SCHOOL LANGUAGE POLICIES: A CASE FOR RESEARCH-INFORMED, COHESIVE POLICIES

Language plays a crucial role in education; yet, on a daily basis teachers and school leaders must navigate a complex web of competing and often conflicting priorities in relation to various dimensions of language education. This includes supporting students’ first language and literacy development, sustaining the teaching of foreign languages in the curriculum and considering the role of other community or home languages spoken by students. However, while issues of language are undoubtedly relevant to all teachers, school-level language policies, which aim to provide explicit guidance underpinned by a clear set of principles, are too often conspicuous by their absence. For example, teachers are often left to their own devices to decide whether and how to allow space for other languages in the classroom; questions arise such as whether to use the first language in the foreign language classroom


INTRODUCTION
Language plays a crucial role in education; it is the means through which students access content knowledge across the entire curriculum and the medium through which they express themselves, negotiate understanding and are assessed in most subjects studied. Yet, on a daily basis teachers and school leaders must navigate a complex web of competing and often conflicting priorities in relation to various dimensions of language education. This includes supporting students' first language and literacy development, sustaining the teaching of foreign languages in the curriculum and considering the role of other community or home languages spoken by students. However, while issues of language are undoubtedly relevant to all teachers, school-level language policies, which aim to provide explicit guidance underpinned by a clear set of principles, are too often conspicuous by their absence. For example, teachers are often left to their own devices to decide whether and how to allow space for other languages in the classroom; questions arise such as whether to use the first language in the foreign language classroom, or whether to 'allow' migrant students to use their home language as a resource for learning. This paper, therefore, aims to provide an overview of some of the key theoretical perspectives on multilingualism in the classroom and to consider some practical implications for schools when developing language policies. While it is broadly reflective rather than empirical, it will draw on some key findings from a project which analysed school policies relating to language across a representative sample of 998 secondary schools in England. By reflecting on key points of intersection and divergence, a case will be made for teachers and researchers to collaborate to develop research-informed, cohesive school-level language policies which incorporate all dimensions of languages in schools.

METHODOLOGY Defining school language policies
To begin, it is important to set this discussion within the wider research context. The field of language planning and policy has constituted a vibrant area of research since the 1960s and, as noted by Hornberger (2006, p. 35), is very much "a field perpetually poised between theory and practice". While early research in the field tended to focus on large scale questions around language policy and planning in, for example, newly emerging nation states, the field has evolved to consider a wide range of domains including the workplace, the family and, crucially for this paper, the school. While there have been a number of influential frameworks within the field (e.g. Cooper, 1989;Spolsky, 2004), one which has been particularly pertinent in exploring issues of policy in educational research is Bronfenbrenner's (1979) ecological systems theory.  Bronfenbrenner, 1979).
As shown in Figure 1, the individual (here envisaged as a school student) is positioned at the centre. The student will have knowledge of one or more languages and will be influenced by their personal beliefs and motivations along with their broader experiences both in and out of school. This student is situated within the microsystem of the classroom with teachers and peers who may (or may not) share the same linguistic repertoire and where they will encounter different types of interactions during the learning process. Each classroom, then, is part of a wider school community with its own policies, priorities and demographics which, in turn is part of the broader education sector within a country.
It is often the case that individual teachers and researchers have little or no influence over these two outer systems, yet these systems inevitably play a large role in shaping language use in schools. At the macro level, decisions will be made about the official or national language(s) of a country which will be (implicitly or explicitly) informed by broader ideologies and the political landscape. For example, gaining independence has had huge implications for the language of instruction in schools in countries such as South Africa and Tanzania (e.g. Brock-Utne & Holmarsdottir, 2004), Smagulova (2021) similarly notes the effect of nationalist ideologies in Kazakhstan on the position of the Kazakh language in schools. In relation to ideologies regarding foreign language learning, the global status of English (and associated ideologies of internationalisation) often lead to prioritisation of English in many countries around the world while, conversely, Collen (2022) notes the negative effects of Brexit (the United Kingdom's exit from the European Union) on attitudes towards and uptake of foreign language study in England. Yet, while such macro-level concerns will inevitably shape government policy around education, such as decisions around the language(s) of instruction and the overall curriculum, this still leaves a lot for individual schools and teachers to decide at the meso and micro level which will shape day-to-day interactions and learning in the classroom.
The key point I wish to make here is to draw attention to the various dimensions of language which should be considered at these meso and micro levels. Firstly, it is important to acknowledge that a key priority for schools will be the development of students' language and literacy skills in their first language(s) which will often (but not always) be a subject in its own right and the primary language of instruction for most other subjects studied. This raises the question about whose responsibility it is to 'teach' the first language and to what extent this is devolved to all teachers, regardless of subject specialism. Secondly, a global rise in migration and transnational mobility has led to increasing linguistic diversity in schools worldwide. This raises questions not only of how to best support newly arrived migrant students to learn the language of instruction (as an additional language), but also about the extent to which their home language(s) should be explicitly acknowledged and supported in schools. At a pedagogical level, teachers are often left to their own devices to decide whether students should be 'allowed' to use other languages in the classroom and how to manage this effectively to create the best possible learning environment for all -this can result in high levels of inconsistency, even within the same school (e.g. Liu & Evans, 2016) which can lead to confusion for students and teachers alike. Thirdly, foreign language learning is a core part of the curriculum in many educational contexts worldwide and this raises questions for schools about which language(s) should be offered and the amount of curriculum time to be allocated. While predominantly Anglophone countries such as the United States, United Kingdom, New Zealand and Australia have been experiencing a decline in foreign language study, other countries are beginning to look beyond English as the 'default' foreign language, for example, in recent years China has seen an increase in the study of languages other than English at both secondary and tertiary level (Gao & Zheng, 2019).
While each of the dimensions of language outlined above are important for schools, some may be neglected, while others may be considered in isolation rather than as part of a wider school language policy. Defined as "an action statement outlining the solutions necessary for addressing the diverse language needs of a school" (May, 1997, p. 229), such policies provide an opportunity to create a contextually-appropriate, cohesive vision for schools which not only outlines key principles in relation to languages, but also suggests practical strategies for achieving this. However, while knowledge of the specific school context is crucial for developing such a document, so too is an understanding of research perspectives around multilingualism in the classroom.

Theoretical perspectives on multilingualism in the classroom
One of the key debates in the field of relevance to school and classroom-level language policies has been around the role of the first language (L1) 2 . This has relevance not only for (foreign) language teachers, but for all teachers in linguistically diverse schools. The aim of this section is therefore to provide a brief overview of some of the prevailing theoretical perspectives which suggest the value of (purposefully) drawing on a range of linguistic resources in the classroom to support learning.
It is important to acknowledge, however, that such views have not always been accepted. Indeed, early views of language learning in the 1950s and 1960s were influenced by theories of behaviourism which saw language learning as the formation of habits developed through exposure and repeated reinforcement. While this could be considered as more straightforward for L1 acquisition, when it comes to the learning of foreign or additional languages such theories often lead to the L1 being suppressed for fear it will 'interfere' with learning the new language. Such views are still evident in many classrooms today where use of the L1 (either among foreign or additional language learners) is ignored, or indeed, actively forbidden. Yet, it is important to remember that multilingual learners are not a tabula rasa and that foreign or additional language learning is fundamentally different to learning the L1.
More recent theoretical perspectives have therefore evolved to reflect this and provide space to explore the affordances of a wide range of linguistic resources in the classroom, three of which will be briefly presented here: cognitive processing theory, sociocultural theory and translanguaging. Cognitive processing theory (e.g. Ellis, 2006) posits that the L1 and any subsequent languages learned are not contained in separate conceptual stores, but that we draw naturally on our knowledge of all languages in our repertoire as a 'resource'. A particularly influential and well-established concept within this is Cummins ' (1979, 2000) linguistic interdependence hypothesis, which suggests that while the surface level aspects of different languages are clearly distinct (e.g. in terms of pronunciation, orthography or syntax), there is an underlying cognitive proficiency that is common. This suggests that certain knowledge and skills from one language can be positively transferred to the learning of another language (particularly in relation to literacy), provided a certain level of proficiency has been reached. This positions the L1 not as a source of interference, but as a potential resource. Indeed, evidence from a more recent study by Forbes (2020) suggests that even beginner or low proficiency foreign language learners can develop effective skills and strategies in the foreign language classroom which can positively influence writing in other languages (including their L1). Positive effects can be further enhanced where collaboration between first language and foreign language teachers is supported by local school policies.
In a similar vein, one of the main areas of inquiry in sociocultural theory (which stems from the work of Vygotsky, 1962) has concerned the question of how language is used as a semiotic tool to regulate ourselves and others. As an intrapsychological tool (i.e. a tool for self-regulation), the L1 mediates what we do through inner voice and private speech (i.e. it is often the means through which we think and process information). As such, even if explicit use of the L1 is 'banned' in the classroom, it is nevertheless always present, whether or not we acknowledge it. As an interpsychological tool (i.e. a tool for regulating others), the L1 can be used for the purpose of scaffolding by the teacher and/or peers. Here, the 'more knowledgeable other' can draw on a range of linguistic resources to guide the learner and help them to build on prior knowledge more effectively than they would be able to do alone. This may take the form of cognitive mediation, for example, using a shared L1 to help learners make connections to concepts they may have already learned through translation (e.g. understanding photosynthesis in an English-medium biology lesson), and also affective mediation, for example, using a shared language to establish positive relationships in the classroom.
Another key perspective which highlights the affordances of a multilingual classroom is that of translanguaging. The concept of translanguaging was first introduced by Williams (1994) in the context of bilingual Welsh and English classrooms to refer to refer to the planned and systematic use of one language for input (i.e. listening or reading) and another for output (i.e. writing or speaking). While definitions have evolved over time, at its core, translanguaging remains a practice that involves the "dynamic and functionally integrated use of different languages and language varieties" (Li, 2018, p. 15), for example, when learners switch between languages or draw on their full linguistic repertoire to convey meaning. This could include both spontaneous translanguaging (i.e. the fluid use of languages both in and out of school) and pedagogical translanguaging (i.e. designed instructional strategies that integrate two or more languages) (Cenoz & Gorter, 2017). Such practices are underpinned by the idea that multilinguals (or those in the process of learning another language) "do not think unilingually in a politically named linguistic entity, even when they are in a 'monolingual mode' and producing one nameable language only for a specific stretch of speech or text" (Li, 2018, p.18). The underlying implication, therefore, is that having knowledge of multiple languages may influence the way in which someone uses or thinks about language more broadly, including their L1.
While this section has only presented a brief overview of some of the key theoretical perspectives on the role of languages in the classroom, what emerges is a strong sense that the L1 (and, indeed, other languages in a student's repertoire) will always be present in the classroom, regardless of whether or not they are explicitly 'allowed'. As such, these languages, when drawn on in a principled and purposeful way, can constitute an important resource and 'stepping-stone' for students to enable them to engage fully with the learning process. However, while there is a growing evidence base in support of this in the applied linguistics literature, this has not yet necessarily fed through to informing decisions made at the level of schools or classrooms. Indeed, in a range of educational contexts school-level language policies are frequently absent and, where they do exist, are often underpinned by monolingual ideologies which do not reflect the multilingual reality of schools today (e.g. Flynn & Curdt-Christiansen, 2018).

Real-world reflections: evidence from analysis of school language policies in England
To further illustrate the above point, this section will present some preliminary findings from a recently completed scoping study conducted in England (Forbes & Morea, under review). The aim of the study was to explore the provision of school-level policies from a representative sample of secondary schools in England and to analyse the extent to which they address the three dimensions of language outlined above (i.e. English -as both a first language and additional language, foreign languages in the curriculum, and other home or community languages spoken by students). In order to compile the dataset, the research team searched the websites of 998 secondary schools (a representative sample of 20% of secondary-level schools across the country) for any documents or relevant webpages which related to any of the above dimensions of language. The final dataset comprised 1,457 separate policy documents.
Overall, it was found that fewer than two thirds of the schools had local policies which made any reference to language whatsoever and those references which were made were largely dispersed across a wide range of policy types. The most common policy types where language was mentioned were special educational needs policies (26% of schools), equality and inclusion policies (23%) and curriculum policies (20%). Despite the fact that, on average, around 20% of students in schools in England speak English as an additional language (EAL), only 6.6% of statefunded schools in our sample had a specific EAL policy. Another notable finding was that only six of the schools had any form of dedicated whole-school language policy. This compartmentalisation of references to language across the various policy documents led to a number of tensions and contradictions.
One of the most common contradictions seemed to stem from the tension between a desire for schools to promote an ethos of multilingualism and the pressure to maintain high standards in English (which often led to the active suppression of multilingual practices). For instance, this was evident in statements such as "all languages, dialects, accents and cultures are valued; however, we aim to teach standard English". There was also huge variation in positions regarding home language use in the classroom, with some policy documents actively encouraging EAL students to draw on their home language to support learning, while another went as far as to suggest that students would be "reprimanded" for speaking in a language other than English.
Another key contradiction we found related to policies around the teaching of foreign languages; while language learning was often framed as a provision 'for all' (in line with being a statutory part of the national curriculum between the ages of 7-14), this did not necessarily equate to being an entitlement for all. This was evidenced by explicit policies to 'disapply' students from language learning found across 117 schools in the sample, in favour of providing additional support for "life skills", "literacy" or, in some cases, simply where students found languages to be "a challenge". Given the growing body of evidence which suggests that learning foreign languages can actually improve literacy skills and meta-linguistic awareness in the first language (e.g. Forbes, 2020;Murphy et al., 2015), such disapplication policies are concerning. What emerged, therefore, were a number of missed opportunities to develop more joined-up and cohesive thinking in relation to languages.

Implications for schools and researchers: towards developing a cohesive school languages policy
In spite of the tensions raised above, the sample contained a small number of encouraging points of intersection between the various language dimensions within policy documents, for example in drawing attention to the role that other languages (either taught or learned at home) can play in improving skills in English. While these statements did not typically come with specific pedagogical strategies to support such connection-making, they provide a useful starting point for thinking about how schools could develop a more cohesive and holistic whole-school languages policy.
It is important to note that the scoping study reported above represents only the first step of a broader project which will involve establishing partnerships between schools and researchers to develop research-informed guidance. As noted by Vanbuel and Van den Branden (2021, p.219), effective reform in relation to school language policies should be "local, context, and schoolspecific, with the school as the 'key site' for educational improvement". Yet this also needs to be informed by a broader understanding of linguistic and multilingual practices in the classroom.
Combining the context-specific and pedagogical expertise of teachers with the theoretical and empirical understanding of researchers is therefore crucial to the effective development and enactment of school-level language policies. However, it is important to acknowledge the challenges involved in such a process. These include, but are not limited to, the need to work within the (often rigid) constraints of broader macro-or exo-level policies, a lack of agreement among teachers about priorities and practices in relation to language, and overcoming the compartmentalisation of subjects which is common at secondary level.

CONCLUSION
Nonetheless, in light of the importance of language in schools for all, as noted in the introduction to this paper, I believe that there is the potential for an effective, cohesive, wholeschool language policy to support both teachers (in making day-to-day decisions about pedagogy) and students (in developing their own linguistic skills and agency). To do so, there is a need to move away from existing silos which consider policies related to the first language, foreign languages or home languages individually and to consider the ways in which these various language dimensions (may) interact. By doing so, the aim is to encourage more joined-up thinking which may, in turn, support students' language development and learning across the curriculum.

FUNDING
The data reported on school language policies in England was supported by a British Academy Small Research Grant [SRG21\211295].