Key questions for this chapter:

• How are language policies done in different settings?

• How is language policy communicated in different social settings?

• Where are the limits of language policy?

3.1 Language Policy in Institutional and Community Contexts

This chapter examines where language policies are done, and how the context in which they are done affects the ways in which they are done. There is a particularly pressing need to ask questions about the context of language policies when looking at the field today. While there was once a relatively clear-cut focus on language policy as the language regulation activities of nation-states, this today seems far too narrow a view, first because there is increasing awareness of the complexity of studying policy in any nation-state and second, more crucially, because language policy research has in recent years engaged with a range of different settings. Families (e.g. Curdt-Christiansen & Palviainen, Reference Curdt-Christiansen and Palviainen2023), businesses (e.g. Barakos, Reference Barakos2020), trans-national organisations (e.g. McEntee-Atalianis & Vessey, Reference McEntee-Atalianis and Vessey2020), online communities (e.g. Hendus, Reference Hendus2015), along with more traditional settings like schools (e.g. Johnson, Reference Johnson and Tollefson2013b) and universities (e.g. Källkvist & Hult, Reference Källkvist and Hult2016), have proved fruitful grounds for the investigation of language policy processes, and have added significant depth to the field by highlighting the diverse dynamics of how people regulate language. For a framework which conceptualises language policy as action (see Chapter 1), this poses a significant challenge: actions, while being exponents of human agency and thus not fully dictated by constraints related to context, are nonetheless situated in particular settings and reflective of, if not necessarily completely determined by, the social structure around them. Following Scollon and Scollon (Reference Scollon and Scollon2004), actions can be seen as being performed in windows opened up by social practices (termed ‘sites of engagement’ in his framework). What this means for language policy is that we need to try and make sense of the social practices that structure a particular setting and which people respond to as they perform language policy actions.

To make this possible, it’s useful to draw a broad, flexible distinction between two types of setting in which language policies are done, institutions and communities (Figure 3.1). This distinction highlights a challenge for language policy in its aim to move away from the relatively narrow conceptual history of ‘policy’, a term which has through its life been primarily associated with institutions. Overwhelmingly, in fact, policy is commonly associated with one type of institution, the state. Looking for instance at what nouns are most common collocates of ‘policy’ in the Google Books (American) Corpus,Footnote ¹ we find the words ‘insurance’, ‘government’, ‘trade’, ‘security’, ‘health’, ‘state’, ‘sound’, ‘energy’, ‘company’, ‘tax’ in the top ten. Almost all firmly situate policy within the domain of the exercise of the power that a state has over its citizens, with the main exception being the collocate ‘company’ (as in ‘company policy’). A few more institutions talked about as having policies also feature further down the list, like ‘school’, ‘party’, ‘agency’, ‘hospital’ and ‘church’, showing that while some flexibility in the concept of ‘policy’ exists, it is nonetheless primarily associated with institutional settings, where the exercise of power over others is a regulated, procedural affair. While this may indeed be the case with some types of policy (e.g. trade policy, health policy, energy policy), existing language policy research has shown that, perhaps as a reflection of the tangibility of language in the everyday life of individuals and communities, actions through which people try to regulate how others speak occur in much less structured settings and in less structured ways.

Figure 3.1 Institutional and community language policies

The distinction between institutions and communities is needed both to draw parallels between these two types of settings, centred on the notion of language policy as action, and to highlight the key contrasts between them. One such area of concurrent parallels and contrasts is related to the mandate that exists for the exercise of power in a particular setting. Whatever their form, language policies involve someone attempting to regulate how others make meaning, but the backing that someone has to do this differs depending on setting. In institutions, the power to impose policy comes from the mandate of the institution itself to exercise power over a particular group of people, its members. In the case of the state, the exercise of power is primarily territorial, in that all those located in the geographic space seen to be governed by a particular state are subject to its power. Physical space matters in other types of policy as well, for instance, in particular types of buildings which are seen to have policymaking power (e.g. hospitals, schools), while elsewhere power may be associated with presence in social spaces (e.g. employment in a company, membership in a club). All these seem to assume that institutional authority stems from universal membership, that is, that all those situated in a particular physical or social space are equally subject to the power of an institution and thus must conform to policy (or leave the space, as is possible in some cases). With communities, there is no such pre-existing universal authority, and anyone seeking to exercise power by establishing policy is thus compelled to provide legitimacy for doing so at the same time. That is, in situations where no pre-existing framework of institutional power exists, one must construct a position of authority for themself in order to try and regulate the actions of others. Often, claims to authority of this type involve invoking a shared identity as a backing, for instance, invoking common identification with a nation as a backing for one’s attempt to create language policy. Such assumed positions of power are, by their nature, more explicit (since a person must openly legitimate their authority) and less stable (since any situationally established authority may also be situationally challenged).

In addition to these conceptual issues, there is also a need to consider practical differences between how policy is done in institutions and communities. In community settings, creating policy may simply involve a normative metalinguistic statement: ‘Don’t swear!’ While such an action is clearly situated in a particular setting and in dialogue with the social practices that structure that setting, the practices of a family, for instance, do not usually prescribe a formal legislative procedure for the creation of policy. In contrast, the creation of policy in institutions involves a relatively clear-cut time-space structure, which is generally legislated within the institution itself. This structure dictates where and when policy actions may take place, generally involving a sequence of events (a ‘policy cycle’) and some assumptions regarding who can play what role in these events (e.g. in a parliamentary committee, see Chapter 4). The rationalisation for such a structure stems from the nature of institutions and the need to legitimate their authority over a social space – because their legitimacy stems from the perceived universality of their authority over all those inhabiting a particular space, institutions are built around a structure which determines how their power is to be exercised with universality. This institutionalised structure also dictates how policies are communicated, most often by prescribing specific features of written texts which are seen to be imbued with the policymaking authority of an institution. Thus, while language policies in a community may simply involve making a metalinguistic statement with normative force, institutional language policies tend to be primarily communicated through a rather narrow set of highly regulated textual genres.

3.2 Language Policy and Institutional Power

As mentioned above, a key consideration in how language policy is conceptualised in institutions revolves around the authority of the institution itself to exercise power over a particular, typically well-defined social and/or physical space and the people located within it. This is most evident in the case of the modern state, which is characterised by the legalisation and society-wide acceptance of a universal mandate to exercise authority, with little or no exception. Thus, Weber defines the state as a ‘compulsory political organization’ (Weber, Reference Weber1978, pp. 53–54) which, through subordinate institutions like the police or military, is able to reserve for itself the ultimate manifestation of power, physical violence (Weber, Reference Weber1994). As Weber finds, the exact legitimation for this authority may vary, and indeed has varied through history, being at one point tied to the ‘divine right’ of particular individuals, or of a particular social class, to rule over others. This particular type of mandate has been largely backgrounded over the last century, as legitimacy has come to rely more on the legalistic aspects of states (i.e. sovereignty over a particular space, presence of legal norms and a bureaucratic apparatus) and their appeals to collective identity, a sense of belonging ‘based on shared duties, rights, and benefits’ (Jessop, Reference Jessop2015, p. 34). Aligned with the latter property, the power of states in large part derives from a common perception that they act ‘in the name of [a population’s] common interest or general will’ (Jessop, Reference Jessop1990, p. 341). Much of the same can be said of other institutions, which while not necessarily having the power to impose a universal mandate on the same scale as states, do often operate in similar ways. Private businesses, for instance, retain much power over the spaces they own and over their workforce, in some cases to the extent that the powers of the state become relativised. Similarly, while membership in smaller institutions like churches may depend on choice, it is often conditioned by acceptance of the legitimacy of their institutional power. This is also the case in online communities, where acceptance of terms and conditions, which include accepting the regulatory power of administrators and moderators, is often a requirement for access (see Chapter 8).

Weber’s definition of the state as a political organisation with a monopoly over violence – that is, it has sole legitimacy to use violence – continues to be widely cited, and indeed is reflective of a broader tendency in scholarly literature (exhaustively studied by Jessop, Reference Jessop1990; Reference Jessop2007; Reference Jessop2015) to conceptualise institutions like the state as things, that is, as entities with powers and agency of their own. This propensity for reification of institutions aligns with a ubiquitous practice in public discourse, namely the representation of institutions like states as actors in their own right. Structures like ‘Israel has launched an attack on Palestine’, ‘Thailand is underperforming economically’ and ‘Slovenia has closed its doors to refugees’ are common across most, if not all, social fields. A similar representation of businesses and other organisations is also common: ‘Microsoft has launched its new product line’ and ‘The law firm Kuits has promoted eight of its lawyers’. In such cases, institutions are represented as agents, almost human-like in their ability to act, their power being represented in a homogeneous manner, detached from any particular individual or group.

This unitary view of power in institutions, while common in scholarly and public discourse, is relatively unhelpful when it comes to the study of policy, especially from an empirical point of view. The key reason for this is that it backgrounds the complex processes that occur inside any institution: the ways that individuals try to exert their influence, how groups struggle for dominance, how compromises are made, how agendas are formed. These are natural in any institution and determine what happens to policies, so they cannot be backgrounded. Jessop (Reference Jessop2007) proposes that, as far as the state goes, the solution is to conceptualise it not as a thing but as a social relation, focussing on different subjects who act as agents of the state and on the relationships that determine when, where and how these agents exercise state power. Thus, the state should be seen not as a monolithic whole, but rather as the sum of its many different parts – an array of ‘various potential structural powers [which through its institutions offers] unequal chances to different forces within and outside of the state to act for different political purposes’ (Jessop, Reference Jessop2007, p. 37). While Jessop deals specifically with the power of the state and does not consider other types of institutions, these same principles can be seen to apply to them: just like the question is not ‘what the US does’ but ‘who, when, where and how acts in the name of the US’, we can also ask ‘who, when, where and how acts in the name of Microsoft’.

An understanding of institutions as social relations, thus, foregrounds the role that individuals play in language policy. What this means is that, aside from dealing with the structural aspects of institutions, namely the way their social practices determine the course of policy, there is also a need to focus on the individual actors who participate in processes of exercising authority in language policy. A study demonstrating this in practice is Johnson’s analysis of how a sub-section of the No Child Left Behind Act (NCLB, adopted in 2002) was produced by the US Federal Government and then implemented in a Philadelphia school district (see e.g. Johnson, Reference Johnson2013a, Reference Johnson and Tollefsonb). A key point of contention in this policy was the position of English-language education (for a summary of such debates in the US, see Lawton, Reference Lawton2013), and by the time the policy text was adopted, it had been phrased in a way which accommodated the arguments both of those advocating English-only education, as well as those arguing for bilingualism. As Johnson (Reference Johnson2013a, Reference Johnson and Tollefsonb) describes, this created a significant amount of implementational space for local actors (cf. Hornberger, Reference Hornberger2005) by allowing them to interpret the text in either direction. In the school district researched by Johnson, one administrator responsible for the implementation of NCLB interpreted it as supportive of bilingual education and was successful in securing federal funding with this argument. However, when this administrator was replaced as part of a reshuffle, her successor took the opposite position, and interpreted NCLB as an English-only policy (Johnson, Reference Johnson and Tollefson2013b, pp. 211–212). Once again, she was successful in securing federal funding with this argument (2013b, pp. 211–212). To Johnson, this indicates how important institutionally empowered individuals, whom he calls policy arbiters, can be in the policy process (see Chapter 6).

These dynamics lead us to the conclusion that ‘power’ is a rather complex phenomenon in institutions. As a general principle, the study of institutional power in language policy revolves around two perspectives, ideally adopted by the analyst at the same time. On the one hand, our gaze must be focussed ‘upwards’, on the structures that define an institution, and on how these provide people with ‘a frame for action without which they could not act [while also] constrain[ing] them to act within that frame’ (Fairclough, Reference Fairclough and Fairclough2013, p. 41). The existence of some type of structural constraint is a defining feature of institutions, as indeed is the general acceptance of such constraints among those acting within an institution. However, language policy scholars must also be sure to look ‘downward’, focussing on the people acting within an institution as its power is exercised. The challenge when adopting this perspective is to try and make sense of the role that individuals, their identities, narratives and beliefs play in the exercise of institutional power, and in turn to understand how individuals like policy arbiters are able to reshape institutions.

3.3 Language Policy and Community Authority

This comment comes from a Facebook group for non-local teachers of English in Thailand, a space where people currently working in the country, or those wanting to work in the country, search for work opportunities. The comment came under a post asking for advice about how to apply for work as a teacher in Thailand. The public profile of the group member who had written the comment states that they are a ‘French expatriate’, whereas the person it addresses, as suggested by the reference to Tagalog in the comment, is from the Philippines. These represent two key groups of non-local teachers of English in Thailand, those arriving from nations in the Global North (typically white, often with limited qualifications or experience in teaching) and those arriving from the Global South (typically people of colour, often with qualifications and experience in teaching). Between these groups, significant inequality often exists when it comes to recruitment, particularly regarding salary: white teachers of English are often preferred and command comfortable upper middle-class wages, whereas teachers of colour are often paid much less despite providing much more specialised types of teaching (e.g. in English-medium programmes). The dynamics of the discourse around this inequality commonly revolve around issues of language, in many cases through the type of statement made above, in which a relatively specific linguistic norm is explicitly being set, namely that a Tagalog politeness particle common in Filipino English (‘po’) is not to be used. Despite the specificity of this norm, we can assume, when considering other dynamics in the group and the political-economic relations it is embedded in, that it implies a somewhat broader devaluation of Filipino English (for more discussion about this Facebook group, see Chapter 8).

In this book, I position statements like the one above under the umbrella of community language policy. The term community here loosely refers to some type of social grouping which is not established through institutional structures for exercising power and instead emerges in other ways. Often, it is members’ participation in a particular type of activity, like discussing English teaching in Thailand in an online space, that brings about the development of communal ties. In these cases, the established term to describe the resultant group is community of practice, defined as a social group ‘created over time by the sustained pursuit of a shared enterprise’ (Wenger, Reference Wenger1999, p. 45). Attention here must be paid to a few key conditions, namely the existence of a community of practice as a ‘real’ community in which members are in a position to interact with each other and develop a shared repertoire of knowledge, as well as their joint engagement in a particular type of activity – a profession, a hobby, a political agenda, etc. – and their shared identification with this activity. Facebook groups devoted to specific topics are good examples of this type of community of practice, since they are generally formed around a specific issue, like English teaching in Thailand, attract members with a shared interest in that issue and provide them with a relatively well-defined discursive space in which they can interact. The role of language policy in such a community is, primarily, to regulate this discursive space by imposing limits on the types of semiotic resources that members can use. The above warning against the use of a resource associated with Filipino English can thus be read as an attempt to control the interaction order of the group, and, in a broader sense, as an attempt to delegitimate the voices of Filipino group members through the stigmatisation of their identity.

While there are many communities where these criteria of shared participation in activities are fulfilled, people also come together – or see themselves as coming together – in much broader groups. That is, aside from the many communities of practice we participate in, we also imagine ourselves as belonging to a multitude of communities which do not involve such direct interaction and instead rely primarily on the notion of a shared identity. I refer here to what Anderson described as imagined communities, that is, communities in which ‘members […] will never know most of their fellow-members, meet them, or even hear of them, yet in the minds of each lives the image of their communion’ (2006, p. 6). Originally, Anderson coined this term to make sense of the emotional belonging people feel to nations, communities which, by definition, transcend the existence of any direct interaction or acquaintance, and are instead centred on the perception of common belonging. In other words, when we identify as ‘Slovene’, ‘German’ or ‘Thai’, we are not doing so primarily because we have formed personal bonds with others seen to belong to the same nation, but because we see ourselves as sharing a natural identity which transcends any achievable level of direct acquaintance. In reality, of course, national identities are neither natural nor transcendent, having often emerged through top-down construction and being reproduced and explicitly enforced by everyday policing (see e.g. Billig, Reference Billig1995; Winichakul, Reference Winichakul1996; Wodak et al., Reference Wodak, de Cillia, Reisigl and Liebhart2009). This type of identity construction, reproduction and policing can also be found in other types of imagined communities, not just those associated with nationhood. The ‘LGBTQ+ community’ have for instance been described as imagined in how members identify with it and in the way that it is invoked (Jones, Reference Jones2007). We could also interpret ‘non-local teachers of English in Thailand’ as a sort of imagined community founded on a common identity and see the attempted prohibition of Filipino English ‘po’ above as an act of policing this identity. Indeed, while we could see the comment as a case of policing the interaction order of the Facebook group, it is phrased as a more general warning to those who ‘come to work here’, invoking a broader identity than that merely of group members.

How precisely the authority to engage in language policy actions is established in such communities depends greatly on the dynamics of each individual community. In the above example, the power dynamic appears relatively clear – the Filipino author of the original post is new to the group and has no experience teaching English in Thailand, while the French teacher responding in the comment assumes the authoritative position of an insider, an experienced community member relaying important information about the Thai language. In Bourdieu’s (Reference Bourdieu and Richardson1986) terms, the author makes use of their accumulated cultural capital, consisting of privileged forms of knowledge or skills acquired either through explicit schooling or implicit learning, to legitimise their own position as a policy actor. In other situations, a person’s social capital – the social network one is able to mobilise as a result of membership in different communities – as well as economic capital – access to assets that may be converted into money – may also determine relations of power. Here, some rethinking of Bourdieu’s conceptualisation is needed: while his interest was in the role of power in the evolution of social fields over time (Bourdieu, Reference Bourdieu1984), and his approach accordingly historical-structural, focussed on macro-level power relations, the relevance of these notions to language policy is from a more strategic, dynamic perspective. That is, we are less interested in forms of capital as a set of pre-existing relations of power in a particular field, and more with regard to how people strategically appeal to them in order to assume positions of power. While such appeals to capital are a general feature of authority construction in policy discourse and may be performed in institutional language policies (see Chapter 4), they are of particular relevance to community language policies, since these do not involve the exercise of power through institutional structures.

A further point to be made about power in community language policies is that the fact that these do not make use of institutional structures of exercising power should not be seen as implying an absence of such institutional structures. People often create language policies in settings where institutional mechanisms exist but do not act through those institutional mechanisms when exercising power. In a Facebook group, for instance, an institutional language policy can be created and enforced by administrators and moderators. The Facebook group described above in fact does have such an institutional language policy, found in a post containing a list of ‘group rules’:

1. 5) We will only publish (OP) posts which are predominantly in English

   Any links or other content posted must also be in English. You may chat in any language in the comments.

What most obviously sets the community language policy above apart from this institutional version is who the policymaker is and how they act – not an administrator or moderator creating a list of universally applicable and enforceable rules, but a ‘regular’ group member assuming the authority to create language policy in a relatively localised interaction. In this case, no direct face-off between the institutional or community language policies appeared to take place despite their somewhat contradictory stance – the institutional language policy being much more inclusive – though significant resistance was observed when other group members attempted to create and enforce similar policies (see Chapter 8).

3.4 Language Policy as Institutional Genre

Above, I discussed how institutional language policies involve people partaking in the exercise of institutional power by acting through legislated structural mechanisms. In institutions, such a structure includes a set of predefined windows in time-space, which determine when, where and who may engage in policy actions. Linked to this is another defining characteristic of institutional language policy, namely the centrality of text in its practices. Indeed, whereas community language policy can involve a number of different types of semiotic resources, institutional language policy can broadly be described as a nexus of institutionalised literacy practices. That is, we can imagine language policy in institutions in very basic, concrete terms as involving the actions of writing and reading, and the social structure around that as determining who gets to write, who gets to read, and the freedom they have in these actions. This means that the study of texts is a key concern for language policy analysis, though the way that texts are studied can differ significantly. From afar, a distinction between two broad perspectives on policy texts can be made following Bhatia (Reference Bhatia2004), who distinguishes a text-internal approach, focussed on the use of semiotic resources characteristic of a type of text, and text-external, oriented to the study of how a text is embedded in social practices.

From the text-internal perspective, the observation can be made that policy texts are often associated with a relatively rigid set of linguistic norms. These often follow conventions connected to legal writing, such as: frequent repetition of lexical items and infrequent use of pro-forms or synonyms; the use of terminology from various different codes; the use of characteristic syntactic structures, such as passives, the restricted use of tenses (e.g. present and future only) and others (e.g. Conley & O’Barr, Reference Conley and O’Barr1998; Maley, Reference Maley and Gibbons1994; Williams, Reference Williams, Salkie, Busuttil and van der Auwera2009). In many cases, these are intended to create a text which balances the need to avoid excessive ambiguity (a significant liability in legal proceedings) with sufficient flexibility and genericity to allow for application to new circumstances, unforeseen by the writers. In general, these conventions constitute a set of genres, distinct stable arrays of semiotic resources that characterise how meaning is made in institutional policy. Such a set can also be called a genre suite (Berkenkotter & Hanganu-Bresch, Reference Berkenkotter and Hanganu-Bresch2011) or a genre chain (Krzyżanowski, Reference Krzyżanowski, Cap and Okulska2013), the latter term referring to the way that different textual archetypes seem to follow each other in many policy settings, from green to white papers to bills, with associated genres like reports, assessments, etc.

In general, the challenge in making sense of such a genre chain when analysing language policies from a text-internal perspective is to achieve what may be called an intertextual reading of different documents. Because language policies are generally communicated through an array of different texts, it is often far too reductive to examine only one text as ‘the policy’, assuming that it represents the essential spirit of a particular language policy. Any ‘policy text’ is generally closely connected to a number of others, some coming before, some at the same time, and other after. Often, while produced under the guise of a particular policy, these may be written by different authors and intended for vastly different purposes and audiences. For instance, a number of language policy texts were produced during 2013–2018 in Malaysia as part of a reform of English language education in the country. Appearing at the centre were two closely interrelated agenda-setting documents (English Language Education Reform in Malaysia: An Agenda for Reform and English Language Education Reform in Malaysia: The Roadmap 2015–2025) produced on the behalf of the Ministry of Education by a team of writers (primarily applied linguists and teacher educators). Both these documents came after a much broader strategic text, Malaysia Education Blueprint 2013–2025, and were followed by a number of other documents, including teaching manuals and curricula. More significantly, all these texts were closely interconnected to two externally produced documents, the Common European Framework of Reference for Languages (CEFR), which was used as a point of reference throughout most of the other texts, and Cambridge Baseline 2013, a commissioned research report produced by Cambridge English Language Assessment for the Malaysian government. Both were referred to extensively to add legitimacy to other texts created as part of the language policy reform, which was positioned as bringing Malaysian education in line with national standards. In such complex cases, increasingly common as policy becomes concerned with taking global standards into account, the question of ‘What is the policy text?’ is difficult to answer in a simple way, and instead the challenge becomes to try and make sense of how the different texts that make up policy are defined through intertextual relationships.

This tendency of policy to be communicated through a multitude of different textual genres underlines the importance of the text-external perspective in the study of language policy. The guiding questions for this type of perspective are ‘What do people do with texts?’ and ‘How is the meaning of a text defined by what people do with it?’, as contrasted with the more text-internal ‘How is the meaning of a text defined by the semiotic resources used in it?’. This perspective is particularly valuable for conceptualising language policy as action, since it allows us to move beyond the confines of relatively static linguistic or semiotic analysis of texts, and instead observe how texts, or pieces of text, are brought to life through human agency. Here, we therefore need to make a difference between the actions that people perform and the texts they make use of when performing these actions. This means looking at texts not as actions but as mediational means, cultural tools which enable action (Wertsch, Reference Wertsch1993). The importance of this distinction is highlighted by Scollon’s observation that a text oriented toward protecting the environment is nothing but ‘black marks on a white page’ unless made use of for something other than ‘to prop up one’s feet’ (Reference Scollon2008, p. 18). For Scollon, this distinction highlights the need to focus policy analysis on action rather than text, seeing texts as meaningful only in a momentary sense, in the context of the specific social actions they come to mediate. This is not to say that texts don’t carry any meaning but that their meaning is, in the spirit of Halliday and Matthiessen’s (Reference Halliday and Matthiessen2004) conception of language as meaning potential – text meaning is only hypothetical until someone uses the text to perform an action (for more discussion, see Chapter 6).

With this distinction between text and action in mind, it’s important to keep in mind that there are significant dangers in seeing institutional language policy as communicated exclusively through particular types of genres or as being associated only with a certain type of language. In reality, many types of text can be imbued with institutional authority. Indeed, one of the pronounced current trends in policy is a tendency toward hybrid genres, in which features of policy texts are combined with those of other types. These types of genres feature heavily in what Mulderrig (Reference Mulderrig2018) describes as ‘nudge’ policy strategies, that is, covert actions taken by governments to try and bring about cultural shifts (e.g. using strategies associated with advertising to run an anti-obesity campaign). See, for instance, these extracts from a glossary of terms related to online learning produced at the University of Manchester in the UK:

At first glance, these appear to be extracts from a relatively conventional glossary, featuring keywords and definitions – aside from ‘Traditional’, other terms included ‘Blended learning’, ‘Face to face and in person’ and ‘Hybrid learning’. Though the introductory text positions the glossary as a collaborative effort (‘we have worked’, ‘we have a common language’, ‘staff and students’), several elements suggest an appeal to institutional authority, particularly the term ‘common language’ and the recommendation that some terms be avoided, which implies that some level of discourse management is taking place. From an intertextual perspective, it’s also worth pointing out that the glossary was produced in 2022 as part of the development of a Flexible Learning Strategy.Footnote ³ While this text may appear to be specific to the emergency provision of online learning that happened during the Covid-19 pandemic, it was in fact framed by the Strategy as part of a broader shift toward ‘increased student choice and flexible learning’ (p. 2). Such moves toward online learning in its various iterations are part of how higher education institutions across the world, as profit-oriented institutions, attempt to solidify their place in an increasingly competitive market. Since this process involves significant reconfiguration of how products are communicated – involving for instance a shift from the previously dominant focus on the ‘experience’ prospective students are part of when physically present on campus (Zhang & O’Halloran, Reference Zhang and O’Halloran2012) toward modes of study where such physical presence is in part or completely eliminated – it’s not surprising to see a level of coercion being exerted over the communicative practices of those acting within the university as an institution.

3.5 Language Policy as Metapragmatic Discourse

In the previous section, I argued that though institutional language policies are most typically associated with particular textual genres, this only tells part of the story, since there is a wide array of what can come to function as a ‘policy text’. This in principle applies to all types of setting in which language policies are done, though it is perhaps most relevant to community language policy, which, as a by-product of not being associated with institutionalised structures of the exercise of power, is also more or less free of the textual constraints associated with those forms of policy. To account for how language policy is done in community settings, a much broader descriptive scheme is thus needed. A useful framing device in this regard is Mortimer’s (Reference Mortimer, Barakos and Unger2016) description of language policy as metapragmatic discourse: ‘Language policy text and talk are discourse about discourse – they describe the uses, users, forms, and contexts of language. In being so, they are bits of discourse about types – types of language, types of people, types of activity, types of situations’ (p. 79). In other words, we can conceptualise language policy as involving explicit semiosis through which social meanings of language, in particular indexical links between linguistic forms and social identities, are created, reproduced and enforced (see Chapter 2).

As an example of the type of discourse that is relevant to this type of conceptualisation of language policy, here is a set of comments from a Facebook group for fans of the opera music programme, ‘La Barcaccia’, aired on Italian national radio. The comments came under a post in which a group member shared a newspaper article by a professor of Italian linguistics. The article argued for the use of the feminine article ‘la’ with the word ‘soprano’, traditionally treated as a masculine noun referring to voice type but interpreted by the writer as referring to the professional role of women, and thus more appropriately seen as a feminine noun. This is part of a broader debate around gender-inclusive language in Italy, beyond the scope of this book but examined in detail by Formato (Reference Formato2019), in which tensions typically arise around whether traditionally masculine nouns, particularly those associated with positions of power (e.g. il ministro, il presidente), can be converted to feminine gender when such positions are occupied by women (i.e. la ministra, la presidente). The spirit of the comments (originally in Italian, presented here in my translation, with notes in [square brackets]), highlights the contentions nature of these forms in Italian society:

1. 1.1 A: @B is quite right: soprano is not a noun but an adjective of register (timbre) and this is what it refers to, not a feminine noun.

   ‘La cantante di registro soprano’ [a female singer of the soprano register] would be the correct definition.

   So masculine, until they decide that little boys have il registro [voice register, masculine] and little girls la registra [voice register, feminine].

2. 1.2 C: It’s a quality of timbre and a span of register of sound. In Italian, neuter is expressed through masculine.

3. 1.3 D: I want to know how many times the distinguished professor who wrote that piece has been in an opera house, how many operas he knows, how many times he has talked about music and with whom.

4. 1.4 E: When I hear ‘la soprano’ it tells me that the person who used it, in general, has no idea about opera!

In these comments, we can observe several of the leitmotifs of the reproduction and enforcement of indexical links associated with metapragmatic discourse. In all the comments, we can see a clear polarity being constructed between ‘us’ and ‘them’, or between an imagined collective Self that writers identify with and which the form of language in question (il soprano) is seen to be representative of, and an imagined Other. In 1.1 and 1.2, these are relatively implicit, as the Other is referred to simply as ‘they’ or can be assumed to represent the inverse of ‘Italian’, that is, non-Italian. In comments 1.3 and 1.4, the Otherness is scaled down somewhat, relating to the collective identity of the imagined community of opera lovers, as both suggest that advocates of ‘la soprano’ lack the requisite cultural capital (i.e. knowledge of opera) that is required to be part of this particular community and, by extension, make decisions about forms of language seen to fall under its authority. In general, all these illustrate a feature that distinguishes language policy from other forms of metapragmatic discourse, namely that they are oriented to the establishment and enforcement of dichotomies of good and bad, or right and wrong, language. While metapragmatic discourse in general can refer to any type of ‘discourse about discourse’, the core concern of language policy is with how people try to exert power by establishing constraints on the types of resources others can activate to make meaning in discourse.

In general, this association of metapragmatic discourse with power is often the primary empirical criterion through which language policy can be separated from other sociolinguistic phenomena. Here, I re-iterate the distinction between policy and practice that is central to conceptualising language policy as action but has not always been made clear in language policy scholarship. By placing focus on metapragmatic discourse, we acknowledge that social practices are key determiners of how semiotic resources are used, but that many practices do not fall into the scope of language policy. The social practices relevant to a particular social space, for example the interaction order underlying customer-seller exchanges in a university cafeteria, are more typically acquired through implicit socialisation rather than being explicitly enforced through metapragmatic discourse. This is a distinction between the framework presented in this book and a number of others, which have included ‘practice’ in its totality under the umbrella of language policy, categorised ‘practice’ as de facto language policy (Shohamy, Reference Shohamy2006), implicit language policy (Spolsky, Reference Spolsky2004) or practiced language policy (Bonacina-Pugh, Reference Bonacina-Pugh2012). Focusing on metapragmatic discourse entails a narrower focus, and assumes that while practices, as elements of social structure, without doubt play a significant role in constraining the scope of people’s agentive use of language in particular social settings, it is when specific people themselves act to constrain the actions of others that language policy comes in.

This highlights a final characteristic of language policy as metapragmatic discourse, the fact that it involves taking up a position of authority over others. Above, I argued that a key concern in language policy, particularly in community settings, is how people appeal to forms of capital to legitimate their own position as a policy actor. To this individual perspective, we must also add a further dimension, namely the way that community language policy often involves invoking a collective mandate. That is, when people attempt to dictate the practices of others, this often involves a level of spokespersonship – taking up the position of a policy actor not on behalf of oneself, but in the name of the community as a whole. There are thus significant parallels between language policy and how individuals in general take up positions of authority over communities. Pels (Reference Pels2000) for instance examined how such positions of spokespersonship are assumed by intellectuals, who, while assuming as individuals a privileged vantage point over society, tend to do so in the name of a broader community and are compelled to continuously balance a level of distance from the community with their own membership within it. In many cases, the community that intellectuals speak for is the nation, as broken down in much detail by Sicurella (Reference Sicurella2020), a tendency to ‘speak for the nation’ is also common in community language policy. Indeed, in the post-Yugoslav space examined by Sicurella, there is a long history of intellectual spokespersonship in issues of language, most typically in the name of nation-building. In Slovenia, a typical space for such engagement were language advice columns, featuring the advice of more or less distinguished writers about what ‘good Slovene’ was. This highlights a related type of spokespersonship, where instead of the nation, what people speak for is the language itself, using its imagined system as a mandate for their language policy actions. This is evidenced in comments 1.1 and 1.2 from the community of opera fans shown above, where the writers position themselves as simply voicing realities inherent in language-as-system, positioning these as facts of nature beyond human intervention. Such disembodiment of language is a feature particularly of nationalist ideology, where the transcendent, structured nature of ‘named languages’ like Italian is a key tenet of nationhood (Anderson, Reference Anderson2006).

3.6 Reflecting on the Intertwining of Authorities in Language Policy

This chapter has broken down the distinction between how language policy is done in institutional and community settings, focussing in particular on the differences in the exercise of authority between these two. However, it must be noted that the distinction between institutions and communities is not to be seen as more than a straightforward analytical heuristic, intended to help analyses of language policy make sense of the context in which language policy actions are performed. The distinction is in particular not to be read as a dichotomy of two mutually exclusive categories, and it is to be assumed that any setting where language policy is done will likely have characteristics associated with both poles of this continuum. For instance, while the power of a state may emanate primarily from its dominance over a particular physical space, this is also routinely backed up by appeals to shared identity in the modern nation-state, meaning that the mandate to exercise power is drawn from both the institutional and community poles concurrently. Similarly, there is much complexity when it comes to how policies are communicated, as institutional policy genres are often appropriated to aid an appeal to community authority, while shifts away from conventional ‘policy language’ and toward more informal modes of expression have also been a feature of the exercise of state power in recent decades (a process described as ‘conversationalisation’ by Fairclough, Reference Fairclough, Keat, Abercrombie and Whiteley1994).

Overall, the distinction between institutional and community language policy thus centres on the type of authority that is invoked as part of doing policy, which by necessity leaves the door open to situations where multiple levels of authority become intertwined. Indeed, it is not unreasonable to say that any imaginable interaction could potentially be subject to the authorities of multiple levels of community and institutional authority (Figure 3.2). The key question for language policy research in such cases is whether there is any contextual cue that suggests a particular authority is relevant to the interaction. Relevance is not to be assumed, since it is for instance perfectly possible that an interaction that takes place within the territory of a particular state is not directly shaped by language policies enacted by the government. Though such links may of course be made by invoking historical context – for example, to explain why an interaction may involve the use of national language rather than a minority language in a context where minority identity was historically stigmatised – the tenuous nature of this type of connection once again highlights the need to distinguish between ‘practices’ as stable forms of action and ‘policies’ as actions by which authority is exercised over practices (see Chapter 2). Complexity arises where policies concurrently invoke multiple authorities, and in such cases complex negotiations can often take place as different layers of institutional and community policy are reconciled (see Chapter 6).

Figure 3.2 Intertwined institutional and community authorities in language policy