaviors modeled by non-digital writing assignments – rather than on all the components of multimodal texts. In digitally influenced environments and societies, however, there is a need to reconsider our approaches to feedback to pay greater attention to the linguistic and nonlinguistic elements of DMC. With the scarcity of research on feed- back in DMC, this article first identifies a gap in multimodal teaching and research regarding the role and focus on feedback in DMC, and, second, provides an assess- ment rubric from which to base formative feedback that addresses both linguistic and nonlinguistic elements to help students develop their multimodal texts. Keywords: digital multimodal composing; digital feedback for multimodal texts; nonlinguistic feedback; assessment rubrics; literacies Idoia Elola, Ana Oskoz 576 1. Introduction The increasing inclusion of digital multimodal composing (DMC) in the second language (L2) and heritage language (HL) classroom, which expands traditional written forms to include other modes of expression (e.g., visual, textual, or au- ral), has brought new concerns to instructors and researchers alike. In recent years, L2 and HL learners have frequently explored and created multimodal texts, such as blogs and digital stories, in which the integration of semiotic resources (e.g., language and visual components, such as images or videos) are intrinsic elements used to generate meaning. Yet, the goal of creating synaesthetic texts (Kress, 2003), in which each component depends on the others for meaning cre- ation, is often ignored during feedback provision (Oskoz & Elola, 2020). Regard- less of the inclusion of DMC in upper-level (Fornara & Lomicka, 2019; Nelson, 2006; Mina, 2014; Oskoz & Elola, 2014, 2016b) or lower-level (Ruiz Pérez, 2022; see also Zhang et al., 2021) courses, the questions instructors are faced with and challenged by concern what the role and focus of feedback should be when cov- ering both linguistic and nonlinguistic features. In L2 learning environments, the provision of feedback, a familiar, well- established practice, generally concentrates on writing conventions (i.e., global issues, such as organization) and error correction (i.e., local issues, such as gram- mar) with the aim of learning to write or writing to learn (Manchón, 2011). Feed- back, that is, stating a problem (e.g., an error or structural issue) and then invit- ing learners to try to fix it, is a traditional technique used to foster linguistic de- velopment that offers learners the opportunity to notice differences between their interlanguage and the target language. Despite the increasing prevalence of multimodal tasks in the L2 classroom and in DMC research (Caws & Heift, 2016; Oskoz & Elola, 2020), the emphasis of feedback provision continues to focus on the linguistic aspects of multimodal texts and often ignores nonlinguis- tic elements. This almost sole linguistic focus on feedback in multimodal tasks and genres may be the result of L2 instructors’ lack of guidelines to assess mul- timodal texts, insecurity with the use and affordances of the digital tools, avoid- ance of nonlinguistic elements that seem to be learners’ personal choices, or lack of curricular objectives that include the reason for using multimodal texts (Oskoz & Elola, 2020). However, echoing Caws and Heift (2016), the roles of in- structors and peers as (digital) feedback providers in a digital context should be “radically changing while, at the same time, becoming more critical,” with the goal of helping learners become “progressively digital[ly] literate” (p. 133). Feed- back should “not only take into account the learner but, ideally, also the tool, the interactions as well as the outcomes” (Caws & Heift, 2016, p. 131). In line with these pressing changes, DMC demands a change in our feedback practices and Reexamining feedback on L2 digital writing 577 focus. Thus, adapting feedback practices for multimodal texts raises important questions: What type of feedback and on which elements of multimodal com- posing should we provide feedback to our learners in DMC environments, and why is this important in current educational settings? How do we provide feed- back on the interrelationships between language and the interconnection of modes that have traditionally not been considered part of the act of writing? In this article, we address the importance of the largely unexplored role of feedback in the L2 DMC domain. First, by examining new definitions of liter- acies, theoretical and pedagogical frameworks, and digital genres, we argue that there is a need to maintain the traditional focuses of feedback (e.g., language, content, structure, and organization) but also to move towards a more multifac- eted, comprehensive, and modern feedback method (e.g., multimodal issues related to identity, agency, and voice). Second, following Lamb’s (2018) emphasis on formative feedback or “feedforward” (as he prefers to say), “where students are provided with correctional advice and guidance during the process of compo- sition, rather than [after] the submission of coursework” (p. 6), we connect form- ative feedback to assessment. Whether coming from the instructor or their peers, providing formative feedback allows learners to focus on linguistic and nonlinguis- tic issues throughout the development of a multimodal text. Third, considering the purpose of multimodal compositions and the need to focus on both linguistic (e.g., orthography, verb tenses) and nonlinguistic issues (e.g., images, sounds), we address the different components that ought to be considered when providing feedback on multimodal texts and suggest criteria that instructors can use when offering feedback on multimodal tasks. Finally, acknowledging the limited re- search on feedback in digital contexts, we conclude by offering ideas for potential future studies that examine the role and impact of feedback in the development of multimodal texts and the creation of multimodal tasks. 2. Factors that shape feedback: Literacies and frameworks 2.1. A new understanding of literacy There is no question, as Hafner and Ho (2020) point out, that even with rela- tively standard practices, like the writing of traditional argumentative and ex- pository essays, for example, there is an increasing proliferation of multimodal texts. These include, among others, video methods articles in the life sciences (Hafner, 2018), visual abstracts in academic articles (Sancho Guinda, 2015), crowd-funding proposals for educational projects (Mehlenbacher, 2017), schol- arly blogs engaging a wide audience (Luzón, 2013), and digital storytelling (Os- koz & Elola, 2014, 2016a, 2016b). These multimodal practices are the result of Idoia Elola, Ana Oskoz 578 expanded definitions of literacies. It makes sense that, because of the internet revolution and the spread of social digital tools, literacies today are considered “social practices that are fluid, sociocultural, multimodal, and dynamic” (Chen, 2013, p. 143), which are implemented “by individuals as parts of larger groups” (Reinhardt & Thorne, 2011, p. 259) and that consider the goals of writers and audiences as well as the social relationships between them (Hafner et al., 2013). The term digital literacy, which pertains to internet- and technology-mediated learning (Chen, 2013, p. 143), converges with other literacies linked to commu- nication skills; fits well within broader social, institutional, and historical do- mains; and may be described in the context of user practices – media, in partic- ular (Barton, 2007). As a result of the expanded view of literacies, the inclusion of multimodality in meaning-making responds to the technological advances of society (Matthewman et al., 2004). The call for including multiliteracies in our curricula is not new. As previous work has already pointed out, to effectively shift into current literacies, it is crucial for learners, instructors, and researchers to acknowledge the need to align in- struction and research with learners’ complex and multifaceted literacy skills (Car- penter, 2009), to assist learners with meeting the challenges of a global economy (Johnson & Kress, 2003), to equip learners with the creativity required in the work- place (Kimber & Wyatt-Smith, 2010), and to act in response to the current digital and multimodal nature of teaching (Jewitt, 2008). In terms of literacies instruction, Lamb (2018) identifies the close and important relationships that exist between multimodality, feedforward and assessment. In fact, if we want to understand the orchestration between the integration of diverse modes and semiotic resources (as a result of evolving tools and their affordances, the genres being produced, and the potential outcomes of multimodal tasks) as well as how multimodal texts will be assessed, the role of feedback becomes more relevant than ever. What is crucial, based on the expanded definition of digital literacies and the inclusion of DMC, is that the notion of what feedback needs to cover in this domain has changed and expanded to encompass the integration of different modes. 2.2. Feedback in connection to frameworks Feedback for DMC is directly or indirectly shaped by the theoretical or pedagog- ical perspectives in which multimodal tasks are framed. Multimodal texts in the L2 classroom have been investigated from different theoretical perspectives, such as social semiotics (Oskoz & Elola, 2016a, 2016b) and multiliteracies (Ruiz Pérez, 2022). Each perspective triggers and provides a unique analysis of the different components of DMC, impacting the manner in which feedback provi- sion is thought out and implemented. Reexamining feedback on L2 digital writing 579 Social semiotic theory, for instance, helps explain the affordances of dif- ferent modes (e.g., visual, textual, or aural) and how those modes separately or collectively contribute to multimodal ensembles such as a digital story or an in- fographic (Bezemer & Kress, 2008; Kress, 2003). Social semiotics investigates how writers (re)design their texts by arranging “available meaning-making re- sources into a multimodal whole, making authorial decisions appropriately for specific audiences and purposes” (Shin et al., 2020, p. 2). The relationship be- tween used modes and intended meanings is known as “synaesthetic semiosis:” the inclusion of oral, aural, and written modes, often combined, when compos- ing multimodal ensembles (Kress, 2003, 2010). This synaesthetic semiosis can occur in two forms during the construction of a multimodal text (Kress, 2003, 2009): first, there is transformation, the actions that reorder and reposition se- miotic resources within a particular mode, and, second, there is transduction, the reorganization of semiotic resources across modes. For instance, when de- veloping a digital story through transformation, learners reconstruct “the syntax or structural complexity of sentences from a narrative (written) story into a dig- ital story script,” and through transduction, learners convert “written narration into spoken language and incorporat[e] images, music, and sound” (Elola & Os- koz, 2017, p. 55). From this perspective, feedback primarily emphasizes the con- tent created within each mode, but it is also important to the interconnectivity between and across modes. From a pedagogical perspective, the concept of multiliteracies in educa- tion (New London Group, 1996) is built upon the intersection of a broader defi- nition of literacy, the evolution of technological tools, and diverse social, histor- ical, and cultural contexts. The New London Group (1996) based its framework on the concept of design (i.e., schematic knowledge, the process of designing a text) and the idea that the redesigned text is the contribution to the world by the meaning-maker, who is engaged in an active, dynamic, transformative pro- cess of meaning-making. Therefore, multiliteracies position the learner at the center of the learning process, able to accept the critical role of agency in it with the objective of “creating a kind of person, an active designer of meaning, with a sensibility open to differences, problem-solving, change, and innovation” (Kalantzis et al., 2016, p. 226). To accomplish these literacies, learners must con- sider the form and function of design (i.e., the different meanings and purposes of a text) and how these meanings and purposes are impacted and altered by the designer and the audience (Mills, 2006; Kalantzis et al., 2016). From this perspective, digital feedback underscores the importance of the meaning-mak- ing process associated with writing and the learner as an agent in the process, thus, also validating Kress’s (2003) notion of the learner as designer. Idoia Elola, Ana Oskoz 580 As L2 instructors and researchers, therefore, we see the need to expand our feedback repertoire beyond the focus on global (i.e., content, structure, or- ganization) and local (i.e., grammar, vocabulary, editing) aspects of the language. As a way to expand feedback, we could consider research grounded in social semiotics (Oskoz & Elola, 2016a, 2020) and the perspectives of multiliteracies (Ruiz Pérez, 2022) that have illustrated how L2 learners shift fluidly across modes (e.g., language and visual images) and have also taken into account the development of coherent multimodal designs. It is therefore essential to include an assessment framework based on curricular objectives that considers multi- modal texts by focusing on both linguistic features elicited, whether in the writ- ten or oral language modality, as well as in additional modes (e.g., visual), and the agency of the designer and the audience. 3. New genres, new tasks The inclusion of digital multimodal texts in the L2 classroom is, without doubt, the result of the increasing presence of digital genres in learners’ professional and personal lives. As Heyd (2016) reminds us, some digital genres are not to- tally new and have gone through a process of remediation in moving from print to a digital environment, resulting in genres that “can be aptly described as ‘hy- brids’ that incorporate both old and new aspects” (p. 95). These hybrid genres include entries in Wikipedia, for example. Other genres, however, are truly new or emergent. Tweeting, for instance, necessitates fully utilized technological ad- vances to create communicative environments that do not have an equivalent in face-to-face communication (Heyd, 2016). Therefore, the inclusion of digital genres and resulting multimodal texts challenges past feedback practices. What was considered appropriate just a few decades ago might not be enough when focusing on current digital multimodal genres; that is, digital genres, such as blogging and tweeting, call for modified conceptions of feedback. The development of a blog entry requires an under- standing of the rhetorical and linguistic characteristics of the digital genre, while the composition of a tweet assumes knowledge of particular linguistic conven- tions and semantic connections, including the value of the hashtag. In these cases, balancing the development of L2 learners’ writing skills, while addressing the evolving multimodal composing conventions through the provision of effective feedback, could be taxing for the instructor. It is for this reason that guiding in- structors and helping them develop a multimodal metalanguage can help them and their learners tackle multimodal texts that are still less familiar in L2 contexts. When providing feedback on multimodal texts, there is also a need to un- derstand that not all of them present the same degree of multimodality. Following Reexamining feedback on L2 digital writing 581 Lim and Polio (2020), we can distinguish between a strong version of multimodal composing – for example, when “linguistic and nonlinguistic modes of expres- sion contribute equally to building communication” (p. 2) – and a weak version, “in which nonlinguistic modes serve supporting roles for language development” (p. 2). Digital stories and infographics, for instance, could be considered exam- ples of strong versions since visual, aural, or oral elements (among others) are needed to create meaning, whereas Wikipedia entries or blog posts, in which the images might support the written text, could be considered weak versions of multimodal texts. Rather than thinking of strong versions as better and weak versions as worse, instructors need to keep in mind that, when employed in the academic context, the integration of different semiotic resources in multimodal texts varies based on their instructional goal (Lim & Polio, 2020). Therefore, the feedback provided needs to consider the degree to which linguistic and nonlin- guistic resources work together in the development of multimodal texts. 4. Feedback interplay with assessment The aim of digital feedback provision is to help learners become efficient com- posers in a variety of DMC contexts. Thus, depending on the purpose of the dig- ital multimodal text, feedback might also touch upon diverse components of the composition. These components include the goals of the task and the expected outcomes while considering the audience’s expectations, the learner’s choice of semiotic resources, as well as authorship – the individual and/or collaborative intellectual contributions, and ownership – the fair use and acknowledgement of external sources (e.g., ideas, images, and other semiotic resources). It is be- cause of the inherent social nature of the texts created in L2 DMC classrooms that learners interact with each other, the instructor, and the wider community, providing a natural and conducive space for the use of feedback through, for example, the use of comments in a blog post, the feedback provided by other authors while developing a Wikipedia entry, or likes and dislikes in a digital story uploaded to YouTube. For a better appreciation of the use of digital feedback, we need to under- stand that a digital multimodal dialogue that makes explicit the connection be- tween technology, multimodality, feedback, and assessment is essential in DMC (Lamb, 2018). In the case of assessment, Hafner and Ho (2020) suggest that its design needs to run parallel with the tools used (e.g., blogs, infographics soft- ware) as well as with formative and summative rubrics. Moreover, they suggest that feedback on DMC requires an understanding of the goals of the task through assessment processes. Similarly, Kalantzis and Cope (2008) propose a model for assessment that includes identity and social cognition (the understanding that Idoia Elola, Ana Oskoz 582 knowledge is the result of social interaction), provides formative assessment, fosters learning, and encompasses multimodal texts. There seems to be a grow- ing consensus that feedback can and should be close to assessment procedures as learners create their multimodal texts through the scaffolding of instructors or peers. That is, feedback provided at various times (i.e., formative feedback) in the development process may draw attention to learners’ semiotic selections (e.g., linguistic, musical, pictorial) and thus help them heighten semiotic aware- ness (Towndrow et al., 2013). To move beyond the sole emphasis on linguistic components and focus on the complete integration of modes and corresponding semiotic resources, Yi et al. (2017) call for assessment procedures that are tailored to L2 writers’ needs to master language in combination with other modes. Hung et al. (2013) further note that a rubric that focuses on the use of different modes may help learners understand the way in which the combination of these modes in the multimodal ensemble support meaning-making. The assessment criteria, as Hicks (2015) suggests, must be flexible enough to reflect the evolving nature of communica- tion tools and technologies, which may initially call for generic criteria that can be adjusted to a range of different types of multimodal composition. As instructors become more at ease with newer digital tools and genres, and begin to understand the impact that both linguistic and nonlinguistic ele- ments have on the L2 composition process, they need to move away from feed- back comments targeted purely at academic written genres and begin to use feedback aligned with semiotic resources or multimedia, to comment on, for example, the development of a digital story. Following Kress’s (2003) notion of synesthesia with digital storytelling, instructors may provide feedback on whether the chosen images correspond to the topic of the digital story, whether there are too many literal (rather than implicit) images, whether the sound/music en- hances or obstructs the story, whether the music volume overpowers the voice of the narrator, and other aspects related to a digital multimodal text (Jiang et al., 2022; Oskoz & Elola, 2016a, 2016b; Yang, 2012). Using an assessment rubric (see Table 1) as a starting point for feedback provision can guide learners through consequent revisions (see Maqueda, 2020; for rubrics on composition as well as complexity and fluidity of semiotic re- sources; Jiang et al.’s (2022) rubric for a genre-based model). These types of ru- brics help instructors and learners reflect critically and equip them with the met- alanguage of multimodality, as well as with an understanding of the affordances of the modes and the media to produce different (digital) genres and texts. For in- stance, Hafner and Ho (2020) point out that although instructors seem to under- stand intermodal relations when assessing multimodal compositions, they do not explicitly acknowledge the notion of semiotic harmony, that is, the orchestration Reexamining feedback on L2 digital writing 583 (or best fit) of the selection of resources from different modes that indicates that the form has the requisite features to be the carrier of the meaning (Hafner & Ho, 2020) when developing a multimodal ensemble. Developing and explain- ing a rubric that considers traditional linguistics issues (e.g., grammar, vocabu- lary) but also audience expectations, choice of semiotic resources, authorship, and ownership helps L2 composers make informed selections of multimodal se- miotic resources to support the overall meaning required for a specific genre and medium. The rubric will also help with the appropriate design of feedback. Table 1 Digital story rubric (summative and formative assessment) Category 4 Points* 3 Points 2 Points 1 Point Audience Engagement The pace fits the storyline and engages the audience. The pace is occasionally too fast or too slow. An attempt is made at pacing, but the audience is not fully engaged. No attempt at pacing is made, and the audience is lost. Semiotic resources Oral narration Narration is clear and well ed- ited. Narration is fairly clear. Narration is often hard to follow. Narration is missing. Music/sound effects (aural) Soundtrack complements and does not overwhelm nar- ration. Soundtrack often over- whelms the narration. Soundtrack is distracting. There is no soundtrack. Visuals All images are clear (and/or origi- nal), and there is a good mix of lit- eral and symbolic imagery. A few images are unclear, and few of them are used symbolically. Many images are unclear, and there is no symbolism. All images are unclear, lit- eral, and/or inappropriate. Grammar Wide range of L2 grammati- cal structures with few or mi- nor errors. Adequate range of L2 gram- matical structures; overuse of simple constructions; several minor errors. Limited range of L2 struc- tures; poor control of grammar; frequent errors. Frequent, persistent L2 grammatical errors; text is difficult to understand. Vocabulary Makes full use of the L2 vo- cabulary about the topic pre- sented. L2 vocabulary accurate but somewhat limited. L2 vocabulary limited, with overuse of imprecise and vague terms. Very limited L2 vocabulary; overuse of imprecise and vague terms. Multilingual and translingual practices (if applicable) Makes excellent use of sev- eral linguistic repertoires in a coherent manner. Makes fair use of several linguistic repertoires in a coherent manner. Makes poor use of several linguistic repertoires. Makes confusing use of several linguistic reper- toires. Genre characteristics Storytelling structure Presents the rhetorical ques- tion to be answered. Either the rhetorical ques- tion or the answer is miss- ing. Rhetorical question and the answer are both miss- ing. There is no clear narrative in the story. Motion and transi- tions (organization) Used at least 4 motion ef- fects; transitions are effective. Used at least 3 motion ef- fects; transitions are mostly effective. Used at least 1 motion ef- fect; some transitions are distracting. Used no motion effects; used no transitions. Mechanics The title appears at the be- ginning and the final credits at the end. Part of the title and/or some of the final credits are missing. Either the title or the final credits are missing. Both the title and the final credits are missing. Authorship Collaboration (if applicable) Author communicated well and participated in a discus- sion of ideas that led to a jointly created product. Author communicated and participated in a dis- cussion of ideas that led to the created product. Communication with the authors was limited, but they contributed to some extent. Author either rarely com- municated or tried to im- pose ideas without listen- ing to others in the group. Ownership Credits and citations All original and non-original images and sounds are cred- ited and cited. Some citations and credits are missing. Many citations and credits are missing. There are no citations or credits. Voice and identity Choice of semiotic resources The choice of language, images, sounds, and tone carefully represent the author’s voice. The choice of language, images, sounds, and tone mostly represent the au- thor’s voice. The choice of language, images, sounds, and tone vaguely represent the au- thor’s voice. The choice of language, images, sounds, and tone do not represent the au- thor’s voice. Idoia Elola, Ana Oskoz 584 Overall digital story The digital story provides a complete account of the au- thor’s message. The digital story provides an adequate account of the author’s message. The digital story vaguely represents an account of the author’s message. The digital story provides a superficial account of the author’s message. Multimodal ensemble Transduction Combining the different modes in a harmonic orchestration, the story is told with exactly the right amount of detail and is not too long or too short. Although combining differ- ent modes, the story is sometimes vague or in- cludes unnecessary detail; it seems to drag sometimes. Despite combining modes, meaning is only duplicated (not enhanced). The story needs more editing and is no- ticeably too long or too short. The story needs extensive editing to make the most of using different modes. Transformation The author maximizes each mode’s potential to tell the story with images (e.g., size, color), sound (e.g., volume), and oral narrative (e.g., pauses, tone). The author generally uses each mode’s potential to tell the story with images (e.g., size, color), sound (e.g., volume), and oral nar- rative (e.g., pauses, tone). The author minimally uses each mode’s potential to tell the story with images (e.g., size, color), sound (e.g., volume), and oral nar- rative (e.g., pauses, tone). The author does not use each mode’s potential to tell the story with images (e.g., size, color), sound (e.g., volume), and oral nar- rative (e.g., pauses, tone). Task completion Multimodal task completion; following the prompt Learner has followed the prompt, and the task ad- dresses all multimodal re- quirements. Learner has followed the prompt, and the task ad- dresses some of the multi- modal requirements. Learner did not follow the prompt closely, and the task addresses few of the multimodal requirements. Learner did not follow the prompt, and the task does not address the multi- modal requirements. Note. *Points in the rubric are subject to change depending on the objectives and goals of the exercise Providing comments on learners’ multimodal work during the formative stages (i.e., formative feedback) can also result in prompt revisions. This effect, as Campbell and Feldman (2017) have observed, is much more suggestive of learning than assign- ing a grade or circling an achieved outcome on an assessment rubric (Lamb, 2018). Lamb’s (2018) emphasis on feedforward means that the instructor’s comments are timelier, enabling learners to take corrective action to improve the quality of the as- signment being submitted for a grade. The provision of feedforward can be more ef- fectively provided through the use of formative assessment exercises, in which learn- ers are able to gauge their progress and understanding in a low-stakes setting (Hounsell et al., 2007). If formative feedback is considered an evolving digital conver- sation between learners and instructors, as well as between learners and other learn- ers, it can promote regular opportunities for encouragement and discussion around DMC production and assessment. Furthermore, the social dimension of emergent technologies, such as wikis and blogs, presents greater opportunities for rich dialogue, facilitating students’ participation, and student-focused interventions that can sup- port ongoing feedback and formative assessment (Hatzipanagos & Warburton, 2009). The rubric in Table 1 (adapted from Oskoz & Elola, 2020), for instance, was de- signed to assess a digital story, a storyline that integrates text, images, and sounds in an online environment, and which is often conceived as a personal narrative (e.g., intercultural experiences when studying abroad; identity awareness as heritage speak- ers). Given that the digital story is a multimodal text, the rubric needs to cover linguis- tic issues, such as grammar, vocabulary, and rhetoric (e.g., content, structure, and organi- zation), as well as additional semiotic resources and the processes of transformation and transduction. Engagement with the audience, as well as learner voice and authorship, are also included. The criteria proposed are by no means exclusive or exhaustive of what can be assessed in different types of multimodal compositions. Reexamining feedback on L2 digital writing 585 5. Feedback in multimodal tasks Using the digital story assessment rubric provided above as an example, the question that remains is what shape and focus the formative feedback will have when L2 learners develop their multimodal compositions. Following Oskoz and Elola (2020), we propose a series of questions that address the criteria for digital stories identified in Table 1 and include rhetorical, linguistic, and nonlinguistic elements, as well as focus on the process of the multimodal ensemble (i.e., transformation and transduction), learner voice, and authorship. Let us keep in mind that as with any other task that we teach in the L2 classroom, we are using the rubric as a guideline that instructors could provide to students to compose the digital story, which include the linguistic and nonlinguistic elements as well as considerations of audience and issues of authorship and ownership, among others. Using this guideline for formative feedback purposes, instructors could reinforce and call students’ attention to, for example, the extent to which they have acknowledged the audience, how successfully they have used semiotic re- sources, or whether they have followed the genre characteristics. Thus, below we present the main seven components of the guidelines with some example questions to direct instructors in the provision of feedback in multimodal texts. Audience Today, composers develop and re-develop content for social networking sites such as Instagram and blogs, either for a real or imaginary audience, with the purpose of creating text that will encourage responses. Taking an audience into account, feedback might include questions such as the following: • Who is your audience? For whom are you writing? • What semiotic resources are you employing to attract your audience? How are you using them? • How do you react to and engage with your audience’s feedback? • What changes will you make based on your audience’s feedback (or lack thereof)? Semiotic resources The multimodal character of digital texts requires learners to integrate different semiotic resources in a meaningful manner. As McGrail and Behizadeh (2017) pointed out, L2 learners must understand “the unique conventions for creating such divergent multimodal compositions” (p. 34) and learn what elements are pertinent to the multimodal genre or their interests. L2 learners also need to learn how to apply “these particular elements and procedures in their own multimodal Idoia Elola, Ana Oskoz 586 designs” (Oskoz & Elola, 2020, p. 174). Feedback, therefore, needs to go beyond the local and global aspects of the language and encompass the additional se- miotic resources included in a multimodal composition, such as images or sounds. Feedback regarding different semiotic resources employed in, for exam- ple, a digital story might include some or all of the following questions. Oral • Does clear articulation and pronunciation from the narrator help the au- dience follow the story? • Does the author use meaningful repetitions or pauses in telling the story? • Does the oral narration facilitate the telling of an engaging story? • Does the oral narration have an emotional effect on the audience? Aural (music, sounds) • To what extent does the music tie in with the theme or the emotional atmosphere of the story? • Does the volume of the music (e.g., soft, loud) or interludes of silence help express the meaning of the story? • Do the sound effects, if any, complement the narrative flow of pictures and words throughout the story? • How do music and sounds reflect and augment the purpose of the story? Visual • Do the selected images correlate with the story being told, and are they integrated meaningfully throughout the story? • To what extent does the quality of the images interfere with or support the message conveyed? • Do images support the theme of the story and help the audience see the story’s main points? • To what extent do the images add to the story rather than merely repeat the written or oral text? Textual • What is the purpose of including written text (e.g., subtitles, words)? • To what extent does the text support or clarify the author’s meaning? Reexamining feedback on L2 digital writing 587 Does the text suit the still or moving pictorial elements that come be- fore, after, or during it? • How does the positioning, font, style, color, and content of the text sup- port the message the author wants to convey? • If text is necessary, is it shown on screen long enough for the viewer to read it? Language • Is the story told with an appropriate amount of detail; that is, is it not too long or too short? • Is a range of (multi)linguistic structures (e.g., verb tenses, subordination, or lack of subordination) evident in the narration of the story? • To what extent does the author use varied vocabulary to enhance and enrich the story? Genre characteristics Multimodal texts have their own idiosyncrasies and diverse practice criteria, which of- ten do not correspond to those of the printed text. When providing feedback for digital genres, there is a need to define the characteristics that make, for example, a good Wikipedia entry or an interesting travel blog, or the rhetorical characteristics of an en- gaging tweet. If these characteristics are clearly defined in the curriculum, we can then provide feedback for the development of different multimodal genres. Following McGrail and Behizadeh’s (2017) set of questions for creating and evaluating multi- modal composition, we propose the following feedback questions for a digital story: · How do you determine what constitutes a good digital story? · How do you effectively integrate different semiotic sources (e.g., visual, textual, aural, gestural) to develop a storytelling genre? Authorship While not always the case, multimodal digital projects tend to be collaborative endeavors that “[call] into question the dimension of content production” (Lotherington & Ronda, 2014, p. 22). This expanded concept of authorship implies the need to include feedback practices that foster and value the achievements and contributions of collective authorship. Providing feedback implies considering both individual and joint contributions. This is perhaps best achieved by combining peer feedback with traditional instructor feedback. Questions that can help instructors or peers provide feedback on authorship might include the following: · To what extent does this project reflect your equal intellectual contribution? Idoia Elola, Ana Oskoz 588 · How has the collective group integrated the voices of all members? · How has communication among all group members resulted in a respect- ful discussion of different ideas and led to a jointly produced product? · To what extent have you played to your own strengths (e.g., linguistic, visual, or aural knowledge) to enhance the quality of the product? Ownership Not to be confused with authorship, it is crucial to acknowledge digital writing ownership through the reuse and remix of semiotic resources. Ownership thus refers to the fair use of external sources by acknowledging and crediting ideas, images, and the eclectic range of semiotic resources that might be included in L2 learners’ multimodal projects. Providing feedback on the ownership of the re- sources provided is of relevance given that “many young people today consider what exists on the Internet as freely available raw material to be used however they see fit” (Chun et al., 2016, p. 69). As Chun et al. (2016) point out, borrowing content from others is not the problem per se but “rather the sense that borrow- ing does not require any acknowledgment” (p. 69). Questions we might ask that focus on intellectual property, copyright, and fair use could include the following: · Have you properly cited sources and credited those whose ideas you have included in your work? · Did you include websites to confirm that the source material was pub- licly available? · What steps did you follow to obtain permission to use copyrighted material? Voice and identity Another goal of multimodality in the L2 classroom is to provide a space for learners’ authorial voices and identities (Cimasko & Shin, 2017). Instructor feedback can help learners find their voices and express their identities by asking them to reflect on their choices of, for example, images, language(s), and colors. When looking at semi- otic resources and the role of synesthesia, research has examined the essential role of multimodality in shaping a learner’s authorial voice and identity (e.g., Cimasko & Shin, 2017; Jiang, 2018; Smith et al., 2017). When a learner is trying to find their voice, the feedback we provide can help them reflect on whether their choice of semiotic resource is appropriate. Questions that we might ask include the following: · Why have you chosen these images as representative of your story? · How does the music you selected represent your feelings and mood? · How does the digital story represent who you are? Reexamining feedback on L2 digital writing 589 · To what extent does this digital story express your voice based on lan- guage choice (e.g., vocabulary), images, sounds, and/or tone? Multimodal ensemble To help learners develop a multimodal ensemble, we need to consider how we provide feedback on multimodal texts and be aware that the process of synaes- thetic semiosis occurs in two forms during the construction of such a text (Kress, 2003, 2009). Transformation consists of actions that reorder and reposition se- miotic resources within a particular mode, whereas transduction involves the reorganization of semiotic resources across modes. In essence, we need to con- sider whether there is semiotic harmony; as we address below, it is important to obtain an orchestration of the diversity of resources from different modes, verifying that the form has the features needed to be the transferor of meaning. Transduction When developing a digital story, users combine, for example, visual (images), au- ral (sounds, spoken words, music), and textual (subtitles) modes. In general, learn- ers start with a narrative of the story they want to tell. Once they have written the story, or as they are writing it, they think of how different modes (e.g., aural, textual) will help create the meaning they want to convey. The feedback provided upon completing the first draft focuses on how the composer can combine different semiotic resources to convey their intended meaning. Because the goal of a multimodal text is to create synaesthetic har- mony, it is important that the learner understand that the integration of modes is directed at meaning and not at enhancing or describing the story. In this case, the composer reorganizes semiotic resources (text, images, sounds) across modes (textual, visual, aural) through transduction. • Are your pauses, sounds, or special effects used judiciously to convey meaning in to your story? • How are you combining different semiotic resources to reflect a digital story? • Is the merging of modes relevant to the meaning of the story? Are the modes balanced with each other? Transformation In addition to combining the different semiotic resources, L2 learners also reor- der and reposition them within a particular mode (visual, aural, oral, or textual). Learners might reconstruct the syntax or structural complexity of sentences Idoia Elola, Ana Oskoz 590 from a written narrative into a digital story script or play with the size and colors of an image to express their intended meaning. Questions that might help learn- ers maximize the use of the different modes include the following: • How do you play with the size and color of images to express your in- tended meaning? • How are you using changes in the volume of your soundtrack to reflect your mood? • How does the use of pauses and changes in tone in your oral narrative convey meaning in your story? • Does the transformation from narrative to script work? It is worth clarifying that feedback on multimodal texts is not, by definition, richly multimodal; on the contrary, many examples of peer and instructor feedback predominantly or entirely depend on comments or guidance expressed through words in isolation. When multimodal feedback is provided, however, it is not restricted to the digital form either; such an idea is immediately dispelled by watching how class- room teachers offer correctional advice or guidance through a varied assemblage of spoken language, silence, gestures, eye contact, and more (Lamb, 2018). 6. What is next? This article has first identified a gap in multimodal teaching and research regarding the role and focus of feedback in DMC and, second, has provided an assessment rubric and examples of formative feedback that addresses both linguistic and nonlinguistic elements to help students develop their multimodal texts. The truth is that the digital world we inhabit calls for the expansion of the role of feedback in response to the affordances of moods, tools, and task outcomes. Reasons why we might have not broadened our feedback type and form might be due to the follow- ing: that our view of multimodal composing is still limited to written texts; that the provision of feedback for multimodal texts is challenging; that we lack familiarity with producing multimodal ensembles; or that DMC is not always aligned with cur- ricular learning outcomes. We argue that feedback for DMC needs to be modeled following theoretical and pedagogical frameworks, genres, and tasks. Most im- portantly, feedback must be articulated in conjunction with multimodal assessment rubrics that guide the way for feedback (re)design. It is important, then, to align feedback with the particular assessments, learning contexts, and knowledge that the learner is attempting to convey. Thus, based on these multimodal ensemble rubrics, feedback can be more clearly tailored to the goals of the multimodal task. Overall, without disregarding print-based approaches to feedback, there is a need in digitally influenced learning environments and society to consider when, Reexamining feedback on L2 digital writing 591 where, and how our approaches to formative feedback should enhance the L2 multimodal representation of academic knowledge (Lamb, 2018). Although we traditionally use print-based or digital feedback for linguistic elements of multi- modal composition, this article advocates for the expansion of our notions of feedback, which should be formative and include nonlinguistic elements, and, more importantly, should cover the multimodal ensemble as a whole. The main questions that remain are how formative feedback on multimodal texts can help learners utilize semiotic modes to advance their multimodal writing abilities and whether feedback on multimodality can help L2 language development (Manchón, 2017). In Oskoz and Elola (2014, 2016a), we saw linguistic improvement as a result of the use of digital stories in the L2 writing class, but this was likely fostered by the task-based approach of the digital story assignment (i.e., the phased instruc- tion that involved the change from an academic narrative to an oral script popu- lated with images and sounds that helped create meaning). Although some feed- back guidelines have been proposed here, there is an urgent need for empirical evidence on how to tailor productive formative feedback to multimodal texts. There is also a need for research on the impact of feedback on linguistic and nonlinguistic development as well as on multimodal ensembles as a whole. Without a doubt, digital multimodal texts are here to stay and will increas- ingly populate diverse academic contexts. It is not far-fetched to affirm that research about feedback on DMC is nonexistent. In the L2 classroom, formative feedback is included because it is part of the learners’ scaffolding and revision processes when writing compositions (Oskoz & Elola, 2014, 2016a), but there is no evidence of the impact of having used instructor or peer feedback on DMC. Furthermore, although feedback in general is not necessarily multimodal, there has been an effort to inves- tigate whether different stages of the composing process and/or final product can be better addressed by multimodal feedback (Ducate & Arnold, 2012; Elola & Oskoz, 2016). It is also imperative to create a research agenda that explores and provides guidelines that are specific to the feedback type (e.g., explicit or less explicit multi- modal feedback for different stages of the composing process, such as drafting the script, choosing images and sounds, integrating different semiotic resources, re- hearsing the oral narrative, and polishing the final product). The current scarcity of knowledge regarding DMC feedback may be discour- aging; however, the potential for L2 multimodal research and teaching exploration is vast. Pursuing this line of inquiry is exciting not just because of the relevance of these multimodal texts but, more importantly, because of their intrinsic evo- lutionary nature, which calls for new ways to visualize feedback for multimodal texts and encourages us to redefine the notion of feedback as we know it. Idoia Elola, Ana Oskoz 592 References Barton, D. (2007). Literacy: An introduction to the ecology of written language. Wiley. Bezemer, J., & Kress, G. (2008). Writing in multimodal texts: A social semiotic account of designs for learning. Written Communication, 25(2), 166-195. Campbell, B. S., & Feldmann, A. (2017). The power of multimodal feedback. Journal of Curriculum, Teaching, Learning and Leadership in Education, 2(2), 1-6. Carpenter, R. (2009). Boundary negotiations: Electronic environments as inter- face. Computers and Composition, 26(3), 138-148. Caws, C., & Heift, T. (2016). Evaluation in CALL: Tools, interactions, outcomes. In F. Farr & L. Murray (Eds.), The Routledge handbook of language learning and technology (pp. 127-140). Routledge. Chen, H. I. (2013). Identity practices of multilingual writers in social networking spaces. Language Learning & Technology, 17(2), 143-170. Chun, D., Kern, R., & Smith, B. (2016). Technology in language use, language teaching, and language learning. Modern Language Journal, 100(S1), 64- 80. https://doi.org/10.1111/modl.12302 Cimasko, T., & Shin, D. S. (2017). Multimodal resemiotization and authorial agency in an L2 writing classroom. Written Communication, 34(4), 387-413. Ducate, L., & Arnold, N. (2012). Computer-mediated feedback: Effectiveness and student perceptions of screen-casting software versus the comment func- tion. In G. Kessler, A. Oskoz, & I. Elola (Eds.), Technology across writing contexts and tasks (pp. 31-56). CALICO. Elola, I., & Oskoz, A. (2016). Supporting second language writing using multi- modal feedback. Foreign Language Annals, 49(1), 58-74. Elola, I., & Oskoz, A. (2017). Writing with 21st-century social tools in the FL class- room: New literacies, genres, and writing practices. Journal of Second Lan- guage Writing, 36, 52-60. Fornara, F., & Lomicka, L. (2019). Using visual social media in language learning to investigate the role of social presence. CALICO Journal, 36(3), 184-203. https://doi.org/10.1558/cj.37205 Hafner, C. A. (2018). Genre innovation and multimodal expression in scholarly com- munication: Video methods articles in experimental biology. Ibérica, 36, 15-41. Hafner, C. A., & Ho, W. Y. J. (2020). Assessing digital multimodal composing in second language writing: Towards a process-based model. Journal of Second Lan- guage Writing, 47, 100710. https://doi.org/10.1016/j.jslw.2020.100710 Hafner, C. A., Chik, A., & Jones, R. H. (2013). Engaging with digital literacies in TESOL. TESOL Quarterly, 47(4), 812-815. https://doi.org/10.1002/tesq.136 Reexamining feedback on L2 digital writing 593 Hatzipanagos, S., & Warburton, S. (2009). Feedback as dialogue: Exploring the links between formative assessment and social software in distance learn- ing. Learning, Media and Technology, 34(1), 45-59. Heyd, T. (2016). Digital genres and processes of remediation. In A. Georgakopou- lou & T. Spilioti (Eds.), The Routledge handbook of language and digital communication (pp. 87-102). Routledge. Hicks, T. (Ed.). (2015). Assessing students’ digital writing: Protocols for looking closely. Teachers College Press. Hounsell, D., Xu, R., & Tai, C.-M. (2007). Balancing assessment of and assessment for learning. (Scottish enhancement themes: Guides to integrative assessment, N. 2). Gloucester: QAA. https://www.enhancementthemes.ac.uk/docs/ethe mes/integrative-assessment/guide-no-2---balancing-assessment-of-and-ass essment-for-learning.pdf?sfvrsn=8e2cf981_10 Hung, H.-T., Chiu, Y.-C. J., & Yeh, H.-C. (2013). Multimodal assessment of and for learning: A theory-driven design rubric. British Journal of Educational Tech- nology, 44(3), 400-409. https://doi.org/10.1111/j.1467-8535.2012.01337.x. Jewitt, C. (2008). Multimodality and literacy in school classrooms. Review of Re- search in Education, 32(1), 241-267. Jiang, L. (2018). Digital multimodal composing and investment change in learn- ers’ writing in English as a foreign language. Journal of Second Language Writing, 40, 60-72. Jiang, L., Yu, S., & Lee, I. (2022). Developing a genre-based model for assessing digital multimodal composing in second language writing: Integrating the- ory with practice. Journal of Second Language Writing, 57, 100869. Johnson, D., & Kress, G. (2003). Globalisation, literacy and society: Redesigning pedagogy and assessment. Assessment in Education: Principles, Policy & Practice, 10(1), 5-14. Kalantzis, M., & Cope, B. (2008). Language education and multiliteracies. In S. May & N. H. Hornberger (Eds.), Encyclopedia of language education (2nd ed., Vol. 1, pp. 195-211). Springer. https://doi.org/10.1007/978-0-387-30424-3_15 Kalantzis, M., Cope, B., Chan, E., & Dalley-Trim, L. (2016). Literacies (2nd ed.). Cam- bridge University Press. Kimber, K., & Wyatt-Smith, C. (2010). Secondary students’ online use and crea- tion of knowledge: Refocusing priorities for quality assessment and learn- ing. Australasian Journal of Educational Technology, 26(5), 607-625. https:// doi.org/10.14742/ajet.1054 Kress, G. (2003). Literacy in the new media age. Routledge. https://doi.org/10. 4324/9780203299234 Kress, G. (2009). What is a mode? In C. Jewitt (Ed.), The Routledge handbook of multimodal analysis (pp. 54-67). Routledge. Idoia Elola, Ana Oskoz 594 Kress, G. (2010). Multimodality: A social semiotic approach to contemporary communication. Routledge. https://doi.org/10.4324/9780203970034 Lamb, J. (2018). To boldly go: Feedback as digital, multimodal dialogue. Multimodal Technologies and Interaction, 2(3), 49. https://doi:10.3390/mti2030049 Lim, J., & Polio, C. (2020). Multimodal assignments in higher education: Implica- tions for multimodal writing tasks for L2 writers. Journal of Second Lan- guage Writing, 47, 100713. Lotherington, H., & Ronda, N. (2014). 2B or not 2B? From pencil to multimodal pro- gramming: New frontiers in communicative competencies. In J. Guikema & L. Williams (Eds.), Digital literacies in foreign language education (Vol. 12, pp. 9-28). CALICO. Luzón, M. J. (2013). Public communication of science in blogs: Recontextualizing scientific discourse for a diversified audience. Written Communication, 30(4), 428-457. https://doi.org/10.1177/0741088313493610 Manchón, R. (Ed.). (2011). Learning-to-write and writing-to-learn in an addi- tional language (Vol. 31). John Benjamins Manchón, R. (2017). The multifaceted and situated nature of the interaction be- tween language and writing in academic settings: Advancing research agendas. In Bitchener, J., Storch, N., & Wette, R. (Eds.), Teaching writing for academic purposes to multilingual students (pp. 183-200). Routledge. Maqueda, C. H. (2020). Heritage and L2 writing processes in individual and col- laborative digital storytelling (Unpublished doctoral dissertation). Texas Tech University. Matthewman, S., Blight, A., & Davies, C. (2004). What does multimodality mean for English? Creative tensions in teaching new texts and new literacies. Education, Communication & Information, 4(1), 153-176. McGrail, E., & Behizadeh, N. (2017). K-12 multimodal assessment and interac- tive audiences: An exploratory analysis of existing frameworks. Assessing Writing, 31, 24-38. Mehlenbacher, A. R. (2017). Crowdfunding science: Exigencies and strategies in an emerging genre of science communication. Technical Communication Quar- terly, 26(2), 127-144. https://doi.org/10.1080/10572252.2017.1287361 Mills, K. A. (2006). Mr Traveling-at-will Ted Doyle: Discourses in a multiliteracies classroom. Australian Journal of Language and Literacy, 29(2), 132-149. Mina, L. (2014). Enacting identity through multimodal narratives: A study of multilingual students. In J. G. Guikema & L. Williams (Eds.), Digital litera- cies in foreign language education: Research, perspectives, and best prac- tices (pp. 141-158). CALICO. Nelson, M. E. (2006). Mode, meaning, and synesthesia in multimedia L2 writing. Language Learning and Technology, 10(2), 56-76. Reexamining feedback on L2 digital writing 595 New London Group. (1996). A pedagogy of multiliteracies. Harvard Educational Re- view, 66(1), 60-93. https://doi.org/10.17763/haer.66.1.17370n67v22j160u Oskoz, A. & Elola, I. (2014). Integrating digital stories in the writing class: To- wards a 21st century literacy. In J. G. Guikema & L. Williams (Eds.), Digital literacies in foreign language education: Research, perspectives, and best practices (pp. 179-200). CALICO. Oskoz, A., & Elola, I. (2016a). Digital stories: Bringing multimodal texts to the Spanish writing classroom. ReCALL, 28(3), 326-342. Oskoz, A., & Elola, I. (2016b). Digital stories: Overview. CALICO Journal, 32(2), 155-173. Oskoz, A., & Elola, I. (2020). Digital literacies for writing in the 21st Century FL classroom. Equinox. Reinhardt, J., & Thorne, S. L. (2011). Beyond comparisons: Frameworks for de- veloping digital L2 Literacies. In N. Arnold & L. Ducate (Eds.), Present and future promises of CALL: From theory and research to new directions in language teaching (pp. 257-280). CALICO. Ruiz Pérez, S. (2022). Multimodal voices in the L2 classroom: Redesigning a dig- ital story, a formative experiment (Unpublished doctoral dissertation). Texas Tech University. Sancho Guinda, C. (2015). Genres on the move: Currency and erosion of the genre moves construct. Journal of English for Academic Purposes, 19, 73-87. Shin, D. S., Cimasko, T., & Yi, Y. (2020). Development of metalanguage for multi- modal composing: A case study of an L2 writer’s design of multimedia texts. Journal of Second Language Writing, 47, 100714. Smith, B. E., Pacheco, M., & de Almeida, C. R. (2017). Multimodal code-meshing: Bilingual adolescents’ processes composing across modes and languages. Journal of Second Language Writing, 36(2), 6-22. Towndrow, P. A., Nelson, M. E., & Yusuf, W. F. B. M. (2013). Squaring literacy assessment with multimodal design: An analytic case for semiotic aware- ness. Journal of Literacy Research, 45(4), 327-355. Yi, Y., King, N., & Safriani, A. (2017). Reconceptualizing assessment for digital multi- modal literacy. TESOL Journal, 8(4), 878-885. https://doi.org/10.1002/tesj.354 Yang, Y.-F. (2012). Multimodal composing in digital storytelling. Computers and Com- position, 29(3), 221-238. https://doi.org/10.1016/j.compcom.2012.07.001 Zhang, M., Akoto, M., & Li, M. (2021). Digital multimodal composing in post- secondary L2 settings: A review of the empirical landscape. Computer As- sisted Language Learning, 1-28.