Policy formation for adult migrant language education in England: national neglect and its implications

This article is about current policy in the coordination of opportunities for adult migrants in England to learn English. People who move to a different country experience a need to learn the dominant language of their new environment, to support their settlement. A willingness to learn the language is a marker of social inclusion from a political perspective too: an insistence that migrants have an obligation to learn and use the language is a recurrent trope in political and media discourse. In the UK, language education for adult migrants focuses on the area of education known as ESOL, English for Speakers of Other Languages. Beyond the rhetoric, policy support for migrants’ learning of English across the UK is inconsistent: there is neither a UK-wide nor an England-specific strategy in policy to support access to ESOL. Where policy exists, it is formed at a local level in the absence of national direction. The aim of this paper is to consider how an important area of adult education appears to have little presence in national policy, and what the implications of this are, for practice. To achieve this, we first follow the trajectory of ESOL policy in England, considering why—despite attempts to address its coordination—there remains a lacuna. Second, we ask what the implications are of this policy gap for ESOL coordination in practice. Analysis of current policy and of interviews with key ESOL stakeholders suggests an enduring condition of fragmentation and lack of coordination to the detriment of students.


Introduction
This article is a critical exploration of the formation and enactment of policy in adult migrant language education in England. Here, as with elsewhere in the UK, language education for adult migrants is largely synonymous with the teaching and learning of English and is commonly known as ESOL (English for Speakers of Other Languages). ESOL is an area that has experienced policy neglect over a sustained period of time. Through a study of national ESOL policy formation we explore the reasons, and through an examination of interviews with ESOL stakeholders we consider the implications for potential students and practitioners. The neglect of ESOL in policy calls for urgent attention at a time when the official environment is inhospitable, and when anti-immigration sentiment at the heart of government is on the rise. For example, at the time of writing it remains Government policy to deport asylum seekers in the UK to the central African country of Rwanda (Home Office, 2022), with the aim of discouraging others to attempt to travel to the UK. There is little support in government for policies to improve the lives of migrants who do come, or a willingness to take responsibility for this. Yet migration and multilingualism shape British society: in 2021 40.6% of the population of London was born overseas (ONS, 2022). Provision of language learning opportunities for adult migrants is therefore not a minor matter affecting just small numbers of people. Moreover, disregard for ESOL is an issue with broadreaching implications for other areas of policy such as employment and public services, not to mention for the well-being of migrants and their families across their lifespans. This paper represents an effort to contest the marginalisation of newcomers, and to point to the role of the field of ESOL and its coordination in enabling them to realise their potential, for the benefit of society as a whole.
In dominant discourses about language and migration in most places, a central ideology holds that for society to be cohesive and stable, its population must share and use one common language. The UK is no different, and the orientation towards language education for adult migrants in practice and in policy is firmly towards the learning of English, and-more recently-of Welsh in Wales (Higham, forthcoming). People who move to a different country thus experience a need to learn the dominant language of their new environment, for their wellbeing, for their independence in society, for their employment and to access services, and generally to support their settlement (Refugee Action, 2017;Court, 2021). Access to the dominant language, alongside the right to maintain the language in which one grew up, is also a linguistic human right (PEN, 1998: Article 13.1). A willingness to learn the language is a marker of social inclusion from a political perspective too, and an insistence that migrants have an obligation to use English has been a recurrent trope in political and media discourse for many years (Simpson, 2019(Simpson, , 2021. Beyond the rhetoric, however, policy support for migrants' learning of English across the UK is inconsistent. The field of ESOL is reactive to changes in patterns of migration, often rubbing up closely and uncomfortably to national policy around immigration, citizenship and social integration (Cooke & Peutrell, Policy formation for adult migrant language education in… 2019). Its treatment at national policy level follows divergent paths in the different countries of the UK. While the governments of Scotland and Wales have developed explicit and funded policies to support and coordinate ESOL provision, no such strategy currently exists for England. Historically, the position of ESOL in policy has been marginal even within further and adult education, itself sometimes referred to as the Cinderella Sector (Daley et al., 2015). ESOL classes that happen in Colleges of Further Education (FE) are viewed as the mainstream (Baynham et al., 2007), yet even here ESOL is somewhat fragmented, provided through multiple funding streams, including the shrinking adult education budget (Migration Yorkshire, 2021) and other sources which are less stable and more short-term. Access to funded classes is subject to complicated rules on eligibility relating to immigration status and income, and even for those who are eligible, waiting lists are long (Court, 2021). Hence an increasingly high proportion of ESOL provision lies outside the mainstream, in the hands of an array of third and voluntary sector providers, supported through small grants and project funding, and the private sector.

Research questions and outline
This paper addresses how the picture of poor coordination and piecemeal funding emerged, and what the implications are, from the perspective of ESOL stakeholders. The focus is the recent historical context, the development of ESOL policy nationally and regionally from the point in 2009-2010 at which ESOL was left outside any national coordinating framework, and when the field experienced an acceleration of funding cuts. Recognising from the outset that there is currently no explicit strategy for ESOL in England at national scale, we pose two questions relating to ESOL policy: 1. Where is ESOL as an area of adult education located in current government policy in England? 2. How do ESOL stakeholders experience government policy in ESOL in relation to its coordination?
The paper progresses as follows. Below we elaborate on notions of structure and agency in policy formation to provide a foundation for later discussion. We then address our research questions. We first provide a historical background of ESOL pre-2010, before examining more recent policy documents, to establish the trajectory of the field of ESOL in national-level policy to the present day. We then present a summary of an analysis of a set of semi-structured interviews with ESOL stakeholders to explore the implications of current policy for the coordination of the field. In our concluding discussion we consider how an understanding of ESOL policy formation in terms of structure and agency in dynamic tension is not helpful when crucial aspects of structure appear to be missing, and end by asking whether and how policy for adult migrant language education might get formed and enacted locally and regionally, in the absence of direction at national level.

Policy formation in ESOL: structure and agency
Practitioners and providers in adult migrant language education, and professionals in allied areas with a stake in the field (e.g., housing, employment, health), need a supportive policy structure so that they in turn can support the students and potential students for whom they have responsibility. These include people in society most at risk of social exclusion: refugees, asylum-seekers and others who have experienced forced migration. Policies that are important include those that relate to the curriculum and materials (i.e., what is taught), teacher education and training, qualifications and assessment regimes, funding, and associated issues of coordination. This is our concern: how provision is organised, what provision is available in a place, how students actually find a class, and how providers connect with each other and to other areas of education, training and beyond.
Varied histories lead to adult migrant language education being understood, experienced and supported in different ways around the world. There are commonalities: in most places where national language policies for the field exist, importance is placed upon teaching and learning the new language for promoting participation in society, and for addressing the communicative needs of new arrivals. These policies tend to align with a common ideological position on the role of one or a small number of national languages in maintaining the strength of the nation state and for promoting homogeneity and social cohesion (Pöyhönen et al., 2018). Simpson and Whiteside (2015) contrast different approaches to state-level policy creation and interpretation for migrant language education. These range from the well-resourced and supported, for example in Australia (Nicholas, 2015) to the fragmented and under-funded, as is the picture of ESL provision at national and state level in the US (Wrigley, 2015; see also Feuerherm & Ramanathan, 2016). In the UK, education policy is devolved, and the governments of both Wales and Scotland have developed explicit strategies to underpin their approaches to ESOL, for example to bring together ESOL providers from across the sector, and to address funding and qualifications at home-nation (rather than UK) scale. The ESOL Strategy for Wales (Welsh Government, 2019) emphasises the role of Welsh as well as English in public life, and promotes a bilingual approach to ESOL provision. Scotland's ESOL Strategy (Education Scotland, 2015) is notable for being collaboratively written in consultation with learners.
A familiar conceptualisation of language policy formation and development distinguishes between macro, meso and micro levels (Baldauf, 2006;Marriott, 2006;Zhao & Baldauf, 2012). For a national policy on ESOL coordination, an idealised version of this model would entail a framework or strategy being formed by central government at the macro level, written up in official documents, enacted and implemented by the range of institutions that provide ESOL at an intermediate or meso level, and interpreted or otherwise appropriated by practitioners (teachers, volunteers) on the ground, at micro level. This, however, can mislead, suggesting as it does a linearity and coherence in policy creation and enactment, a sense of something being designed and created at the top and then passed down the levels in turn. Recent work in language policy research contests Policy formation for adult migrant language education in… this simple image (Martin-Jones & da Costa Cabral, 2018;Tollefson & Pérez-Milans, 2018). Far from being unified actors, governments and states prove themselves fragile and heterogeneous assemblages (of people, practices, technologies, texts, discourses etc.). Policy texts encounter countervailing forms of power and agency throughout the networks in which they circulate from their instigation, hence disrupting their development and uptake. Nonetheless a model of policy that recognises a hierarchy points to important differences in the resources and authority of national versus local institutions. Therefore it is still relevant to a normative assessment of accountability. Concerning ESOL coordination in England, at the macro, national government level responsibility has been abrogated, and no coherent overall strategy for the field is articulated.
It is of course well-established that language policies that are communicated topdown by centralised authorities are also appropriated, subverted and interpreted in new ways by practitioners and those on the ground (Johnson & Johnson, 2015). Ethnographically informed studies of language policy demonstrate that policies themselves can emerge in local contexts of practice: in such cases language policies can be regarded as processes (Ricento & Hornberger, 1996) and as locally situated sociocultural practice. Indeed, as Johnson (2013) and others have suggested, the meaning of a language policy is not just derived from a policy document; it also emerges within discourse at one layer and 'across multiple layers of language policy interpretation and appropriation' (Johnson, 2013: 119). This explains, as Johnson and Johnson found, how nominally identical educational programs, funded under the same language policy, 'end up being different in practice ' (2015: 221). The idealised model sketched out above therefore also masks the agency of social actors at meso and micro level, i.e., their capacity to make free choices and take individual action. Social theories of agency rest on a dualism of agency, the capacity to act, as well as structure, the social relationships within which action occurs: agency exists in a symbiotic and dynamic relationship with the structure within which it might be claimed. Hence as Liddicoat and Taylor-Leech (2021) note, agency is 'the ability of individuals to influence their contexts rather than merely react to them'.
As the same authors say, policy can be reinterpreted, then, or even rejected as a form of resistance (Giddens, 1984;Ortner, 1984) or as an exercise of choice (Pickering, 1995). Studies of grassroots language policy formation (McCarty, 2011) suggest that the rejection of what is imposed top-down in favour of the development of a policy from below can be both desirable (particularly when what is imposed top-down is highly contested on ideological grounds) and feasible. Examples of imaginative initiatives in ESOL in the UK demonstrate this: advocates of socialjustice approaches to pedagogy can develop them locally, in the name of emancipation and the exercise of voice, when fissures in structure open up for them to do so. This was the case following the end of the Skills for Life policy in England in 2009, which we discuss below, the concomitant abandonment of a centralised curriculum with prescribed teaching materials, and the development of alternative participatory, student-centred approaches. The project Whose Integration? (Cooke et al., 2015) involved teachers and a university researcher working with ESOL learners in London to challenge the circumstances of the students' potential social exclusion. Our Languages, a multilingual approach to ESOL pedagogy in London led by the same 1 3 team (Cooke et al., 2018), had as its basis a recognition that a shared language is vital to social life, but linguistic diversity is also central. 'Both can be enhanced by education, enriching both the individual and society' (https:// ourla nguag es. co. uk). The organisation Heart and Parcel (https:// heart andpa rcel. org) supports women living in Greater Manchester to settle and connect across communities by developing English language through the medium of food. Beyond the Page (https:// beyon dthep age. org. uk) brings women from different backgrounds together-as they put it-to break down the barriers of language and cultural difference (Macdonald & Watson, 2022). Enterprises such as this, often taking place outside formal places of education, in community-based centres and with third-sector organisations and NGOs, demonstrate that-as Hornberger (2020: 122) says-'even when top-down policies begin to close ideological spaces, implementational spaces carved out from the bottom up can wedge them open.' These initiatives do not address our specific focus, the (lack of) coordination of language education for new arrivals. Ideological positions firmly held in political rhetoric regarding the need for migrants to learn English are not accompanied by a national strategy that supports coherent structured opportunities for them to do so. There is evidence to suggest that where adult migrant language education operates outside state education systems, that is, where there is an absence of guidance and support from above, policies on coordination can be successfully developed locally and bottom-up, involving cooperation between ESOL providers and other stakeholders. Feuerherm and Oshio's (2020) description of a grass-roots partnership approach to ESOL policy formation at a city scale in the US bears similarity to local responses which exist in some cities in England, notably Manchester, Nottingham and the boroughs of Hackney and Newham in London. These are few and far between though, and are not part of a coherent system. There is just one serious coordinating effort at regional scale, Learning English Yorkshire and the Humber (LEY&H: www. learn ingen glish. org. uk), which developed from the work of a partnership instigated and led by this paper's first author in the city of Leeds when the regional coordinator for refugee resettlement incorporated ESOL provision into its larger and long-term refugee support programme. LEY&H, in addition to providing a dynamic directory of provision, also has a space for learning and teaching materials, runs regular training events and seminars for tutors, and hosts a virtual staff-room. In other regions and at national scale, ESOL in England exists in a state of fragmentation.
Later we attend to the views of stakeholders on a picture of a lack of coordination and of fragmented provision in practice, one where structure is weak or lacking. People do not act in isolation from each other but are part of a complex constellation of agents (Mayntz & Scharpf, 2001) who, through action and discussion, mediate policy or respond to a lack thereof. Badwan (2021), in a study of the agency of stakeholders in another language policy context of neglect (in her case in Higher Education in Tunisia), explains how agency can be exercised, rejected and contested at an individual level. She concludes that while individual agency enables flexible responses to changing local circumstances, it can also cause problems such as 'inconsistency, uncertainty, and the reproduction of social inequalities ' (2021: 99). In other words, the absence of top-down policy can be seen as a 'double-edged Policy formation for adult migrant language education in… sword'. When we examine stakeholder perspectives on ESOL coordination in England, we suppose that the same pertains in this neglected context.
So we turn to our two questions. In the next section we answer the first, concerning the location of ESOL in policy in England, by tracing its trajectory. In the section after, through an examination of interviews with 20 individuals with a stake in ESOL in England, we explore how they experience government policy for the field with reference to its coordination.

ESOL in national policy in England
We place the current picture of policy on ESOL coordination in its historical and political context. First we provide a brief history of ESOL in England up to the end of the first decade of the century. Then-through an analysis of relevant major policy texts-we look in more detail at the policy moves from 2010 and the election to power in that year of the Conservative/Liberal Democrat coalition government. We highlight the policy-related documents that refer to some kind of national strategy for ESOL since, noting in particular how language learning for migrants has become aligned with discourses about their social integration. Our critical examination of ESOL policy formation in this section aligns with Tollefson's historical-structural approach (1991), in that our gaze is more upon language policies themselves and their power, and less upon the agency of specific policy actors (see also Johnson, 2018: 62).
The field of ESOL emerged in the mid-twentieth century, in concert with the arrival in the UK of migrants from former British colonies-particularly the Indian sub-continent and the Caribbean-who had a right to settle in Britain in response to the demand for labour following World War II. The language learning needs of these migrants were typically addressed ad hoc, and classes were organised on a voluntary basis. In the late 1970s and 1980s, provision of English as a Second Language (ESL), as it was still then known, became more organised and better funded, with classes in Adult and Further Education colleges and workplaces (Cooke & Simpson, 2008;Rosenberg, 2007;Simpson, 2015). More recent migration differs in range and scale from the earlier post-war mobility, at least in part due to processes of globalisation associated with late modernity (Appadurai, 1996;Giddens, 1999;Vertovec, 2007). In reaction to larger numbers of people from across a spectrum of places and backgrounds, the field of ESOL grew, achieving a central place in England's adult education policy for the first time in the early 2000s, under the New Labour government elected in 1997. A review of basic skills (the Moser Report, DfEE, 1999) recommended implementing a national strategy, Skills for Life, to reduce the number of adults with low levels of literacy and numeracy. Skills for Life was overseen by the Department of Education and Skills, and from 2007 by the newly-formed Department for Innovation, Universities and Skills. ESOL was incorporated into Skills for Life in 2001, reflecting a political and indeed societal preference for highly-skilled over low-skilled migrants, and indexing the nation's utilitarian perspective on migration and its close association with labour market policy. The inclusion of ESOL as a 'skill for life' brought with it the creation of a statutory national curriculum, classroom materials to support the curriculum, teacher-training and inspection regimes, and qualifications mapped against national standards. It can therefore be viewed as a moment where a government took macro-level responsibility for many aspects of ESOL in England. We note though that ESOL's incorporation into the Skills for Life framework only came about through sustained lobbying by grassroots activists in the sector, indicating the complexity of policy formation at national scale.
Demand rose fast for expanded ESOL provision, particularly after the eastward expansion of the European Union in 2004 and migration to the UK from the new accession countries (Mallows, 2006). In 2006, enrolments in government-funded Skills for Life ESOL classes peaked at 500,000, and in 2008 funding under the Skills budget reached £300m (Paget & Stevenson, 2014). In 2006 an enquiry on ESOL, More than a Language, carried out by an advocacy body for adult education, NIACE, noted the high cost of English language provision, and from this point government support for ESOL began to decline. In 2007 ESOL fee remission became restricted to 'a needs-based eligibility criterion' (Paget & Stevenson, 2014: 38). Funding for classes in FE colleges was to continue in reduced form, still administered via Skills for Life by the Skills Funding Agency. An increasing amount of provision now took place outside formal structures though, filling the spaces left by the withdrawal of mainstream courses because of this lack of funding. The New Approach to ESOL (DIUS, 2009) required ESOL outside Further Education colleges to be coordinated at the level of local authorities and councils. The result was that the Government relinquished both immediate responsibility for, and control of, much of the field. The election of the Conservative-led coalition government in 2010 brought with it a period of 'austerity', a deficit reduction programme consisting of cuts to public spending including local government funding. This severely compromised local authorities' ability to fulfil their new obligation to coordinate English language provision for adult migrants. Within FE, ESOL and its funding remained the responsibility of DIUS and from 2009 the Department for Business, Innovation and Skills (DBIS, 2009(DBIS, -2016, but outside FE, there was no clear oversight. Thus much of the field was left in a state of fragmentation (Simpson, 2012): gaps in provision were filled by voluntary and community bodies and the private sector, operating outside any formal coordination structure.

ESOL in national policy since 2010
The effect on the field of ESOL of severe and sustained funding cuts coupled with a lack of coordination was profound. Soon after the change of government in 2010, in work led by the first author, a study of local ESOL provision in the northern English city of Leeds (HENNA, Simpson et al., 2011) highlighted the incoherence of provision locally and city-wide. For the researchers, 'an overarching conclusion is that the erosion of the cohesive framework afforded by Skills for Life is likely to lead to a return to the fragmented picture of ESOL provision of previous times; [and] the pattern of multiple funders and combinations of providers and centres is likely to remain characteristic of ESOL under the proposed new funding regime.' They noted Policy formation for adult migrant language education in… that this, combined with the absence of sustained funding for many providers, raised questions of continuity, consistency and quality of tuition for the benefit of students. For example, pathways through learning need to be meaningful, clear and coordinated for it to be successful, yet the 'lack of continued and stable funding streams disrupts progression routes both between ESOL courses of different levels, and from ESOL into training and work ' (2011: 5).
From around this time, mention of English language education for adult migrants in policy in England became associated less with the discourse of skills and more with a longstanding ideology of linguistic homogeneity evident in political rhetoric. Hereby, multilingualism is regarded as a 'problem' and something that must be 'managed' (Hogan-Brun et al., 2009), the learning of the English language is seen as a prerequisite for integration and social cohesion (Blackledge, 2006;Cooke & Peutrell, 2019;Cooke & Simpson, 2012;Simpson, 2019Simpson, , 2021, and the failure of a migrant to learn English is considered an emblem of an unwillingness to integrate, a failure to pay the proper 'debt of hospitality' (Vigouroux, 2017). The co-option of ESOL into discourses of homogeneity was associated with a general hardening of the stance towards migration following the election to power of the Conservative-led coalition Government in 2010. A commitment to reduce net migration first featured in that year's Conservative Party manifesto. In 2013 Theresa May, as UK Home Secretary, introduced a new Immigration and Naturalisation Bill, highlighting that policy creates categories of migrant, who can then be treated in law in certain ways according to the category that they happen to fall into. Among other things, the purpose of the new Bill was 'To make provision about immigration law; to limit, or otherwise make provision about, access to services, facilities and employment by reference to immigration status' (UK Government 2014). May's aim for the bill was to create-in her words-'a really hostile environment for illegal migrants' (Kirkup & Winnett, 2012).
The notion that knowledge and use of English is the answer to problems with integration and social cohesion is present in the key documents that have contributed to shaping current policy on ESOL in England, to which we turn now. The analysis that follows focuses on the documents that punctuated policy debates around the field at national level between 2014 and 2020 that either call for some kind of strategy for the field or set out Government plans for its coordination. The texts are: Our choice to focus upon these particular texts relates to their prominence in public debate, indicated by the amount of discussion about them in the education press (Exley, 2017), in the news area of the website of the ESOL teachers' organisation NATECLA, and on the online forum ESOL-Research. The intertextual 1 3 and interdiscursive connections across the documents are clear. The lexical links between them are especially prominent, as the texts draw progressively closer to actual government policy formation. Particularly notable is the repetition of those lexical items that relate to a discourse that was dominant through this period, that of integration. Within a text, lexical cohesion is considered a dimension of textual cohesion (Halliday & Hasan, 1976), which can be identified through sequences of related words (lexical chains). Between and across texts (i.e. intertextually), as Blackledge (2005) describes, policy formation might be studied through the tracking of discourse chains. In his research into the UK's 2002 Nationality, Immigration and Asylum Act, Blackledge notes that discourse about language and migration is recontextualized and transformed in increasingly legitimate contexts, gaining authority as it travels. An ideology which 'clearly privileges the English language above the other languages of England' is 'most strongly evident as argument moves closer to the centre of Government' (2005: 225), until it becomes enshrined in law. A similar process is at play here, with the progressive cementing in political discourse of the relationship between English language education for adult migrants, integration and social cohesion.
Before we go on, we should note that UK Government documents on integration vary in their focus. For example, the Casey review (2016) encompasses all of Britain (and occasionally Northern Ireland) and refers to British society and Britishness, whereas the Integrated Communities Strategy (2018) also refers to integration into Britain and British values, but presents proposals for England (Court, 2021: 33).
The fractured state of ESOL provision post-Skills for Life soon became widely recognised. In 2014 the think-tank Demos published a book-length report, On Speaking Terms (Paget & Stevenson, 2014), which would be widely cited in future arguments for supporting the field. The report highlighted the 'paradox' of 'an identifiable ESOL need and the withdrawal of state support' (p. 11). 'Current ESOL policy,' argued the authors, 'suffers from fragmentation, a lack of clarity about the aims and intended outcomes of learning, disagreement over the analysis and description of English language levels and abilities, and a general tendency to take a short-term view' (p. 5). A national ESOL strategy would address the situation, and in calling for this, the authors' argument foregrounded integration and social cohesion: 'A coherent ESOL policy should be fit to … promote a more integrated and socially cohesive society' (pp. 9-10). This position was elaborated in the conclusions of the report, with a three-point rationale for supporting ESOL in policy (p. 81): it saves public money, it benefits the economy, and it promotes social integration ('the desirability of a better integrated, more socially cohesive society').
In the 2015 general election the political scales tipped further to the right, with an increased majority for the Conservative Party, and the anti-immigration discourse of the hostile environment became more prominent in policy and in the media. Here, a campaign of misinformation about migration was fought by sections of the national press in the run-up to the June 2016 Brexit referendum. Front page headlines such as 'Britain is a Migrant Magnet', 'We Must Stop the Migrant Invasion' and 'Britain Must Ban Migrants' (all from the right-wing anti-EU newspaper the Daily Express) indicate how unpleasant the tone of the debate was. Media rhetoric, and the pandering to it by politicians, doubtless played a role in the outcome of the Brexit vote. By the time of the referendum, the idea of leaving the EU had become associated with discontent, fear and anxiety about immigration, stirred up by the media and exploited by right-wing populist but increasingly mainstream politicians, not an uncommon situation across Europe at the time (Wodak et al., 2013). Following its election, the Conservative government commissioned Louise Casey's Review into Opportunity and Integration (2016). The Casey review linked social and economic exclusion to lack of access to the English language, termed the 'common denominator' (p. 94). Insufficient competence in English was positioned firmly as a social problem connected discursively to crime: 'Central and local government should develop a list of indicators of a potential breakdown in integration. These might include incidences of hate crime or deficiencies in English language' (p. 167). Prominent in the Casey review were Pakistani and Bangladeshi Muslim women, seen as being uniquely challenged and problematic because of cultural, religious and social barriers to integration, including their failure to learn English, competence in which was consistently framed in deficit terms. 'Pakistani and Bangladeshi ethnic groups,' says Casey (p.14) 'have the lowest levels of English language proficiency … and women in those communities are twice as likely as men to have poor English.' Othering, negative stereotyping most often associated with race, ethnicity, foreigners or minority groups (Holliday, 2005), is strongly evident in monolingualist discourse such as this, where a language-based shortcoming is identified in the Other. The process of othering, i.e. defining the 'problem' in linguistic, cultural and religious terms, deflected attention from the role of policy in creating barriers and worsening inequalities, as noted by Bassell (2016).
Activist practitioners have long been influential in ESOL policy formation (Rosenberg, 2007), and the post-2010 period was no exception. For ESOL practitioners hoping to influence policy formation, persuasion, pressure and lobbying validates the role of practitioner knowledge as a component of policy-making processes. Practitioner activism supported the field to grow in prominence in the New Labour years: for instance, as we remarked earlier, lobbying by practitioners had ensured that ESOL was included under the Skills for Life umbrella in 2001. As Lo Bianco suggests, 'Seeing policy making as essentially a kind of argument and narrative describes a large part of how policy making actually happens, and it has the additional benefit of validating citizenship participation' (2019: 163). From the start of the coalition Government there had been grassroots reaction against cuts to ESOL funding that would exclude certain categories of student, including asylum seekers. The ESOL teachers' organisation the National Association of Teachers of English and Other Community Languages to Adults (NATECLA) instigated the Defend ESOL campaign in 2010, and the practitioner-led pressure group Action for ESOL published the ESOL Manifesto in 2012. While noting the social and economic arguments for a fully-funded ESOL, the ESOL Manifesto also stressed the anti-racist and human rights purposes of provision. It called for coordination too: 'ESOL provision should be accessible, comprehensive and integrated. It should reach out into the community and provide well-constructed but flexible routes onto academic and vocational courses' (p. 9). NATECLA itself published the document Towards an ESOL Strategy for England (NATECLA, 2016) in response to the incoherence of ESOL provision in national policy in England, summarising the calls for effective coordination of the field from both Action for ESOL and the Demos report. The organisation carried out a concerted campaign to lodge its messages in policy, including meetings in parliament. An ESOL strategy, said NATECLA, would allow local authorities to provide a comprehensive service, and enable anomalies in provision to be ironed out. The document also noted that immigration was a major concern in public perception; that there were uncertainties about the implications of the Brexit vote; and that social integration remained a key plank of government rhetoric if not planning.
A well-trodden route towards national policy formation is the work of an All-Party Parliamentary Group (APPG), which collects evidence on a focal topic. Practitioner activists provided evidence to inform the report of the APPG on Social Integration Integration not Demonisation (2017). This report adopted an orientation towards ESOL which linked it to human rights and aspiration, framing the need to support the field according to the notion that 'the ability to speak English is required in order to enjoy the basic freedoms which British society is built upon and is crucial to social mobility' (p. 5). True to its title, its call for a strategy-drawing upon the Demos report and NATECLA's (2016) document-invoked integration as well, which for the APPG equated with assimilation into an economically productive workforce. The strategy would not involve top-down coordination by central government; rather, the proposal was for 'a new statutory duty on local authorities to co-ordinate and optimise ESOL provision in their areas-sign-posting learners to suitable provision and facilitating a positive dialogue between language training providers' (p. 19).
The APPG report, along with the Casey review and the Demos report (itself heavily cited by Casey) informed what remains the main policy development relating to ESOL coordination, the Integrated Communities Strategy Green Paper published by the Ministry of Housing, Communities and Local Government (MHCLG, 2018). A Green Paper is a preliminary report of government proposals, a step on the way to law that is a statement of intent rather than a commitment to action. The Green Paper continued the focus on 'segregated communities' and the concern for shared 'British values' emphasised by Casey. It stated a commitment to developing good practice in ESOL, and was accompanied by an Integrated Communities Action Plan (MHCLG, 2019), which included not a promise of national coordination but guidance on effective practice in coordinating ESOL locally 'to support greater joinup of different providers and types of provision' (p. 13). It mentioned 'successful local approaches' to linking up providers, citing examples in Hackney, Leeds and Nottingham.
The policy documents discussed here that were created at or near the centre of government (the Casey review, the APPG report and the MHCLG Green Paper) embed ESOL within a discourse of integration. The lexeme INTEGRATE appears in one form or another 219 times in the Casey review, 400 times in the APPG report and nearly 300 in the Green Paper (not to mention in the actual titles of all three). The modernist dogma of 'one nation one language' (Joseph, 2006;Wright, 2004) persists in these documents, underpinned by a belief that in order for society to be strong, stable and cohesive, its population must share a common language. Later documents cite earlier ones, suggesting their influence Policy formation for adult migrant language education in… (the Casey review cites Demos 24 times; the APPG report mentions the Casey review 33 times, and the MHCLG Green Paper cites it 19 times). As noted above, neither the APPG report nor the MHCLG strategy propose coordination at a national scale (cf. Skills for Life) in their recommendations, and a national strategy for ESOL in England shows no sign of materialising, despite a recent renewal of lobbying efforts by NATECLA. Since the publication of the MHCLG Green Paper, advice for coordinating ESOL nationally has mainly come in the form of reports by an independent policy, research and development organisation, the Learning & Work Institute (L&WI). Presaged by the conclusions in the Green Paper, the L&WI has a firm focus on local and regional partnerships. Its Framework for ESOL Local Coordination (2020), commissioned by the MHCLG, the Department of Education and the Home Office, includes a local ESOL Partnership guide, comprising 'Effective practice guidance on how to best support the development of an ESOL partnership in your local area.' It cites the Leeds example, alongside initiatives in Bristol, Luton, Suffolk and Newham.
If national direction no longer exists, so funding for ESOL provision is also scant. The national government funding for ESOL from the Education and Skills Funding Agency (ESFA) adult skills budget is channelled towards Further Education (FE) college provision. This funding fell from £203 million (2009)(2010) to £90 million in 2015-2016 (Martin, 2017;see also Migration Yorkshire, 2021), and the demand for accessible, freely available classes even within FE consistently far outstrips supply. Much ESOL provision sits entirely outside central government education policy, in the hands of non-governmental and non-profit-making organisations, including charities, voluntary and community groups.

Stakeholder perspectives
To answer our second question, we turn now to how this policy lacuna is perceived by a range of stakeholders, ESOL teachers and providers, local, regional and national ESOL policy actors and other interested parties, and what its consequences are for practitioners, providers and students themselves. The perspectives were ascertained through a thematic analysis of a series of semi-structured interviews conducted in February and March 2020, principally by the second author, supported by the first author and another colleague. The initial motivation for carrying out these interviews was to generate evidence in support of an Impact Case Study (ICS), an aspect of the Research Excellence Framework, the system for assessing the quality of research carried out in UK universities (www. ref. ac. uk). The ICS is a narrative which describes how research resulted in a beneficial change in society, taking evidence from users of the research as a basis for a claim of impact. The focus of our ICS was a practice-and policy-oriented programme of ESOL research led by the first author which included the establishment of the Leeds (and later Yorkshirewide) ESOL coordination initiative mentioned above. Participants also gave their informed consent to the use of the interview data, in suitably anonymised form, in an academic paper with a policy focus.

Participants
A purposive approach to sampling was adopted. There were 20 interviewees in total, recruited through the professional networks of the first author. They were chosen to reflect the range of people who might speak with authority on ESOL policy, and fall broadly into three groups: ESOL teachers and course providers; ESOL policy actors working in local or regional government or for ESOL professional or support organisations; and people working outside the field of ESOL but with an interest in its coordination. They are listed in anonymised, summary form in Table 1. The inclusion criteria were that they had extensive knowledge of ESOL provision and policy formation at local, regional or national scale (for the ESOL policy actors and practitioners), or (for the other stakeholders) experience of working with migrants in education, employment and housing contexts.

Procedure and analytical approach
There were 16 individual and two pair interviews. Interviews took place in person, over the telephone or online and were audio-recorded. The interviews were semistructured, with the use of follow-up questions and probes (Kvale, 2018;Mann, 2016) and took a theme-based approach (Kvale, 2007). Our aim-abiding by the underlying principle of interpretivist research-was to understand the issues from the perspective of the participants. The schedule was organised around the topics of current challenges to successful ESOL provision in the UK generally and in the participant's regions, knowledge of efforts to coordinate ESOL provision, and views on the new regional coordinating resource, Learning English Yorkshire & the Humber (LEY&H). The data were first transcribed without using sentence punctuation, which was added to the extracts reproduced here to aid readability. To establish and organise patterns in the data, transcripts were analysed thematically, broadly following the phases outlined by Braun and Clarke (2006: 86-93). This involved systematically generating initial codes, collating codes into potential themes, reviewing themes to generate a thematic 'map' of the analysis, and defining the themes.

Results
The analysis generated a thematically-organised narrative. This allows us to draw out an interpretation of participants' individual and, importantly, their collective experience of, and views on, ESOL coordination in policy and in practice. Here, we summarise the aspects of the narrative that examine participants' perspectives on the need for coordination of ESOL, on what coordination might offer and for whom, and on the optimum scale at which the field might be organised. Direct quotations are in italics or in block quotes.

Fragmented provision
The stakeholders share a common understanding of the complexity of current ESOL provision, remarking upon its unconnected and uncoordinated nature. The interviews convey a sense that ESOL providers operate in isolation from each other. As Robert, who works in refugee employment support, says, there is some kind of fragmentation, if I could put it like that. People do their own thing. Describing efforts to address a need for communication across the sector, Farah, a committee member for a national ESOL professional organisation, notes that people quite often didn't know what everyone else was doing. The effects of fragmentation and a lack of communication are felt by students wishing to find classes as well as those who support them, suggests Susie, who works for a Local Authority: ESOL is very difficult to navigate. If providers and tutors struggle with this, then from a learner's perspective it is a real challenge. Without coordination, in the form of knowledge of provision, it is claimed that there is little clarity about where students can go, to find a class. Jack, from an asylum seeker support organisation, echoes the notion of challenge when communication and coordination are absent: There's loads of people doing different stuff but it's also a bit higgledy-piggledy. It's occasionally, you'll go round the corner and think, how did they even manage to keep that up? So it's inspirational and it's frustrating because without a great deal of shared understanding of what we need, it's a challenge.
Responsibility for the fragmented picture, says Christine, who works for a regional ESOL support organisation, lies with a lack of recognition of the extent of ESOL need at central government and at regional scale, what she describes as the main higher-level issue: It's hugely underestimated how many ESOL learners there are and how significant the need. The picture of a lack of funding for provision, provision of different types and with fundamentally different funding models (from government funded to volunteer-led classes), and no overall understanding or coordination of the field, make it difficult to improve, as an area of practice. Helen, who like Susie works for a local authority, notes that with inadequate piecemeal funding, it is hard to provide consistency in provision and that there is a lot of fatigue around trying to create positive change for the sector.

Benefits of coordination
The stakeholders also align in their understanding that the purpose of a meaningful strategy for ESOL at national scale would be to enable and support coordination of practice and provision at more local scales, in the towns and cities where prospective students are attempting to find somewhere to study. Central to these coordinating efforts would be the mapping of ESOL, identifying where ESOL classes currently take place and presenting the findings from the process in a publicly-accessible way. The results of any mapping activity would increase knowledge of provision for potential students and for those who are assisting them to find appropriate classes. As Farah puts it, the priority is about getting information out there for people to access.
ESOL providers also consider how coordination, where it exists, enables them to better aid students who are already in classes. They see their role as supporting students' progression through ESOL to other areas of education or training, and ultimately to greater independence. With coordination, they say, they can use knowledge about what is happening outside their own institutions or organisations when providing support, perhaps by directing their students to more appropriate provision, and consequently making room for new students. Conversely, without knowledge of provision elsewhere, teachers and providers are concerned that students will continue attending the same class year after year, seen as particularly an issue for women with young children in informal classes, who miss opportunities to progress.
For providers, a coordinated ESOL field is also cost-effective and can save resources by enabling collaboration and cooperation. This is compared with a culture where they are in competition with each other for students and hence for funding. With reference to local coordination efforts, Andy, with responsibility for ESOL at a national education policy organisation, explains: a lot of these initiatives, whether it's around referring learners to different providers or sharing a standard initial assessment, really require a high level of trust between providers. I've mentioned the historic context was one of competition between providers not of collaboration so it really does take a lot of work to get to that point.

Policy formation for adult migrant language education in…
For ESOL tutors, coordination has the benefit of them being able to connect with each other. Tutors working for small organisations in the third or voluntary sector, or even employed by a large FE college or a Local Authority but based in a small centre, can be quite isolated. As Christine says, there's no staffroom for ESOL tutors. Fragmented provision and a lack of coordination exacerbate this isolation, with the detrimental effect that they do not meet other providers or practitioners working in similar ways in similar contexts, to gather as a group, to network. Were such connections widespread, say the participants, this would enable learning together to share knowledge and teaching resources, and the joint organisation of workshops and training events. It would also support ESOL tutors' involvement in campaigning for the field, and in efforts to enhance social justice for ESOL students beyond their language learning experience, for example through taking action for affordable housing.
ESOL learners' lives outside their classrooms bring them into contact with a range of people who work outside the ESOL sector but for whom ESOL integrates with their activities. These stakeholders tell a unified story of how coordination of ESOL is lacking, and for whom better coordination-in terms of knowledge of provision-would support their work. For example, the clients of an asylum-seeker housing organisation need to be helped to find an informal class, as they are ineligible to attend a state-funded class within the first six months of their arrival. Another example: a university-based employability officer would benefit from a database of providers to whom she could direct undergraduate student volunteers and those looking for placements. Without knowledge of ESOL provision, people in these and equivalent positions across sectors cannot do their jobs effectively. Linda, who works for a regional-level migrant integration organisation, was involved in the UK Government's Vulnerable Person's Resettlement Scheme (VPRS), which provided Local Authorities with funding to support Syrian refugees. She explains how she was tasked with identifying services for Syrian refugees in her region, including ESOL classes: I started looking at what the Local Authority partners had already got in place for the Syrians and I think a bit naïvely I just assumed that they'd all have stuff in place that they'd know who to liaise with, if there were any gaps. It became very clear very quickly that […] some of them [Local Authorities] had .pdf documents which some individual normally had the burden of trying to compile and update and maybe circulate maybe every September, and within a few weeks it would be outdated again. Other places had nothing at all.
So without knowing what ESOL classes existed, the funding from the VPRS could not be used effectively. In sum, then, the central purpose of coordination is felt to be to gain an overview of what provision exists, primarily so students can be directed to an appropriate class where there is one, but also for other disparate purposes.

Local and national coordination
Stakeholders recognise that local coordination across partners from different sectors (ESOL and beyond, including but not restricted to Local Authorities) is both needed and lacking. Andy's organisation has been commissioned by the Government to advise upon the strategy on local ESOL partnerships. He outlines the benefits of local coordination of provision by cross-sector groups of partners, maintaining that good partnership working and cooperation needs to be in place between providers and other agencies, principally to assess local needs. This might be done through gathering intelligence about provider waiting lists or local labour market information. Coordinating hubs developed through partnerships, says Andy: have a really valuable role in doing that data gathering, they can also start to map some of the supply and demand and identify gaps and duplications and build links to potential outreach and engagement partners. So some of those organisations, they can really help get people who stand to benefit the most into provision.
Sally is charged with the coordination of ESOL provision in her city and is responsible for leading the development of a city-wide coordinating hub for ESOL. She stresses that the strategic partnership responsible for the hub's development is crucial. It involves different sectors in coordination at local level, including, for example the Job Centre, whose advisors commonly refer non-English speaking jobseekers to ESOL classes. Once you have those professionals on board, then you can start thinking about what works best for the particular area.
Notwithstanding the benefits of organisation at local scale, the stakeholders agree that to address the picture of fragmentation and lack of coordination, there should be an overall national ESOL strategy of some kind. They describe the reasons for such a strategy, what form it might take, how it might be implemented, and the likelihood of one emerging. The overarching purpose of a coordination strategy, they say, would be to address issues of access, to give a picture of what provision there is on the ground and to identify gaps. At present, even if a potential learner is eligible for a funded ESOL class, finding out what is actually available locally is difficult. A national strategy should support coordination at a local level, say the participants, to encourage more locally-appropriate and flexible approaches to delivery of provision, such as community and family-based ESOL.
Participants are realistic about how difficult it is to sustain an effort at coordinating ESOL locally, even though there are a small number of effective efforts in some cities. Interviewees describe how attempts to map provision are motivated by a visible need for coordination but then come to a halt when the money runs out. Sustainability is therefore a main concern. Any coordinating effort needs to be designed in such a way that it can continue, to avoid the situation described by Ursula, who chairs an ESOL professional organisation: [it is] important with projects that are funded for a certain amount, and then the funding is taken away, that whatever database or website they create is long lasting. Someone else comes along they get different funding and they're like "oh we're going to start again from scratch." This, say the participants, implies a sustained funding stream and support from policy-makers at more powerful levels, however. Linda articulates a general pessimism, in an echo of the conclusions from our examination of the policy trajectory for ESOL coordination, with her comment: I don't know at the moment whether I'm getting the feeling there's any appetite for more funding from central government for it. While participants are realistic about the likelihood of financial resources that central government might allocate to ESOL, some note that a strategy for coordination is cheaper to put in place and fund than extensive provision, and that some current funding for ESOL could be diverted to support a strategy. Their view is that lobbying in this direction is likely to be more effective than applying pressure for more funding. As David, an ESOL tutor and migrant rights activist, says, an appeal for a strategy for ESOL is more winnable than simply saying, sort of, reverse the cuts.

Discussion and conclusion
At the scale of the city and (in London) of the borough, there exist examples of effective efforts to connect and coordinate ESOL provision locally: in Manchester, Nottingham and Bristol, and in Hackney and Newham in London. The grass-roots partnerships behind these initiatives have, through the creative use of small-scale funding, made the results of mapping activity publicly available in different ways to the benefit of ESOL students and those who support them. There is just one serious coordinating effort at regional scale, the Learning English website in Yorkshire and the Humber (LEY&H), mentioned earlier. To end, therefore, we return to our earlier discussions of structure and agency in ESOL policy, and of the trajectory of policy in ESOL coordination. In our examination of policy documents we identified a sustained emphasis on the importance of learning and using English to support social integration of migrants, stressing the benefits of cohesion not diversity. At the same time-and indeed in the same document in the case of the MHCLG Green Paper-responsibility for provision and coordination of ESOL was presented as a local rather than national concern. In the absence of direction at national government level, in the form of a coherent overall strategy, those initiatives motivated by local actors exercising agency are the de facto current ESOL policy. Their formation can thus be regarded as rhizomatic. Following Deleuze andGuattari (1980/2004), the emergence of local partnerships has been multiple and non-hierarchical. They have originated when people with a common interest link with one another, establish connections, identify (sometimes very modest) sources of funding, map provision and make the results of mapping available to the public. But while grassroots, bottom-up approaches to coordination are valued by ESOL students, providers and other stakeholders, they are typically small-scale efforts that remain precarious, sparse, scattered and unconnected. Most ESOL providers nationwide are still working in an isolated way.
How, then, might a more comprehensive policy for coordinating adult migrant language education get formed and enacted across all the localities and regions where it is needed? The stakeholders in our study point to the importance of the pivotal interaction between the local and the national. One city or regionwhile it might provide a useful model for other places-has neither the scope nor responsibility to coordinate beyond its specific area. For local coordination activities to develop, flourish and connect with each other, they need fertile soil, the support of a coherent national policy of direction and implementation at national scale.
The signs that this might happen are not encouraging. It is more likely that national policy for the field will face continued neglect, and local and regional partnerships, where they exist, will continue to be left to fend for themselves. ESOL receives attention only tangentially in national government discourse, and rarely by name, as a weapon in the integration arsenal, and for its role in building a socially cohesive (and inevitably English-speaking) nation. Government support for ESOL in the third-sector (outside FE) typically comes in the form of piecemeal project funding, much of it to support volunteers. A particular trend since 2015 is for attention on ESOL in policy to rest on specific groups of forced migrants who are allowed to settle in the UK (currently Ukrainians, and before that Hong Kongers, Afghans and Syrians), for whom access to the English language is identified as a particular need. The schemes to support these migrants are jointly developed across government departments: the Hong Kong British Nationals (Overseas) Integration Programme, which includes funding for Local Authorities to provide ESOL, was jointly developed by the Home Office, the Department of Levelling Up, Housing & Communities, and the Foreign Office, but notably not the Department for Education.
We observe therefore a continued lack of clarity about where responsibility lies for ESOL in England, coupled with the lack of appetite centrally for a wellsupported and over-all approach to adult migrant language education. An effective counter-narrative might locate ESOL as more of a concern for national education policy than for policy on integration and immigration. For now, though, the picture painted in our earlier work (Simpson, 2012(Simpson, , 2015 persists: of poorly-resourced ESOL centres, especially those in the third sector, and of provision itself lacking cohesion within and beyond local areas. The rhizome might be an appropriate metaphor to describe ESOL policy formation. It does not, however, represent a practical approach to the provision of the comprehensive support that is needed to enable adult migrants to gain access to appropriate language classes and to navigate routes through their learning. Indeed the replacement of a hierarchical model of policy with a rhizomatic model would perhaps be welcomed by the small-state ideologues in decision-making positions in central government. In a call for ESOL coordination at national policy level, it is hardly helpful: without such coordination the social inequalities that pertain for potential students-among whom are the most disadvantaged in society-continue to be reproduced.