What beliefs about writing guide EFL curricula? An analysis of relevant policy documents for teaching English at German secondary schools

Although curricular analyses are of crucial importance for curricular development, documents for teaching writing in English as a foreign language [EFL] in Germany have not yet received much attention. Our study explores what beliefs about writing guide educational policy documents for teaching EFL in Germany. Using Ivanič’s (2004) discourses of writing [DoW] as a theoretical lens, we conduct a curricular analysis of 11 curricula for lower secondary school (Year 9/10) and their guiding superordinate documents. Our data suggest that none of the documents offers a comprehensive conceptualization of writing. The skills and the genre discourses are predominant; the process, the creativity, and the social practices discourses receive little attention, while the sociopolitical discourse of writing is missing. Differences in discourse frequency in curricula among school types and federal states may hinder transitions between educational tracks. The relative neglect of the procedural character of writing and the lack of acknowledgment of the social dimension of learning to write further suggest a misalignment between curricula and current research into (foreign language) writing pedagogy. This may be particularly detrimental for less proficient or socially disadvantaged learners. Implications for curricular development are outlined, suggesting that a more comprehensive approach to writing that covers all six DoW is important. In particular, we highlight the potential of process-oriented writing and also refer to the need to address digital literacy in writing.


Introduction
It is well known that writing in general is a crucial competence and a key factor for participation in many areas of life, and thus needs to be adequately supported at school (e.g., Graham and Perin 2007). Fostering writing in English as a foreign language [EFL] in particular has become increasingly necessary in Europe due to globalization and digitalization processes but is also important in enabling students to become socially responsible agents who participate in universal discourses. The latter is a core educational concern (Klafki 2007) and central to multiple contemporary discourses of writing that recognize writing as a socially or politically situated practice (Ivanič 2004). Modern communicative EFL pedagogy similarly aims at preparing students for using (both oral and written) language in authentic contexts, ultimately enabling them to make their voices heard in (cultural, political, societal) discourses (e.g., W. Hallet 2011).
However, little is known about the relative importance placed on writing (and the social dimension of writing) in EFL classes in Germany. Existing evidence suggests that students mostly engage in oral activities (Helmke et al. 2008) and that writing receives less in-class support (Porsch 2010a), which could indicate a lack of teacher preparedness for teaching writing observed in other contexts (e.g., Parr and Jesson 2016). Therefore, curricula and supporting documents are relevant, as they inform on expected standards. By setting student goals, curricula specifically offer evaluative measures and guidance for teacher practice (Hericks and Kunze 2008;Richards 2017). Importantly, policy documents inform textbooks 1 , which in turn assert a major influence on how languages are taught in Germany (Bieswanger 2008). Thus, an exploration of policy documents is vital and also a necessary precondition for curricular development.
Yet, a systematic analysis of EFL writing in German policy documents is still missing. The present article seeks to address this research gap by analyzing policy documents that inform teaching practices and set the expected standards. We examine secondary school curricula for teaching English (Year 9/10) in four German federal states and their guiding superordinate documents. We draw on Ivanič's (2004) discourses of writing [DoW], which offer a comprehensive analytical framework to scrutinize underlying beliefs about writing in educational documents.

Theoretical background
Current German educational policy documents have been impacted by the educational reforms that were initiated after the 2001 PISA study, with the English curricula being no exception (see Helbig-Reuter 2005). German students' unsatisfactory performance undermined faith in school effectiveness and equity and was accompanied by a paradigmatic shift in education with the introduction of standardized, competence-based policy documents (Drieschner 2009;Kolb and Raith 2018). The new documents prioritized students' attainment of normative, objectifiable competence levels over specific content (Drieschner 2009;Elsner 2018). In addition, policy documents introduced so-called Operatoren, that is, key verbs that signal specific expectations and text formats, to standardize assessments (Tesch 2020), such as the end-of-year-10 final exam, which is nowadays centrally administered by most states, including those selected for this study (see e.g., Schulgesetz für das Land Nordrhein-Westfalen 2005).
From an organizational point of view, curricula in Germany are intended to implement the more general learning objectives that are established by superordinate policy documents (Böttcher 2003). The Bildungsstandards [Educational Standards] (KMK 2004a), that were published by Germany's Standing Conference of the Ministers of Education and Cultural Affairs [KMK] during the post-PISA reforms, are one of the superordinate documents that inform curricula. These standards stipulate requirements for the Mittlerer Schulabschluss [MSA], that is, the end-of-year-10 graduation diploma (W. Hallet 2011), independent of specific school types (Bellenberg 2012), and were thus intended to allow students to transition upwards between school types (Cortina 2003). Regarding teaching English, Germany's Bildungsstandards (KMK 2004a) are further informed by the Common European Framework of Reference for Languages [CEFR] (Council of Europe [COE] 2001). Both the Bildungsstandards and the curricula reference the CEFR levels of language competence in their student expectations. The organizational relationships of these documents are illustrated in Fig. 1. Matters of education policy are decided by Germany's federal states. Thus, despite efforts to standardize educational policy documents across Germany, the alignment of state curricula with the KMK's (2004a) Bildungsstandards remains the responsibility of the individual state (Böttcher 2014). Therefore, differences in the curricula are to be expected.
In terms of content, contemporary English curricula in Germany claim their commitment to the communicative language teaching [CLT] approach (see e.g., North Rhine-Westphalia's 2019 Curriculum for Gymnasium), which has been highly influential in English language teaching [ELT] in Germany (Doff 2018;Elsner 2018) and around the world (see e.g., Butler 2011). CLT approaches are based on the idea that knowledge about a language does not suffice for its successful use in communication, insomuch as it lacks a performative component (Hymes 1972). Instead, these approaches suggest learners should engage in interactive settings that employ authentic tasks and materials, which is believed to best promote meaningful communication (Hiep 2007;Savignon 2002). CLT approaches also draw on the work by Canale (1983) and Canale and Swain (1980), who define communicative competence as the integration of "grammatical [...], sociolinguistic, [...] strategic" (Canale and Swain 1980, p. 28) and "discourse competence" (Canale 1983, p. 9). Thus, when framed within CLT, writing is more than producing grammatically and orthographically correct sentences or texts. According to modern understandings of CLT, communicative competence in English should enable students to contribute to and participate in global discourses and thus help shape societal negotiation processes (e.g., W. Hallet 2011).
Interest in authentic communicative activities is also highly relevant to the taskbased approach to language learning and teaching [TBLT], which has emerged from the communicative paradigm and is now often considered to be one of its most relevant manifestations (e.g., Abendroth-Timmer and Gerlach 2021). TBLT structures teaching around relevant tasks and initiates learners' active use of the target language by engaging them in problem solving (Ellis 1999;Long 2016;Van den Branden 2006). It has been suggested that the potential for learning in such TBLT  settings is best exploited through genre-oriented writing tasks, as mastery of genres is closely associated with achieving an intended meaning in writing-a concern that is central to TBLT approaches (e.g., Byrnes 2014). In Germany, the TBLT approach has become highly relevant in EFL education (Müller-Hartmann and Schocker-von Ditfurth 2011). It may thus be assumed that the influence of both CLT and TBLT is also evident in EFL writing curricula in Germany.

Analyzing beliefs about writing in educational policy documents: Ivanič's (2004) discourses of writing [DoW]
Since "[w]riting curriculum documents are underpinned, consciously or sub-consciously, by particular conceptualizations of writing, together with perceptions of how children learn to write and how teachers should teach writing" (Peterson 2012, p. 260), analyzing documents for such underlying beliefs can provide key insights into what policy makers consider high-quality writing and writing education. To identify such underlying beliefs about writing, Ivanič (2004) established a frame-work using six DoW 2 , (i.e., skills, genre, process, creativity, social practices, and a sociopolitical DoW). Following Gee's (1996) definition of the term 'discourse', Ivanič (2004) describes DoW as "constellations of beliefs about writing, beliefs about learning to write, ways of talking about writing, and the sorts of approaches to teaching and assessment which are likely to be associated with these beliefs" (p. 224). Authentic documents or practices, however, may comprise "hybrid instantiations" (Ivanič 2004, p. 240) of multiple discourses.

Skills
This discourse focuses on applying linguistic knowledge in writing and is recognizable in its concern with "spelling, punctuation and grammar, in expressions such as 'correct ', 'accurate', 'proper', [...] and in an emphasis on accuracy in assessment" (Ivanič 2004, p. 228). Such prioritization of formal accuracy in writing might be traced back to the earliest beginnings of EFL teaching, which derived its form-focused instructional methods from grammar-translation approaches to teaching Latin (see Doff 2018). Research today increasingly recognizes texts as more complex than this discourse implies.
Genre Genre-based approaches here center on the writing products and different text types, which serve specific social functions and may differ depending on purpose and context (Ivanič 2004). Although the use and defining lines of the genre terminology remain a topic of debate (for a discussion see e.g., Derewianka 2003;Kress 1993), according to Ivanič (2004), genre-based approaches are embedded in the view that characteristic features of text types need to be explicitly taught in order to enable students to use text types appropriately for different purposes such as "recounting, describing, informing, instructing" (p. 233).
Process Flower and Hayes's (1981) pivotal work outlines how writers perform various taxing cognitive processes while composing a text. According to Ivanič (2004), procedural aspects of writing contain both cognitive aspects, that is, the implicit "mental processes of meaning-making" (p. 223), and more explicit practical elements, such as "planning, drafting and revising writing" (p. 231). The process discourse describes approaches to writing that direct the focus to these cognitive and practical aspects (Ivanič 2004), which are recognized as cyclical (e.g., Bruton 2009;Sokolov 2014;Zamel 1983).
Creativity Creative approaches to writing recognize the text as the creation of the author's mind (Ivanič 2004). Creative writing aims at initiating students' "selfexpression" (Ivanič 2004, p. 239), thus allowing writers to share their thoughts and perspectives, independent of external constraints (Ivanič 2004). This approach is particularly useful in diversity-sensitive education that seeks to empower students, as it explicitly values "writing which represents the [...] 'voice' of learner writers from minority and disadvantaged backgrounds, challenging elitist ideas of what counts as an interesting narrative or topic to read about" (Ivanič 2004, p. 229).

Social practices
The social practices DoW is concerned with the event of a social interaction in which a piece of writing is produced and suggests that writers need to engage in authentic activities that are significant to them (Ivanič 2004). Often, these approaches pose communicative problems to be solved by the learner which can include potential external expectations the learner might be confronted with in real life, as for example those of "someone in authority such as an employer, or imposed by bureaucracy" (Ivanič 2004, p. 235).

Sociopolitical
The sociopolitical DoW features "a strong sense of the social, cultural and political location" (Ivanič 2004, p. 239) of writing. It centers on the interrelations of text production and discourse communities and values learners' agency in writing (Ivanič 2004;Little and Erickson 2015;Tardy 2006). In the classroom, the sociopolitical DoW can be explored through concepts such as critical language awareness (Ivanič 2004), which mainly aims at sensitizing students toward a range of issues including "worldviews; ideologies; diversity issues involving power, class, gender, race, sexuality; how discussions of language are frequently invisibilized; and, finally, discourse features and techniques" (Taylor et al. 2017, p. 1). Fostering students' awareness of sociopolitical concepts, enabling them to challenge discourses and thus informedly and autonomously participate in societal negotiation processes is a key concern of this DoW (Ivanič 2004).

Empirical studies using Ivanič's (2004) DoW to analyze curricula
In recent years, Ivanič's (2004) DoW framework has frequently been used to analyze educational documents concerning underlying beliefs about writing (see e.g., Jeffery and Parr 2021;Peterson 2012;Peterson et al. 2018). In the following, we focus on four different analyses in countries whose documents are especially interesting because they place a strong emphasis on writing and cover a wide range of discourses. Note, however, that these analyses focus on writing in the regular language of instruction. Table 1 illustrates the discourse representation identified across these international examples of curricula, contrasting it with a recent analysis of curricula for writing in German (Müller et al. 2021). It furthermore needs to be acknowledged that the emphasis on discourses depicted in the below table is based on the interpretations made by the respective authors, some of whom count and thus quantify discourse frequency (Canada, Germany, and New Zealand), while others choose a qualitative approach to establish emphasis (Denmark and the United States). Comparability of results from different countries can therefore not be guaranteed. Accordingly, Table 1 must be viewed as an approximate representation of the discourses in these countries. Ontario's language curriculum for grades 1-8 exhibits a strong focus on the process discourse, with 38.9% of its text segments referring to process (Peterson et al. 2018, p. 508), despite the skills discourse being impactful at the same time. References to the process discourse in Ontario's curriculum, however, mainly concern practical issues ( Peterson et al. 2018). These foci are followed by references to genre, creative writing, and the social practices discourses. However, stipulations regarding the latter two discourses are far less detailed and include less precise examples than those referencing the dominant discourses.
An investigation of the Danish writing curriculum for grades 1-9 by Elf and Troelsen (2021) finds all DoW represented to varying degrees, although the sociopolitical discourse occurs only implicitly. There is a tendency toward the process discourse in some curriculum areas, but this is contradicted in others. The authors further indicate limitations to their application of the DoW framework. For example, Elf and Troelsen (2021) point to a section called "joyful writing" (p. 181), which defies easy allocation to any one of the DoW, since it is only partially categorizable by a combination of the creativity, social practices, and the sociopolitical discourses.
New Zealand's curriculum for Level 1 English exhibits a balanced proportion of all DoW except the sociopolitical discourse (Peterson et al. 2018). It foregrounds the creativity and process discourses, including both practical and cognitive aspects of the latter. Peterson et al. (2018) identify this focus on creativity and process as following the idea of "personal growth through language" (p. 511) that is commonly known in New Zealand. The authors furthermore indicate that references to genre and social practices tend to occur in a hybrid discourse, which they call "Genre as Social Process" (Peterson et al. 2018, p. 511). In reference to these findings, Parr (2021) concludes that New Zealand's curriculum aligns with the well-established idea that "social and cultural practices shape literacy learning" (p. 32).
In the United States, too, some curricula make ample references to different discourses. Wilcox et al. (2021) examined curriculum materials for grades 9-12 from five American states. While the state curricula are most similar in representing the skills, process, and genre discourses, California's curriculum recognizes all DoW, including the sociopolitical discourse, which otherwise frequently seems to lack representation. California's curriculum thus frames writing as a socially situated event rather than the mere application of linguistic rules. Moreover, Wilcox et al. (2021) identify explicit recognition of the state's diverse student demographic, including linguistic and cultural heritage, in the Californian curriculum. Issues of power concerning varieties of English are explicitly addressed in California and, interestingly, To reiterate, it seems that references to the sociopolitical discourse are most commonly identified in analyses of US curricula. Creative writing and the social practices discourse are acknowledged frequently across the examined countries, albeit their stipulations are often described as less precise. The process discourse is very present; in some curricula, even its more implicit cognitive components are included, as for example in New Zealand (Peterson et al. 2018).
In contrast to the multifaceted view of writing presented in these international curricula, German curricula for writing in the subject German show a relatively narrow conceptualization of writing. Müller et al. (2021) examine curricula and educational standards for the Hauptschule diploma, "which define federal minimum competencies that must be obtained to enter the professional sphere" (p. 130) across six German federal states and find that these documents are mostly informed by the genre and skills discourses. In contrast to international curricula, the process discourse receives comparatively little attention. In addition, the documents lack any mentions of two discourses: the social practices and the sociopolitical discourses. The former is present in all other international examples mentioned here and the latter is present in two countries (United States, Denmark).
In sum, it appears that conceptualizations of writing are comparatively narrow for the subject German and fail to address sociocultural issues. However, to our knowledge, no prior analysis has been conducted to explore conceptualizations of writing for EFL teaching in Germany. The present article addresses this research gap and explores the presence of the six DoW in policy documents for EFL teaching in Germany.

Aims and research questions
The present article explores beliefs about writing in EFL Curricula for German secondary schools. In addition, we analyze superordinate policy documents informing the Curricula. The research questions below guide our analysis.

RQ1
To what extent are Ivanič's (2004)  RQ3 Do the Curricula or the superordinate documents exhibit further writingspecific stipulations or other aspects of writing that cannot be allocated to Ivanič's (2004) DoW, and if so, which?

Design
We conducted a curricular analysis of 11 EFL Curricula for secondary schools in four German federal states using the theoretical lens of Ivanič (2004), by identifying writing-specific proficiency statements in the documents and allocating them to the DoW, thus operationalizing the documents' underlying beliefs about writing. The states selected were North Rhine-Westphalia, Germany's most populous state, located in the west; Bavaria, a southern state; Saxony, located in the east and former member state of the GDR; and Bremen, a northern city state. The sample is thus demographically diverse, includes regional variations, and achievement differences as seen in large-scale assessments of students' performance (e.g., Stanat et al. 2016). The superordinate documents we used for comparison included the Bildungsstandards (KMK 2004a) and two editions of the CEFR (COE 2001(COE , 2018. We investigated the documents' writing-specific requirements for the MSA diploma, or CEFR level B1 respectively.

Data analysis
The procedure for the analysis employed by this study followed the examples set by previous studies that use Ivanič's (2004) DoW. We adapted, in particular, the analytical steps outlined by Müller et al. (2021) for the Main Analysis. We added two further steps, which in the following we refer to as our Minor Analyses I and II. For each document, we followed the analytical steps illustrated in Fig. 2.
Following Müller et al. (2021), in the Main Analysis, the writing-specific proficiency statements in the documents were identified. Sections qualified for analysis when their titles included the terms 'writing', 'written production', 'production of text' or relevant synonyms. In the common case of bullet point listings, each bullet point was considered an individual statement. Each of the statements listed in bullet points was then allocated to one of Ivanič's (2004) DoW, based on which discourse grasped the statement's essential feature best. In instances where a statement did not fit one of the discourses, it was allocated to the extra category 'other aspects of writing'. Following that, the total number of proficiency statements referencing writing was counted (NDoW) as well as the number of statements that were allocated to each DoW or to the extra category (nDoW), so that percentages could be calculated.
In the second phase of analysis (Minor Analysis I) we added onto the procedure outlined by Müller et al. (2021) and focused on statements referencing multiple discourses. Ivanič (2004) describes these statements as cases of "hybrid instantiations" (p. 240), where a main discourse is supplemented with components of other discourses. To calculate percentages, the total number of hybrids (NHybridDoW) was counted, and the frequency of combinations (nHybridDoW) was recorded.
In a third step (Minor Analysis II), references that were allocated to the extra category 'other aspects of writing' in step one (NOtherDoW) were revisited. We compared these statements for similarities and explored the potential for creating new categories. Thus, when possible, new categories were established, and the references

Findings
For research question (1) To what extent are Ivanič's ( 2004) DoW represented in the Curricula?, our analysis finds that 128 proficiency statements in the Curricula can be allocated to the DoW by Ivanič (2004) and 6 statements to the category 'other aspects of writing'. The Curricula are dominated by the skills discourse, to which an overwhelming 44% of the 134 proficiency statements can be allocated. This is followed by the genre discourse, to which 30% of the statements refer. These two, the skills and the genre DoW, are also the only DoW that can be found in every Curriculum. The process and creativity discourses are each represented by 9% of the proficiency statements, followed by the social practices discourse, exhibited by 4%. Manifestations of the genre discourse occur the second most often in the Curricula and are rather elaborate, presenting several text types that students should master such as descriptions, reports, and comments, in addition to letters. Accordingly, statements mostly refer to descriptive, informative, and argumentative writing purposes. Letters are presented in connection to social purposes of varying specificity, including letters to the editor, letters of complaint, but also personal and formal letters whose purposes are not specified.
Special attention needs to be paid to the representation of the process discourse, or lack thereof, across the Curricula. No Curriculum mentions underlying cognitive processes of writing. All process-related statements refer solely to practical issues, with a focus on planning, while revision receives little attention. Only one of the two process references in North Rhine-Westphalia's (2019) Gymnasium Curriculum explicitly features revision. In the case of the state's Gesamtschule Curriculum (North Rhine-Westphalia 2004a), revision is mentioned once per track (basic level G-track and advanced level E-track), but only in statements that simultaneously contain elements of other discourses and were therefore identified as hybrids. Although there is further mentioning of the 'correction' of texts in other Curricula (Bremen 2010;North Rhine-Westphalia 2004a, b, 2011, these may be more adequately allocated to the skills discourse because statements such as "make comprehensive linguistic corrections" [umfassende sprachliche Korrekturen vornehmen] (North Rhine-West-phalia 2011, p. 33) suggest that process-oriented teaching is not their core concern. Bavaria's and Saxony's Curricula mention neither revision nor correction.
Moreover, the process discourse tends to be featured more in the academically oriented Gymnasium than in less advanced school types. For example, North Rhine-Westphalia's (2019) Curriculum for Gymnasium shows a distinctly larger proportion of the process discourse (17%), than the state's Hauptschule Curriculum (North Rhine-Westphalia 2011), which only features the process discourse in 8% of its statements. An exception to this observation can be found in Bavaria, where the Curriculum for the least advanced secondary school type, Mittelschule (Bavaria n.d.c), exhibits the state's largest share of references to the process discourse. Bavaria's other two Curricula (Bavaria n.d.a, n.d.b) do not feature the process discourse at all, thus inevitably making the still rather small share of 8% process discourse in the Curriculum for Mittelschule (Bavaria n.d.c) the state's largest share.
Those few statements that can be identified as references to the creativity discourse are largely superficial and vague. For example, Bremen's (2010) Oberschule Curriculum expects students to be able to "creatively modify texts" [Texte kreativ verändern] (p. 19) and North Rhine-Westphalia's (2004b) Realschule Curriculum requires students to "employ basic means of creative writing" [einfache Formen des kreativen Schreibens einsetzen] (p. 35). In most of the instances, the Curricula merely mention the adaptation of a text, or the creation of an alternative beginning or ending.
References to the social practices discourse deal with vocational aspects, such as completing application forms or writing a curriculum vitae. Instances of the social practices discourse can be found in some of North Rhine-Westphalia's, Bavaria's, and Saxony's Curricula. In addition, this discourse is more frequent in Curricula for school types whose students are expected to cease schooling and enter the professional world after graduating with the MSA diploma, such as North Rhine-Westphalia's Hauptschule (2011) or Gesamtschule G-Track (2004a) and Bavaria's (n.d.c) Mittelschule. In direct contrast, Saxony only shows representations of the social practices discourse in its Curriculum for Gymnasium (2019a).
The Curricula exhibit fluctuating degrees of discourse frequencies across the different secondary school types of a given federal state. The representation of the genre discourse in Bavarian Curricula, for example, varies from 15% in the state's Curriculum for Mittelschule (n.d.c) to 75% in its Realschule (n.d.b). In Saxony, the Curriculum for Oberschule (2019b) is dominated by an extensive focus on the skills discourse to which 75% of its statements can be allocated, while the state's Gymnasium (2019a) Curriculum features the skills discourse in 36% of its stipulations.
Likewise, comparing similar school types across the states paints a picture of heterogeneity. The school type Oberschule, for example, can be found in Saxony and Bremen. Their respective stipulations, however, vary greatly. While Saxony's (2019b) Oberschule shows a substantial share of 75% skills, only 40% of the statements in Bremen's (2010) Oberschule can be allocated to the skills discourse.
The analysis further reveals "hybrid instantiations" (Ivanič 2004, p. 240) of discourses in the documents. In total, hybrids can be found in all four federal states, albeit not in every Curriculum. All in all, the pairing of genre as a main discourse  (Ivanič 2004) in the documents' writing-specific proficiency statements with creativity as a sub-discourse occurs most frequently, accounting for seven out of 16 total instances, which equals 44% of all hybrids, followed by the pairings of genre/skills (two instances, 13%) and skills/genre (two instances, 13%). Five more pairings can be identified, each of which occurs only once.
Concerning research question (2) To what extent are the stipulations about writing in the Curricula aligned with those in Germany's superordinate educational policy documents?, our analysis reveals that of the 49 proficiency statements in the Bildungsstandards (KMK 2004a) and the two editions of the CEFR (COE 2001(COE , 2018, 47% can be allocated to the genre discourse, followed by skills with 41%. A total of 6% of the statements refer to the writing process, 4% to creativity, and finally, 2% to the social practices discourse. Both the sociopolitical discourse and the extra category 'other aspects of writing' are absent from all superordinate documents. In both editions of the CEFR (COE 2001(COE , 2018 the genre discourse accounts for the largest share of statements, while the Bildungsstandards (KMK 2004a) are dominated by skills. Furthermore, the Bildungsstandards (KMK 2004a) do not exhibit any representatives of the creativity discourse. They do, however, include references to the social practices discourse, and an outstandingly large share of 19% of its proficiency statements can be allocated to process-both of which are absent from either edition of the CEFR (COE 2001(COE , 2018. No other document, including the Curricula, matches the Bildungsstandards' (KMK 2004a) share of 19% process discourse, with one third of the Bildungsstandards' process-related stipulations mentioning revision.
When comparing the findings from all superordinate documents to the findings from all Curricula, a clear similarity in terms of their tendency toward skills and genre is noticeable. The skills statements in the superordinate documents are disproportionately elaborate, as is the case in the Curricula. Regarding genre, both document types also mention the same writing purposes (mostly descriptive, informative, and argumentative) and text types (mostly letters, reports, and essays). On average, Curricula feature the creativity (9%) and the process discourses (9%) more than their superordinate counterparts do (4% creativity and 6% process), despite the Bildungsstandards' (KMK 2004a) obvious focus on process. None of the documents references the sociopolitical discourse. Fig. 3 provides an overview of the distributions of the DoW per document and in total.
Regarding research question (3) Do the documents exhibit further aspects of writing that cannot be allocated to Ivanič's ( 2004) DoW?, our analysis reveals that the documents contain six instances of proficiency statements that do not fit into the DoW framework and were categorized as 'other aspects of writing'. All occur in Curricula from North Rhine-Westphalia and concern vague stipulations for digital writing. There is one instance each in the Curricula for Hauptschule (2011) and Gymnasium (2019). There are two references in the Curriculum for Gesamtschule (2004a) (one per track), and two instances in the Curriculum for Realschule (2004b). Most of these references mention the use of software and other digital technology for writing without specifying these requirements any further. The Realschule Curriculum (2004b) further allocates the use of software in text production within the scope of the skills discourse, as it expects students to successfully use digital technology to scour their texts for errors and correct them: "use a spellcheck program" [ein Rechtschreibprogramm für den Fehlercheck einsetzen] (p. 38).

Discussion
Our investigation of policy documents for EFL writing in German secondary schools uncovers a misalignment between the documents' underlying beliefs about writing and current research into writing pedagogy in general, and foreign language writing pedagogy in particular. None of the documents exhibits a comprehensive approach to writing, which is discussed along six key findings below.
(1) The documents are guided by a focus on the skills and the genre DoW, with skills being the most prevalent. Every document is dominated by one of these two DoW, and they also constitute the only two DoW that occur in every document.
As the skills discourse is the most elaborate, there is a risk that teachers assume that teaching linguistic knowledge, including such aspects as grammar, orthography, and punctuation, is also the most important. While such a focus is understandable in the CEFR's descriptive levels of language competence, the Curricula may require a slightly different focus. Excessive attention to skills could result in overly formfocused teaching, which does not appropriately cater to learners' needs and may even increase writing anxiety.
As genre-based pedagogy was originally rooted in the idea that mastery of generic forms can function as a gatekeeper to their associated discourse communities (Kress 1993; see also W. Hallet 2016), the strong presence of the genre DoW in the documents might suggest an effort to foster students' command of influential genres. In this sense, Abendroth-Timmer and Gerlach (2021) point out the importance of keeping genre connected to learners' real-world communicative needs. Writing in a specific genre requires students to recognize textual conventions which, in turn, can function as a safety net that may help less proficient learners conceptualize texts more globally (Firkins et al. 2007) and pay more attention to deeper level issues in writing (Kay and Dudley-Evans 1998). The central position of TBLT in German EFL practice, which is often associated with genre-oriented writing instruction (e.g., Byrnes 2014), might be influential here. In addition, the standardized Operatoren in German EFL education might be another factor associated with the strong focus on genre observable in the documents. In North Rhine-Westphalia, for example, the list of standardized Operatoren includes the verb "discuss" (QUA-LIS NRW 2021, n. p.), which requires students to write argumentative texts of a given structure-a task which fits into the genre discourse. These argumentative writing purposes mentioned in the documents carry potential for engaging students in authentic negotiation processes, which seems especially beneficial for democratic participation. Nonetheless, literacy in genres requires learners to access a multitude of genres and associated social discourse communities. Arguing, describing, and informing feature in the documents the most often, yet they only represent a small part of the genre discourse. Narrative writing, for example, is lacking major reference in policy documents, although it has the particularly helpful potential to gauge EFL writing competence (see Gerlach and Götz 2020). Furthermore, Tesch (2020) notes how the list of Operatoren published by the KMK and the German Institute for Educational Quality Improvement [IQB] narrows text formats to those most relevant to the high school graduation exams.
The predominance of the skills and the genre DoW might thus be linked to the post-PISA educational reforms, which are frequently associated with a tendency to adapt pedagogy to the demands of standardized performance assessments (e.g., Gruschka 2019). For teaching English in particular, researchers have voiced similar concerns (e.g., Keller 2013; Tesch 2020). Since in writing, matters of genre and skills are especially easy to assess, teachers might run the risk of reducing writing pedagogy to these two discourses.
(2) The process DoW is more present in the Bildungsstandards (KMK 2004a) than in the Curricula. In the Curricula, it is lacking full representation, particularly concerning revision.
Planning, reviewing, and revising are key factors in fostering students' writing competence and are part of the process DoW. However, they can be difficult to master, especially when writing in a language other than the L1. As a meta-analysis by Silva (1993) shows, planning and revision cycles are substantially easier to conduct in the L1 than in other language contexts, and L2 writers often plan less and revise less automatically than L1 writers. Although Silva (1993) does not distinguish between L2 and FL learners, these findings might be particularly important for FL learners, who may have even less intuition and more insecurity. Therefore, one would expect curricula to highlight the centrality of the writing process. It is thus concerning that, within our sample, Curricula for less advanced school types often feature the process discourse the least, which is similar in documents for the subject German (Müller et al. 2021). This exclusion seems to contradict the guidance set forth in the Bildungsstandards (KMK 2004a, b), both for writing in the subjects German and English and may point to a larger problem with how federal states use the Bildungsstandards to create their curricula. It further indicates a tendency of writing pedagogy in Germany to exclude revision and multiple drafts from classroom writing practices, regardless of the target language, which ties in with findings by Porsch (2010a) that teachers rarely incorporate drafting into lessons and students receive little feedback on their writing. The lack of planning, reviewing, and revising in German documents points toward a product orientation, as opposed to a process orientation observed in other countries such as Canada (Peterson et al. 2018) or New Zealand (Parr 2021;Peterson et al. 2018). Although findings from these countries are admittedly not set in an FL writing context, the lack of attention to process-oriented writing in our documents is concerning, since FL learners, in particular, benefit from revising and feedback activities (see the meta-analysis by Biber et al. 2011) which are part of the process approach.
In addition, areas of writing that present obvious opportunities for a focus on the writing process remain unexploited to this end or are still permeated by references to skills. Digital contexts are one such area of writing that could be used as part of the process discourse to simplify revision cycles (see Hyland 2003;Sokolov 2014). In the documents, however, references to the usefulness of digital software in writing are limited to error correction (skills discourse), and neglect opportunities for revising deep level textual elements (process discourse). Sensible employment of digital technology could be better used to steer EFL classrooms toward the benefits of process-oriented writing instruction, especially in times of school closings, in particular, but also of digital communication, in general, which is likely to increase even more in the future.
(3) Creative writing and the social practices discourses are underrepresented in the documents, both in the Bildungsstandards (KMK 2004a) and the Curricula.
Previous criticism of the Bildungsstandards (Porsch 2010b;Zydatiß 2005Zydatiß , 2008 warned that by narrowing language use to utilitarian purposes other creative, literary, or aesthetic facets might be left out. For writing in particular, this fear remains relevant as our findings reveal that creativity statements are scarce in the analyzed documents. Creative writing is frequently suggested to clash with the demands of standardized education, since it inherently is "anathema to systems based on control" (Maley 2013, p. 168). This might explain the lack of creative writing mentions in the present documents and parallels the findings by Müller et al. (2021) for writing in German. When creative writing is mentioned in the EFL documents, the statements are largely vague and frame creativity as a method to get a text product rather than as a process of personal development. Moreover, creative writing often occurs in hybridity with a predominant genre statement, which highlights the fact that creativity is not sufficiently explored for its full potential. In Ivanič's (2004) framework, creativity gives students the power to make their voices heard through writing as long as there is not an end goal constraining self-expression. Writing curricula from Germany thus contrast with other countries, where creativity constitutes an integral part of the conceptualization of writing and education in general, for example: New Zealand's curriculum, which foregrounds creativity and "personal growth through language" (Peterson et al. 2018, p. 511), and the Danish curriculum, which dedicates an entire section to "joyful writing" (Elf and Troelsen 2021, p. 181).
Writing competences which can be allocated to the social practices DoW are rarely mentioned in the documents, and when mentioned, they only concern generic vocational practices (e.g., completing application forms). While it is certainly necessary to prepare students for life after school, limiting the idea of students' social practices to situations of the professional world might not adequately represent their life experience. Engaging in authentic communicative settings constitutes one of the main ideas of contemporary CLT (Hiep 2007;Savignon 2002). Curricula should therefore broaden their ideas of social practices to include more diverse social settings. These could, for example, include communicative writing on social media, in online forums (Matz 2014), or gaming communities. In this way, identification with an authentic task may indeed be achieved and thus heighten students' sense of producing meaningful work.
(4) The sociopolitical discourse is absent from all policy documents.
Despite the KMK's (2004a) claim to promote students' intercultural ability to act, the Bildungsstandards, and subsequently the Curricula, fall short of explicitly addressing the sociopolitical dimension of writing students need in order to participate in societal and cultural discourses. In the sense of Klafki (2007) the German notion of Bildung implies the educational idea of fostering students' self-determination, co-determination, and the ability to have solidarity (p. 10) with others who may be denied fundamental rights. Similar to this idea, the central task and goal of communicative ELT is to prepare students for responsible participation in political, societal, and cultural discourses (W. Hallet 2011;Matz 2014). Therefore, we would expect the Curricula and especially the Bildungsstandards (KMK 2004a), which claim to set standards of Bildung, to include references to the sociopolitical DoW, and yet it is completely absent from both types of documents. For writing in particular, our analysis supports views of previous critics. Wolfgang Hallet (2011), for example, criticizes that the sociopolitical and sociocultural framing of teaching English, which was initially intended to guide the Bildungsstandards, is in fact not sufficiently apparent. Similarly, Zydatiß (2008) identifies a lack of sophisticated competence expectations for students' intercultural ability to act and suggests that the Bildungsstandards do not deliver on their promises.
Unfortunately, research on writing in a plethora of countries suggests that sociopolitical aspects of writing do not yet receive proper attention with US-American (Wilcox et al. 2021), and to some extent Danish curricula (Elf and Troelsen 2021), being exceptions. The fact that our findings for EFL writing align with those by Müller et al. (2021) for writing in the subject German suggests that this lack is also a widespread issue in Germany and might affect multiple language contexts. We must, therefore, raise the same concerns as Sturk and Lindgren (2019), who also observed a lack of the sociopolitical DoW in the Swedish context: "Children may leave school with limited experiences of how to make their voices heard, develop critical awareness toward text and information, and participate in social and political change" (pp. 525-526).
(5) Digital literacy is largely neglected in the documents.
References to writing in the context of digital literacy are limited to the use of software in error correction, leaving the potential of digital media in process-oriented writing development (e.g., Döring and Busse in press) largely untapped. However, in order to embrace digital literacy (see e.g., Belshaw 2012), teaching would have to move beyond the use of specific writing tools and enable students to practice critical thinking when engaging in online discourses, which are mostly written discourses that have become increasingly challenging to navigate (A. Hallet 2021). Students therefore need to learn how to be aware of their online consumption, consider information critically, and present their resulting views in a reasoned manner.
Although the need for increased promotion of digital literacy has been acknowledged in many areas of education, and thus to some extent also in recent policy documents (see KMK 2017 and accompanying state strategy papers on digitization efforts), there is as yet a lack of specific guidance for writing in this context. Although the companion volume to the CEFR (COE 2018) acknowledges the importance of online communication and interaction, clear specifications for writing are missing. Addressing digital literacy and its implications for writing should thus be a key task for policymakers in upcoming curricular revisions, which is particularly important in light of increased populism, false information, and ongoing heated debates in discussion forums during the pandemic.
(6) The analysis reveals varying discourse frequencies between Curricula for the different school types within one federal state and between identical school types in different federal states.
Although discourse frequency shows similar tendencies across state averages, Curricula within one state often show great variation (e.g., Bavaria's Curricula for Mittelschule n.d.c, and Realschule n.d.b). Inconsistencies in frequency are similarly pronounced for identical school types across states (e.g., Curricula for the school type Oberschule vastly differ in Saxony and Bremen). These substantial differences in the DoW underlying the Curricula suggest that, at least concerning writing, they are not yet aligned closely enough to ensure a smooth transition between tracks. Similarly, any move between states may be constrained by largely divergent emphases in the writing curricula of different states. Existing standardization efforts, indicated by such measures as the state-administered end-of-year-10 examinations or official lists of Operatoren, do not yet seem to have contributed sufficiently to writing curriculum alignment between states. Although further examination of additional sections in the documents is needed, our findings concerning writing support Bellenberg's (2012) claim that the detachment of school type and diploma, which was meant to ease movement between tracks and states, did not yet successfully result in compatible curricula.
As outlined by the six key findings above, our analysis offers important insights into the beliefs about writing that guide EFL documents in Germany. We pursued a novel approach as, to our knowledge, the DoW by Ivanič (2004) have not been used for systematically analyzing German EFL documents, despite their significant impact on writing pedagogy. There are, however, limitations to this study that we need to acknowledge. First, we focused on Curricula from only four federal states. Despite this small sample, the Curricula and school types selected constitute a demographically diverse sample, which provides a good overview of national documents for ELT. Future research should explore whether the observed tendencies can also be found in the other 12 state curricula. Secondly, the Curricula differ in the lengths of their proficiency statements, which in some cases might have only impacted their detailedness, whereas in other instances this may have led to more cases of hybrid discourses. In addition, other raters might include sections of the documents that formulate overall aims of ELT which we excluded from our analysis for not specifically addressing writing. Furthermore, vocational schools, such as the Berufskolleg, were excluded from the analysis since their pedagogical orientation may differ from general schools. Future research should also investigate whether the lack of attention to sociopolitical concerns, crucial for CLT, is a writing-specific issue or one that also affects other areas of EFL pedagogy.
Finally, our analysis is limited insomuch as the link between policy papers and actual classroom practice is mediated by a host of variables that require additional in-depth analyses not covered within the scope of this paper. In addition to researching teaching practices, a closer look at relevant textbooks (and other widely used teaching materials) might prove insightful, as they may function as a "secret curriculum" (Bieswanger 2008, p. 36). Yet as most states, including Bavaria, Bremen, and North Rhine-Westphalia from our sample, require textbooks to comply with curricular requirements, one may assume that curricula still have considerable influence on several aspects of language teaching.

Conclusions and implications for curricular development
In sum, our curricular analysis provides insights into the values regarding writing that underpin the examined documents. Previous analyses have mainly focused on writing in the regular language of instruction (for an exception see Jeffery and Akhmedjanova 2021), thus making more research into EFL documents necessary, especially considering the increasing importance of English in international communication. Overall, our study reveals that none of the documents covers all six DoW identified by Ivanič (2004), and that all documents are guided by a focus on skills and genre, which suggests a misalignment with both the current state of research on writing and the communicative turn in EFL pedagogy. The process discourse lacks full representation in the documents and, when mentioned, plays a minor role. As some of the analyzed policy documents are likely to be revised in the near future, we recommend that greater emphasis be placed on process-oriented writing, including revising and subsequent formative feedback, which are crucial for promoting K writing competence, especially for weaker learners in general and for FL learners in particular (Biber et al. 2011).
What is perhaps most concerning, however, is the relative absence of crucial aspects of writing, such as the creativity and social practices DoW and, especially, the sociopolitical discourse. Writing pedagogy in a democratic learning environment should recognize these dimensions of writing and encourage students to find their voice and claim their agency. Especially in the face of increasingly challenging public debates (see also A. Hallet 2021), young people need to learn how to critically examine information and express their perspectives in writing. Critical thinking in writing should therefore be given a more prominent position in future curriculum guidelines. Moreover, an effort to further standardize conceptions of writing and teaching writing across tracks and states would increase equity for students and should play a central role in upcoming revisions. However, standardization should not narrow the conceptualizations of writing in the documents to the demands of assessments. Instead, revisions should focus on fostering students' communicative writing competence by describing a comprehensive approach to writing that covers all six DoW and additionally addresses digital literacy.
Funding Open Access funding enabled and organized by Projekt DEAL.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4. 0/.
Conflict of interest K. Peltzer, L. Siekmann, J.M. Parr and V. Busse declare that they have no affiliations with or involvement in any organization or entity with any financial interest or non-financial interest in the subject matter or materials discussed in this manuscript.