State recognition for ‘contested languages’: a comparative study of Sardinian and Asturian, 1992–2010

While the idea of a named language as a separate and discrete identity is a political and social construct, in the cases of Sardinian and Asturian doubts over their respective ‘languageness’ have real material consequences, particularly in relation to language policy decisions at the state level. The Asturian example highlights how its lack of official status means that it is either ignored or subjected to repeated challenges to its status as a language variety deserving of recognition and support, reflecting how ‘official language’ in the Spanish context is often understood in practice as synonymous with the theoretically broader category of ‘language’. In contrast, the recent state recognition of Sardinian speakers as a linguistic minority in Italy (Law 482/1999) illustrates how legal recognition served to overcome existing obstacles to the implementation of regional language policy measures. At the same time, the limited subsequent effects of this Law, particularly in the sphere of education, are a reminder of the shortcomings of top-down policies which fail to engage with the local language practices and attitudes of the communities of speakers recognized. The contrastive focus of this article thus acknowledges the continued material consequences of top-down language classification, while highlighting its inadequacies as a language policy mechanism which reinforces artificial distinctions between speech varieties and speakers deserving of recognition.


Introduction: contested languages and language policy from above
While named languages are ambiguous and arguably misleading constructs, in the sphere of 'top-down' language policy and planning they retain their power, with classification of a speech variety as a 'minority' or 'regional language' 1 carrying with it real material consequences. The case of Sardinian in Italy offers particular insight into the blurred divisions drawn between speech varieties, with this article focusing on the recent classification by the Italian state of Sardinian speakers as a 'linguistic minority' 2 and the subsequent effects of this recognition, specifically in relation to education. In Spain, the Asturian language shares an ambiguous legal status, most notably due to the absence of a declaration of official status for the language. This article will discuss the precise meaning of 'official languages' within the Spanish context, as well as how the ambiguous legal status of Asturian affects language policy decisions at the state level. By comparing and contrasting the cases of Sardinian and Asturian, this article provides further insight into the position of what Nic Craith has termed 'contested languages ' (2006: 108) within the unavoidably hierarchical language order of the state.
Contested languages are understood as 'forms of communication whose linguistic status is or has been disputed in the recent past' (Nic Craith 2006: 106). In many cases referred to as dialects, these forms of communication are widely perceived to have a lower social and cultural prestige, which can be both a consequence and cause of their contested language status. The aim of this article is not, however, to attempt to resolve such disputes or to attempt to offer my own definition of the distinction between a language and a dialect. Rather, the subsequent discussion and analysis depart from the view that the idea of a named language as a separate and discrete entity is a social construct and the consequence of a discursive project (Creese and Blackledge 2015;Makoni and Pennycook 2007;Nic Craith 2006;Woolard 1998). Or as Otheguy, García and Reid state most emphatically, 'a named language cannot be defined linguistically, cannot be defined, that is, in grammatical (lexical or structural) terms. And because a named language cannot be defined linguistically, it is not, strictly speaking, a linguistic object ' (2015: 286).
State recognition for 'contested languages': a comparative… As a social construct, the factors which influence the attribution of language status are not just about language (Blommaert 1999b: 429), and consequently, social, cultural and political factors in both contexts are given particular attention in the subsequent discussion. Equally, to assert that the category of 'language' is a social construct is not to deny its import, both to the speakers who identify with it and to external entities, such as the state authorities addressed in this article. Indeed, as well as investigating the political and social conditions under which Sardinian and Asturian have seen their status as language asserted or challenged, this article sets out to address the material consequences that the recognition of language status, or its absence, can have. The focus here is also explicitly on the state as the object of investigation. This is not intended, however, to exaggerate its position or the importance of 'top-down' policy measures in relation to language status or to suggest that legitimacy for a language can only be acquired through political authorities such as the state. Indeed, as will be discussed, there are limitations to what such external recognition can achieve and, while not the focus of primary research for this article, the aim is not to suggest such recognition is a substitute for speakers' own efforts at self-affirmation (Nic Craith 2006: 113). Nevertheless, to accept that a named language is a social construct is to also acknowledge the central role that political actors and authorities, particularly at the level of the state, have historically played in sanctioning and legitimizing such constructs. In this sense, the article seeks to investigate the state's current influence and effect of its decisions in relation to language status as inarguably still a key site of governance, even in an age of globalisation and multilayered decision-making (Williams 2007: 19).

Methodology
The article is based on a broader comparative study of the language policies of the central states of Spain and Italy between 1992 and 2010 in reference to linguistic minorities. The comparative approach is intended to pay attention to the distinct contexts of state action in Spain and Italy, both in terms of governance and what Schiffman has termed the 'linguistic cultures ' (1996: 5) in which the respective language policies are grounded. In both cases, three sources of primary data were identified for the purposes of this study: official state documentation and legislation, elite interviews with political and institutional representatives, and state-wide and regional newspapers. 3 The interviews were semi-structured and two types of respondent were identified: political representatives and civil servants within state and regional institutions, and expert commentators, primarily from the academic field. Although only a limited selection of interview data is used in this article, it should be clarified that the selection of respondents demonstrates a strong bias towards those in favour of recognition and support for minority languages. This is partly inevitable since those who have particularly useful knowledge on existing language policies concerning minority languages tend to be those who have taken a particular interest in either a specific minority group or multilingualism in general. At the same time, the triangulation of the interviews with other primary sources is intended to ensure the validity and reliability of research materials (Wodak 2006: 174), by preventing any single bias or view from dominating and by pinpointing gaps and inconsistencies between different accounts.
This triangulation of data and use of interviews allowed for consideration of concrete laws and regulations, as well as less visible practices and informal statements of intent. As language policy studies have stressed, it is essential to consider both formal or 'overt' forms of policy, such as constitutional clauses and language laws, as well as more implicit or 'covert' forms of policy (Kaplan and Baldauf 1997: xi;Schiffman 1996: 2;Shohamy 2006: 45-50;Spolsky 2004: 4). As a result, this article addresses not just stated policies or regulations but also the debates or 'discursive struggles' (Blommaert 1999a: 8) surrounding these policies, as well as how and indeed whether they are implemented. While not claiming to be able to offer comprehensive and complete coverage of all of these aspects in relation to the selected cases within the confines of this article, the analysis departs from the need to recognize the potential gaps between stated or 'official' policies and how they operate in practice. Acknowledging potential limitations of the research, as a project conducted by a single researcher and without the resources for extended fieldwork, it was not within the scope of the project to conduct either a survey or ethnographic study of the language users concerned. Nevertheless, while maintaining a critical awareness of potential limitations, the research makes use of the most reliable and extensive surveys conducted in both contexts and made possible by the work of large teams of researchers (Llera Ramo and San Martín Antuña 2003;Oppo 2007b). Equally, while the focus of the article is on top-down language management, its findings emphasise the importance of further research in both cases into local understandings of language and have the potential to act as a stimulus for such investigations.

Sardinian and Asturian: the sociolinguistic context
To provide a brief introduction to both languages and their contexts of use, Sardinian is spoken on the island of Sardinia and has traditionally been treated by the state as one of the Italian 'dialects', the local Romance varieties spoken across Italy. Although recent decades have seen important examples of cultural production in the Sardinian language, particularly in the form of popular and folk music, limited textual production (Spiga 2007: 75) and the absence of a unified standard form have most notably caused doubt regarding its status as a language. The absence of a widely accepted standard is also tied to the strong attachment of speakers to distinct varieties of Sardinian used across the island, with southern Campidanese and central Logudorese often considered the main varieties of the language (De Mauro 1987: 114). The situation is, however, further complicated by the existence of the Gallurese and Sassarese varieties spoken in the north of the island, which are closer to Corsican and the Tuscan dialect of mainland Italy (Blasco Ferrer 1984: 182-186;Paulis 1998Paulis : 1217. Other local varieties spoken on the island include Algherese, a variety of Catalan spoken in Alghero, and Tabarchino, a variety of the Ligurian dialect. All of these local varieties are used by speakers alongside Italian and some speakers may also use or know more than one local variety, reflecting the complex linguistic repertoires of the island's population (Spiga 2007). There is also no clear estimate for the number of speakers of Sardinian, partly as it depends on which varieties are included under this label, but a recent and extensive study carried out by the regional Observatory found that 68.4% of respondents spoke one of the local language varieties (Oppo 2007a).
Overall, the number of speakers of Sardinian is likely to be somewhere around the million mark, making it the largest legally recognised linguistic minority group in Italy (Council of Europe, 3 May 1999: 33;Euromosaic, 12 January 1995a). Despite this seemingly high number of speakers, Sardinian is more likely to be spoken by older generations and those who live in small villages (Oppo 2007a: 10), and is also most used in the family context (Oppo 2007c: 5). However, as is common with languages traditionally restricted primarily to the private sphere, there appears to be a strong emotive link to the language, and desire to see it promoted and maintained (Valdes 2007: 52). This desire is also linked to the development of the cultural movement known as 'neosardismo', which has been particularly active since the 1970s and led to the proposal of a series of unsuccessful laws to protect Sardinian, eventually resulting in Regional Law 26 for the Sardinian language in 1997 (Cossu 2001: 39).
Supported also by the inclusion of Sardinian in the 1999 state-wide Law 482 discussed below, this new legislation has facilitated greater institutional and financial support for the language from the regional government, with the establishment of a Service for the Sardinian Language and Culture (Regione Autonoma della Sardegna 2012). Both the regional and national laws have also stimulated recent attempts to introduce a unified standard form of Sardinian, with the largely unsuccessful introduction of the Limba Sarda Unificada by the regional government in 2001 (Calaresu 2003), followed by the introduction of the Limba Sarda Comuna in 2005. This has had some success in written documents produced by the regional and local administrations (Bolognesi 2007), but is still far from gaining wide acceptance among speakers and continues to be a much-debated subject within the region (Tufi 2013;Valdes 2007: 62-63). As primarily internal regional debates and without space to discuss in detail, standardization will not be a primary focus of discussion here, but the absence of a widely accepted standard form has been perceived as an obstacle to the greater visibility of Sardinian in the public sphere, particularly in its written form.
Asturian, 4 spoken in the northern Spanish region of Asturias, is in a remarkably similar situation. Again, some continue to refer to Asturian as a dialect of Castilian (Herreras 2006: 113), although, like Sardinian and other Romance languages, it originated as a dialect of Latin. In contrast to other regions of Spain, a significant cultural and/or political movement in support of the language did not develop until the 1970s, when the association Conceyu Bable began to make specific demands for the inclusion of the language in schools (Bauske 1998: 60-61). Continued doubts over the status of the language were also caused by the fact that standardisation came later than for the other languages of Spain, and was not undertaken until the 1980s by the newly created Academy of the Asturian Language. While still struggling to gain wide acceptance by speakers, standardization has not been the same focus of controversy and debate as in Sardinia, potentially reflecting the fact that standardization is more widely accepted in Spain as a necessary step in the promotion of use of regional languages. At the same time, the use by many speakers of a mixed form of Asturian and Castilian known as 'mecíu' or 'amestáu', a feature common to minority language contexts, has also encouraged the belief that Asturian is a dialect of Castilian (Cano González 2002: 51;de Andrés 2002: 49;Viejo Fernández 2004: 171).
The introduction of a Regional Law for the Use and Promotion of Bable/Asturian in 1998 has, however, reinforced the presence of Asturian in schools and the regional administration, but subsequent regional governments have not fully exploited or developed the provisions of this Law. In reference to the contemporary context of use, according to recent studies around a third of the regional population speaks Asturian, which is to say over 300,000 speakers (de Andrés 2002: 172;Llera Ramo and San Martín Antuña 2003: 188;Viejo Fernández 2004: 171). There are, however, clear indications of decline in use since a previous study conducted in 1991 (Llera Ramo andSan Martín Antuña 2003: 99), which is largely related to the low social prestige of the language and may also reflect the continued absence of a significant regionalist political movement in support of the language.

The acquisition of 'language' status
The ambiguous legal statuses of both Asturian and Sardinian have been a cause of significant debate in recent decades. In Spain and Italy, it is the respective Constitutions which provided the initial legal framework for the recognition of specific languages and their speakers, demonstrating also the salience of language policy questions in the histories of both states (Spolsky 2004: 59). To begin with the Italian Constitution of 1948, while noticeably not establishing the official status of the Italian language, Article 6 explicitly stated that 'The Republic safeguards linguistic minorities by means of appropriate measures' (Constitution of the Italian Republic 1948). However, as it did not specify the linguistic minorities to be protected, in the following decades this article was only enacted for what are often termed the 'national minorities' (Dell' Aquila and Iannàccaro 2004: 51), which is to say those on the borders who were typically assumed to speak the official language of the neighbouring state. This was primarily due to the pressure from neighbouring states (Coluzzi 2007: 31), and the protection of these groups and their languages was also explicitly stated in the regional Statutes of those regions concerned. 5 However, the Sardinian Statute, which defines the island's system of regional government, contains no references to the language (Special Statute for Sardinia 1948 6 ). Sardinia and other Italian regions did attempt to introduce their own laws to protect their respective local language varieties from the 1970s, seemingly in response to the absence of any state action or guidelines (Toso 2005: 261). These laws were often either returned to their senders by the government or rejected by the Constitutional Court. For example, in 1994 a Regional Law for the 'Safeguarding and promotion of the culture and language of Sardinia' was successfully challenged by the central government at the Constitutional Court, on the basis that the Law exceeded the legislative powers of the region in the area of education (Judgement 290, 4-13 July 1994). The Judgement provides no recognition of the existence of a Sardinian language, and also demonstrates how the lack of such recognition was a clear obstacle to ensuring the presence of Sardinian in schools.
With the repeated rejections of regional laws of this nature, growing pressure was also placed on the central government to introduce legislation to clarify the founding principle of Article 6 and, in its absence, the regions did not appear to have the power to attribute the status of linguistic minority to a specific group, and consequently to enact significant language planning measures. As Piergigli, an expert in constitutional law, asserts, without official recognition by the state: the linguistic minority is placed, so to speak, in a pre-legal state of mere existence, which does not allow it to benefit from any measures of protection apart from those of a general and generic nature, available to any citizen or individual. (2000: 631-632) 7 Various attempts were subsequently made to draw up a national law for the protection of linguistic minorities, most notably with Bill 612 in 1991, but these were repeatedly rejected by the state parties or were subject to the instability of various Italian governments (De Mauro 1996: 440). The widely debated Bill 612, for example, was passed by the Chamber of Deputies, but the government fell before the Senate had a chance to vote (Coluzzi 2007: 57). Nevertheless, the Sardinian government was able to successfully pass the previously mentioned regional law for Sardinian in 1997 and the fact that it went unchallenged by the central government appears to reflect a growing acceptance by external actors of the language status of Sardinian.
This was officially confirmed with the successful passing of the Italian Law 482 of 15 December 1999, titled 'Regulations concerning the safeguarding of the historic linguistic minorities'. Rather than providing a definition of this term 'historic linguistic minorities', a decision was made to list the specific minorities to which the Law referred: the Republic safeguards the language and culture of the Albanian, Catalan, Germanic, Greek, Slovene and Croatian populations and of those which speak French, Franco-Provençal, Friulian, Ladin, Occitan and Sardinian. (art. 2, Law 482, 15 December 1999) The content of this list was, however, the focus of much criticism, with Toso arguing that it merely served to legitimize a new form of hierarchy in relation to Italy's rich linguistic heritage (2004, 2008: 42-43). The exclusion of certain groups, particularly the Rom and Sinti groups, was a notably controversial issue (see Orioles 2003 and Palici Di Suni Prat 1999 for further discussions). 8 The inclusion of the Sardinian and Friulian communities, however, also caused particularly intense debate due to the contested status of their respective languages.
The inclusion of both languages had caused similar controversy in the previously unsuccessful Bill 612 in 1991 and, although the passing of Law 482 attracted less debate, many still objected to their inclusion. In the Chamber of Deputies several objections were raised to their inclusion, principally on the basis that they constituted 'dialects' rather than 'languages'. The Deputy Roberto Menia, for example, argued that: the Constitutional source makes no mention of Sardinians and Friulians as linguistic minorities, much less 'historic', as the title of this Bill affirms. This is obvious, since the people of Sardinia and Friuli are Italian populations, just like Sicilians and Lombards, who speak a dialect 'distant' from Italian. (Roberto Menia, Chamber of Deputies, 25 May 1998) 9 Such views were echoed in the Senate, most notably with the opposition of Giulio Andreotti, the former Prime Minister of Italy (Senate of the Republic, 25 November 1999). Even after the Law had been passed by both chambers of parliament, Andreotti went so far as to ask the President of the Italian Republic not to promulgate the Law on the basis that Sardinian should not be included for recognition (La Nuova Sardegna, 12 December 1999). While unsuccessful in his request, these examples demonstrate that some political representatives of the state strongly objected to the classification of Sardinian speakers as a 'linguistic minority', which can be understood as a recognition of the status of their speech variety as a 'minority language'. Such objections were tied to the belief in a clear distinction between languages and dialects, with many political representatives believing Sardinian could not be considered as distinct from Italy's other local Romance varieties excluded from recognition. In common with Sardinian and Friulian, the other regional varieties spoken in Italy, such as Lombard, Piedmontese and Sicilian, all derive from Latin rather than Italian. Linguistic experts have consequently cast doubts on the distinct status of Sardinian, with Pellegrini writing in 1977 that 'If we must consider the Sardinians and the Friulians as clearly foreign to the Italo-Romance linguistic domain, then we must also discuss the position of so many regional varieties in relation to the national language and culture ' (1977: 18-19). More recently, linguists such as Toso and Grassi have criticised this distinction between 'dialects' and 'minority languages', with Grassi describing it as 'artificial and, ultimately, misleading' (Grassi cited and translated in Parry 1994: 180 andToso 2008: 18). Although the linguist De Mauro has attempted to make a clear distinction between the dialects and the minority languages to be legally recognised (28 November 1991), Parry suggests that the exclusion of these other varieties responded more to 'pragmatic' concerns, with any attempt to include all of the 'dialects' likely to increase hostility to the passing of such a law (1994: 181). This is, in fact, reflected in the hostility to the original Bill 612 in 1991 in both the press and in parliament, with some vocal critics mistakenly believing it did refer to all of Italy's dialects. Opposition to the recognition of Italy's dialects was also tied to fears of political fragmentation with the consolidation of the separatist Northern League movement in Italy in the 1990s, illustrating how a number of political factors affected decisions concerning the attribution of language status (see Savoia 2003 for a detailed discussion of the debate surrounding Bill 612).
Nevertheless, to focus specifically on the Sardinian case, a certain consensus on the 'language' status of Sardinian does seem to have emerged, first among experts writing specifically on Sardinian (Blasco Ferrer 1984: 174;Wagner 1951: 58), but now shared by many linguistic experts across Italy and most significantly by De Mauro (1987: 114) who played an instrumental role in the design and implementation of Law 482. These experts typically focus on the linguistic distinctions between Sardinian and Italian (Orioles 2003: 87), illustrating how 'scientific' arguments still play an important role in discusions of linguistic legitimacy, although the isolation and island status of the Sardinians have inarguably also helped to emphasize the distinctiveness of the Sardinian-speaking minority. There has, in addition, been a growing, although still limited, acceptance and awareness among the speakers of Sardinian that their language variety constitutes a 'language' deserving of recognition (Marcato 2004: 68;Toso 2008: 90). Demonstrating the role of cultural and political factors, both internal and external perceptions that Sardinian should be recognized as a language are also related to the previously mentioned 'neosardismo' cultural movement which has developed since the 1970s (Toso 2008: 27). In sum, a certain, if still fragile, consensus had formed within Sardinia and within the academic community, which by 1999 appears to have reached the political arena of the state with the passing of Law 482.
In reference to these debates and any attempt to make judgements and assign a status to a speech variety, it is necessary to reassert the socially constructed nature of these labels. While both laymen and linguists may focus their attention on the internal dynamics of languages, or on particular features such as the existence of a standard form, this appears to be an attempt to hide or deny the importance of political and cultural factors in determining the status of a language variety. However, to accept the constructed nature of these labels does not mean to deny their importance or the very real material consequences of such classifications (Makoni and Pennycook 2007). Equally, although speakers themselves can attempt to confer the status of language on their own variety, the top-down recognition of the state, and other political bodies, should not be ignored (Nic Craith 2006: 107). Official recognition can remove legal obstacles, as illustrated by the previously rejected regional laws, to the implementation of language policy measures. State recognition also has the potential to confer legitimacy on speakers of marginalised languages and to empower language groups (Shohamy 2006: 63), as will be further discussed in the following section.
To turn, however, to the Asturian case, the language shares an ambiguous legal status, but unlike Sardinian still lacks such explicit legal recognition by the state authorities. The Spanish Constitution does, in fact, establish much more explicitly than the Italian Constitution the status to be accorded to Spain's distinct languages, with Article 3.1 establishing Castilian as the official language throughout the state (Spanish Constitution 1978). Article 3.2 goes on to clarify that 'The other Spanish languages will also be official in the respective Autonomous Communities 10 in accordance with their Statutes'. This allows for Spain's 'other' languages to be official within specific regions alongside Castilian. The clause does not, however, specify to which languages this refers, except that they are characterised by being 'Spanish' as in 'of the Spanish state' (Mar-Molinero 1990: 54). Despite this ambiguity, the clause does appear to be fairly prescriptive with the use of 'will be' suggesting, as Milian i Massana argues, that 'the other Spanish languages must be official, and not just potentially can be ' (1984: 134). However, as we will see in the Asturian example, this has not been the case.
Before discussing the Asturian case, it is important to clarify what 'official status' actually means within the Spanish context, since evidently labels such as 'official', 'national' or 'minority' are, like the term 'language' itself, political and social constructs (Williams 2005: 35). As such, the meaning and effects of officiality vary across contexts and in Spain obtaining 'official status' appears to have become the sole form of state recognition for speakers of other languages. 11 Essentially, in the Spanish context, a language is official when it is used by the public authorities as a 'normal' language of communication, both with members of the public and within the organisation itself (Judgement 82/1986). In relation to the education system, the Constitutional Court further clarified in 1994 that the official status of a language also meant ensuring its presence as a compulsory subject in schools (Judgement 337/1994). Consequently, official status in the Spanish context is considered to be the highest guarantee of the rights of speakers to use their language (Pérez Fernández 2006b: 29). However, the fact that Article 3.2 states that Spain's language will be official 'in accordance with their Statutes' has meant that it has been left to the individual regional Statutes, and consequently regional political representatives, to make the declaration of official status, rather than the Constitution or the state authorities (Judgement 87/1997). Some have argued that this phrase actually refers to the possibility to regulate and define the scope of official status, rather than to prevent its declaration (Milian i Massana 1984: 134). Nevertheless, by providing for the Statutes to declare explicitly the official status of a language, rather than doing so in the Constitution, this formula has resulted in a legal ambiguity which has been highlighted by the Asturian case.
The Asturian Statute is notable for the absence of a declaration of official status for the Asturian language. It does not ignore the subject entirely, explicitly providing for exclusive powers for the region over the protection and promotion of Asturian (arts. 4 and 10.1.21, Statute of Autonomy of the Principality of Asturias 1981). However, the Statute and its subsequent reforms also avoid any classification of Asturian as a 'language' (Pérez Fernández 2010: 175). By avoiding the naming or classification of Asturian as a language, this can be seen as a way for the regional government to avoid the apparent prescription in the Constitution that Spain's other languages 'will be' official. As Milian i Massana argues, the only justification for not making Asturian official is to not consider it a 'language', but rather as a mere 'linguistic variety' or as a dialect of Castilian (1984: 137).
Nevertheless, the 'promotion' and 'protection' of Asturian in the Asturian Statute still provides a legal recognition of the language at the regional level and entails obligations on the Asturian government to take action. This 'semi-official' recognition was reinforced by the introduction of a Regional Law for the Use and Promotion of Bable/Asturian in 1998. This Law attempted to provide a clearer legal framework for language planning measures in favour of Asturian (Asturian Law 1/1998;Pérez Fernández 2005: 44). In common with Spanish regions with co-official languages, 12 the Law does provide for forms of intervention in many spheres including the education system, the media and the administration. At the same time, while referring to Asturian as a language, in contrast to similar laws in other Spanish regions, it avoids the designation 'lengua propia'. This term, interpreted to mean a region's 'own' or potentially even 'rightful' language, is particular to the Spanish context and is used 12 These include Catalonia, the Balearic Islands, the Basque Country, Galicia, Valencia and Navarre.
Südtirol, and Slovene in certain towns in Friuli-Venezia Giulia. While co-official status is clearly important, Law 482 confirms that the Italian state does not see it as the sole form of possible recognition.
Footnote 11 (continued) not only in language laws in Catalonia and the Basque Country, for example, but also in the 2009 Languages Law of Aragon which recognizes both Catalan, spoken in a small area of the region, and the Aragonese romance variety estimated to have under 30,000 speakers (Campos Bandrés et al. 2016). Although some have criticized the term 'lengua propia' for suggesting that a region has only one 'rightful' language (Herreras 2006: 367;Siguán 1993: 91), given its widespread use across Spain its absence in Asturian legislation is notable. Instead, Asturian is referred to as the 'traditional language' of the region, a qualification which in contrast to the term 'own language' appears to cast doubts upon the role of the language in contemporary society.
Nevertheless, the Law has meant that language policy has taken on a more formal and institutionalised role in the regional government, with the creation of the General Directorate of Cultural Promotion and Language Policy in 2003 representing, according to its Director, 'the highest political recognition in the history of the recovery of Asturian and in the history of Asturian democracy' (SI1). 13 A limited form of recognition of Asturian was also provided by the state in its ratification of the European Charter for Regional or Minority Languages in 2001, which mentions the existence of languages recognised in regional Statutes, although as clearly distinct from Spain's 'official languages' (Instrument of Ratification, 15 September 2001). Nevertheless, despite what can be described as a form of 'semi-official' recognition, the position of Asturian remains ambiguous and has been subject to repeated challenges.

State recognition: consequences and challenges
This section will address both the effects of explicit recognition by the state in reference to the Sardinian case, as well as the obstacles and difficulties faced in the absence of such recognition in the Asturian case. To begin with Asturias, the previous discussion of its legal status may appear a merely technical or legalistic exercise, but in reality the ambiguous legal status of Asturian both reflects its social and political status, and reinforces doubts over Asturian's 'languageness' (Blommaert 1999b: 431). The contested status of Asturian was made most evident in a prominent language debate surrounding the attempt to introduce a degree at the University of Oviedo, in the capital of Asturias, for the study of Asturian philology. Asturian had, in fact, been present in the University of Oviedo since 1985 as an optional subject in the postgraduate teacher-training qualification and from 1996 the subject of 'Asturian language' became an optional subject in undergraduate degrees. However, the Faculty of Philology at the University decided to create a specific degree in Asturian philology in 1996, which was approved by the governing body of the University of Oviedo in 1997 (Cano González 1999: 115;Pérez Fernández 2006a: 274;Viejo Fernández 2004: 185).

State recognition for 'contested languages': a comparative…
Despite the autonomy granted to universities in Spain, the creation of any new degree must also be approved by the central Council of Universities (art. 28.1, Organic Law 11/1983), which is responsible for ensuring the proposed degree satisfies certain national requirements, such as the number of subjects and hours per course. However, after the request was sent to the Council of Universities, the petition was rejected in 1998, on the basis that 'From a scientific viewpoint, Asturian cannot be listed as a "language" in the curriculum' (cited in Congress of Deputies, 31 March 1999). With the legal support of the recently passed Regional Law for the Use and Promotion of Bable/Asturian, a second request was made to the Council in 2001, which was supported not only by the University but also the regional parliament of Asturias (Viejo Fernández 2004: 186). The legal and administrative dossier to support this request was also extended with reports signed by professors of philology from Spain and Europe. Nevertheless, the request was again rejected in 2001, and the Council went further in April 2002, demanding the removal of the optional subjects of 'Asturian language' and 'Asturian philology' in undergraduate degrees (de Andrés 2002: 154-155;Viejo Fernández 2004: 186). This despite the fact that in 1996 the Council of Universities itself had previously approved the content of these degrees and the inclusion of these subjects.
The members of the Council itself are the deans from the different universities of Spain, who may have no formal training in the subject area concerned, and for this reason it is often expected that the Council will respect the expertise of the representatives from the faculty and university making the request. In this case, however, Viejo Fernández argues this rejection appeared to be an 'outright attack on academic freedom ' (2004: 187). The Council used the ambiguity over the legal status of the language to justify the rejection of the degree in Asturian philology. The Council refused to accept that Asturian was a language and even went on to claim in 2001 that 'the differences between languages and dialects are well-known' (cited in de Andrés 2002: 110), demonstrating the common misconception that there are clear answers to such questions.
Although the Council of Universities is an autonomous institution, particularly from the central government, it can still be considered an institution of the central state and, as a powerful academic institution, its findings have important consequences in placing additional doubts on the status of Asturian. The significance of the decision was made evident in a debate in 2000 in the Spanish parliament concerning a proposed resolution to condemn the Council's decision. The Asturian parliament had already unanimously adopted a resolution of a similar nature (Cano González 1999: 120). However, in the central parliament the resolution was rejected by both the leading parties, the Popular Party (the PP) and the Spanish Socialist Party (the PSOE). Of interest were the similar reasons given for their opposition. Firstly, the representative from the PSOE stated that 'it is not appropriate to bring to the political sphere a question which belongs in the academic and scientific sphere' (Montseratt Palma i Muñoz, Congress of Deputies, 20 September 2000). Even clearer was the representative from the PP, who stated that: many different legal, scientific and linguistic reasons lead them [the Council] to reject the creation of the degree of Asturian language due to the fact that they do not consider Asturian to qualify scientifically as a language. And I want to make very clear that the distinction is from a scientific perspective. (Alicia Castro Masaveu, Congress of Deputies, 20 September 2000) The repetition of the word 'scientific' in these statements demonstrates the important role played by supposed 'experts', whose views are invoked and automatically assumed to be objective and 'scientific' rather than political (Blommaert 1999b: 429-430). In fact, the refusal to accept the Asturian degree and the classification of Asturian as a language by the Council appears to respond more to political than supposedly 'scientific' criteria (if, as earlier discussed, there can ever be any 'scientific' criteria concerning the status of a language). The fact that the Council showed no evidence or justification for its decision, and that it was not required to do so by any other body, is particularly questionable. The lack of clear justification is confirmed in the 2005 report by the Council of Europe which called on 'the Spanish authorities to clarify the situation, and in particular to explain the reasons for the Central Council's decision' (Council of Europe, 21 September 2005: 23), with no evidence of any response to this request from the Spanish government.
It is also notable that the representatives from the two main Spanish parties accepted the 'scientific' judgement of the Council of Universities, but not that of the Faculty of Philology in Oviedo, or of the other experts who endorsed the report and the creation of the degree. Interestingly, the influential sociolinguist Joshua Fishman also weighed in on the subject, describing it as an 'anti-democratic' and 'arbitrary' decision (Fishman cited in de Andrés 2002: 155). Despite what might first appear to be the admirable aims of the politicians to avoid political intervention into academic matters, given the extremely controversial nature of this decision and unclear circumstances in which it was taken, the argument that it was based on 'scientific' evidence appears far from justified. This is not to suggest that linguistic or sociolinguistic experts are themselves ever 'neutral' or 'non-political' in their interventions in such language debates, nor that such disputes can be resolved on linguistic grounds (Blommaert 1999b;Otheguy et al. 2015). However, what is noticeable here is the ultimately unchallenged 'scientific' authority of the Council of Universities as a centralised institution associated with the state, and the fact that if Asturian were an 'official' language, a legal and political rather than linguistic classification, it is extremely unlikely that its validity as an object of study would have been questioned.
Furthermore, the ambiguous legal status of Asturian as a non-official language automatically excludes it from gaining any recognition or support from the state. This was made most evident with the creation in 2007 of the Council and Office for Official Languages to coordinate the language policy of the Spanish state (Royal Decree 905/2007). As the name suggests, these bodies were created to ensure the appropriate usage of Spain's 'official' languages in state institutions. Consequently, as confirmed by a representative from the Office, 'The protection of non-official minority languages does not fall within the remit of the Council. It only covers the protection of official languages' (SI2). In sum, Asturian as a non-official language is typically ignored by the institutions of the central state or is subject to continued challenges to its status, demonstrating how 'official language' in the Spanish context often appears to be understood as synonymous with the theoretically broader category of 'language'. This lack of official recognition clearly places Asturian in the dubious and ambiguous position of a 'contested language', a categorisation which is likely to have some influence on speakers' own perceptions of the subordinate status of Asturian within Asturias itself. This appears to be supported by the 2002 sociolinguistic survey of Asturias which showed increasing signs of apathy since the previous study in 1991, with Llera Ramo and San Martín Antuña suggesting the lack of progress in gaining recognition for Asturian had led to 'weariness, resignation and a certain frustration in public opinion ' (2003: 314).
If we turn to the Sardinian case, however, we are also reminded of the limitations in terms of the effects that explicit legal recognition by the state can have on speakers of that language. In theory, action at a regional level to promote the local language can be legitimized and encouraged by state recognition (Telmon 2007: 318), with the potential to challenge the typically weak and marginalised position of minority languages. This is particularly important in relation to the education system, since the approach of schools towards different speech varieties is critical in determining the value accorded to them (Bourdieu 1991: 57). This explains why states have commonly viewed the education system as essential in ensuring the establishment of an official national language, as well as typically the exclusion of other varieties, which has traditionally been the case in Sardinia (Berlinguer and Mattone 1998: xxxvii;Euromosaic 1995b). The 1997 Regional Law for the promotion of the Sardinian language did, however, lead to the creation of regional funds for schools to encourage the teaching of the language. The central Ministry of Education also initially contributed to these regional funds in 1999 to encourage the promotion of the language (Depau and Zucca 2005: 283). However, the passing and implementation of Law 482, which established education as one of its key areas of action, would signal a more direct form of state intervention. Law 482 provides for the use of the minority language alongside Italian in pre-schools for 'carrying out educational activities' and for the use of the language as a 'teaching instrument' in primary and secondary schools (art. 4, Law 482, 15 December 1999). Article 5 explains the role of the central government, and specifically the Ministry of Education, which is tasked with issuing the general criteria for the implementation of Article 4, with an annual budget of around 1 million euros to be assigned to national and local projects for the study of minority languages and cultures (Portelli 2006: 132). To fulfil this role, the Minister for Education at the time, Tullio De Mauro, established a commission which met in November 2000 and included linguistic and sociolinguistic experts such as Leonardo Savoia, Vincenzo Orioles and Tullio Telmon (Savoia 2001: 17). The criteria drawn up by the commission were then sent in a circular letter to all of the relevant schools in 2001 and the commission met again in July 2002 to assess the projects submitted. The process was then repeated annually, although with the members of the commission changing.
Initially, there was a strong sense of optimism over the new direction the Ministry of Education was taking, signalling a clear move away from the stigmatisation of minority and local languages to an active promotion of their use. As Tiziana Sinesi from the Ministry of Education clarifies, the purpose of these projects was 'to tackle those prejudicial attitudes associated with the use of a local variety within an institutional setting ' (2010: 117). The early circular letters also received a positive response from schools. Tullio Telmon, a member of the first commission, stated that his experience 'had been quite encouraging because we could see a clear will from many schools, numerous schools, to insert something into their school curriculum' (II3). There was also a significant increase in the first few years in the number of projects proposed and approved, with Sardinia seeing the most significant increase in projects funded, from none in to 53 in 2007-2008(Ministry of Education, Universities and Research 2008.
Nevertheless, despite these initially positive responses, the reality is that many projects have focused on seemingly folkloric subjects, sometimes failing to provide more than passing references to the language. This focus has been most notable in Sardinia, with 40 out of 67 schools in a 2010 study declaring that most projects focused more on the local culture than the language itself (Iannàccaro 2010: 109). While cultural projects may not necessarily be negative, this cultural focus also tends to concentrate on traditional and archaic elements. This is evident from some of the titles of projects which received funding in Sardinia, such as: 'Food through time: memories of yesterday in the flavours of today' and 'Journey in search of our past' (Ministry of Education, Universities and Research 2003, 2008. Although Law 482 itself does refer to both the language and culture, the implication is that the two should coincide with the language remaining the primary focus, when in reality the language appears to be increasingly marginalised in such projects. In particular, although Article 4 of Law 482 provides for the potential vehicular usage of Sardinian as a language of instruction, Sardinian schools appear to have made no use of this provision, demonstrating how official declarations do not necessary guarantee implementation. The reasons behind such an approach, however, may not lie with the central government, but rather the individual schools, as well as the parents and local community which play an influential role when schools draw up proposals for projects. For example, when the researcher Iannàccaro contacted schools in towns which had declared themselves as Sardinian-speaking communities, six of the schools even declared that no minority language existed in the area. As he explains, 'It appears that the directors of these schools do not agree in considering Sardinian as a language and they see it instead as an Italian dialect ' (2010: 102). Furthermore, although a recent survey in Sardinia found that most respondents were in favour of some use of the local language in schools, the majority were also opposed to its usage as a vehicular language, and believed it should only be taught as a subject itself or to study the local culture (Valdes 2007: 53-54). The main reason for opposition to the use of the language in schools was that the language was not 'appropriate' (Valdes 2007: 53), demonstrating the continued belief that Sardinian is unsuitable for the transmission of knowledge and complex ideas.
This reminds us of the importance of correspondence between language planning measures and the practices and ideologies of the community concerned if attempts to promote language maintenance or transmission are to be successful (Schiffman 1996: 5;Spolsky 2004: 218). The absence of a widely accepted standard form of Sardinian also appears to reinforce the view that it should not be used as a language of instruction in schools. In this respect, Law 482 has been criticized for failing to acknowledge or investigate such local realities and for appearing to assume the prior existence of a standard written form for all of the languages recognised (Toso 2008: 49-53). Locally, there have been attempts to address how Sardinian can be used in schools in the absence of a widely accepted standard form, with teachers instead encouraged to use the local form with which they and the students are familiar (Iannàccaro 2010: 270-271). This illustrates, however, the limitations of overarching state-wide legislation, and in the Sardinian case it would appear top-down attempts by the state towards a more bilingual approach to education are likely to be controversial, and may encounter significant opposition from parents if the state authorities do not engage with local language practices and attitudes.
The Ministry of Education does, however, appear to have recognized that funded projects which take a primarily folkloric approach, treating Sardinian as merely a 'historic' language, are unlikely to convince younger generations of its current or future relevance (Iannàccaro 2010: 361-362). The circular letter sent to schools in 2008 encouraged the use of the Content and Language Integrated Learning teaching method, which focuses on the vehicular use of a language to teach other content and with the stated aim of transforming 'the historic minority languages into "living" languages' (Ministry of Education, Universities and Research, 23 July 2008). Nevertheless, despite these seemingly positive messages from the Ministry of Education, such efforts would require significant investment, for example to develop the necessary teaching materials and to engage more closely with local contexts of use, while in reality there are several signs that state-led action is all too often ineffective and minimal in both scale and impact. On a financial level, the original sum of just over a million euros was widely felt to be insufficient and there had been a major reduction in these funds to around half of the original sum by 2006 (Morelli 2006: 17). This sends a clear message that the promotion of minority languages in the education system is far from a priority. The focus on individual projects approved annually also leads to instability since, as the Council of Europe reported, it 'makes it very difficult to ensure continuity both in the learning process and working methods' (24 February 2005: 29). Experts in Sardinia would appear to agree, with Giuseppe Corongiu, Director of the Regional Service for the Sardinian Language and Culture, explaining that 'in schools the presence of the language is episodic, […] they do these projects and then the funding disappears and the project is over, Sardinian disappears' (II1). The ineffectiveness of these projects is suggested also by the 2006 Sardinian language survey, which found that 44.6% of schoolchildren at the time had never experienced use of the local language in school and only 17% had experienced regular use (Oppo 2007a: 40).
Consequently, central government action is having a very minimal effect and Iannàccaro's report on the implementation of Law 482 in education across Italy concludes that a new approach may be required (2010: 358). However, given the rapidly diminishing funds at the state level, it would appear that the impetus and funding will still need to be found primarily at the regional and local levels (see Depau and Zucca 2005 for details of activities initiated at these levels). The conclusions we can draw from state action concerning Sardinian in the education system are thus not as optimistic as may first appear, demonstrating very clearly the gap between official policies enshrined in law and the realities of implementation. Interestingly, Renato Soru, former Regional President of Sardinia who implemented key language planning measures for Sardinian during his time in office, is particularly sceptical of the potential role of the state and the impact of state legislation: 'the law does not count for anything […], what matters more is if I speak in Sardinian, if my friends speak in Sardinian, if we speak in Sardinian. Laws do very little' (II2). Soru here appears to be dismissive of the idea that speakers' language practices may be influenced by external actors and legislation. However, while state laws and action may not necessarily have an immediate impact on speakers, their potential as powerful mechanisms for affecting language practices should not be too easily dismissed (Shohamy 2006: 59-60). The danger illustrated here, however, is of 'top-down' policies which demonstrate limited engagement or dialogue with the intended targets of such policies who, without such consultation, may not even be aware of the change in legal status of their speech variety.

Conclusion
The cases of Sardinian and Asturian demonstrate that while legal recognition by the state is certainly not the sole or necessarily most effective means of improving the status of a language, without legal recognition such a task does become more challenging. At the same time, the contrastive focus of the article reveals that the forms such recognition may or should take and the consequent effects of recognition respond to the wider national and regional contexts. In the Spanish case, the statewide language policy regarding Spain's minority languages appears to have created a linguistic culture where the legal category of 'official language' has become near synonymous with term 'language' itself. Nevertheless, while the institutions of the state may be responsible for ignoring non-official languages, it is also important to remember that the declaration of 'official status' is the responsibility of the regional government of Asturias. It is perhaps hardly surprising that the institutions of the central state have doubts over the status of Asturian, given that the regional political class has refused to recognise Asturian as official or even to use the word 'language' in its own Statute. This is a marked contrast to other regions in Spain and belies the common assumption that regional governments or politicians are always more effective or favourable in their approach to regional or minority languages.
The example of the Italian Law for linguistic minorities, on the other hand, demonstrates an alternative model where the state itself, admittedly after significant delay, is actively supporting regional efforts to improve the status of the Sardinian language. Nevertheless, there is the danger of such top-down measures reinforcing a dependent and passive relationship, which places Sardinian in a precarious position particularly in light of recent fluctuations in state funding. The lack of priority given to implementing Law 482 may also reflect the continuation of a linguistic culture across Italy which, with the exception of the border regions with powerful 'national minorities', has often shown a reluctance to invest significant energy in language policy (Tosi 2001: 20). The Sardinian case also proves that legal recognition is not necessarily a panacea for improving the prestige and status of a minority language (Henrard 2003: 41), with this recognition having a State recognition for 'contested languages': a comparative… limited effect on speakers' views of its validity as a language of instruction. In particular, this challenge highlights the ineffectiveness of legislation at the centre which fails to fully investigate the local contexts of use and attitudes of the different communities of speakers recognized, and which fails to envision more enduring and effective ways of engaging with local educational institutions, which arguably have the greatest potential to influence such attitudes.
In both cases, it would appear unrealistic to expect institutions at the centre to provide the main impetus for improving the prestige and status of a language, even if it may well have been the institutions of the state which in the past ensured the marginalisation of these languages. Nevertheless, the most notable failing of both states is in the rigid distinction imposed between speech varieties and speakers deserving of recognition, namely 'official languages' or legally recognised 'linguistic minorities', and other typically unspecified varieties and speakers excluded from recognition or support. The controversy over the recognition of Sardinian would appear to stem at least in part from the fact that other regional varieties (the so-called 'dialects') are excluded from any form of state support (Toso 2008: 45-46). A more flexible approach, with an attempt to encompass the full multilingual reality of the state, could avoid such rigid divisions and allow the recognition of Sardinian to be seen not as discriminatory to speakers of other varieties denied recognition, but as part of a wider policy to support the full range of diverse linguistic practices within the state. Equally, in Spain, the idea that only 'official' languages deserve any form of state recognition presents a clear obstacle to the wider recognition of Spain's linguistic diversity. The exclusion of the languages of migrant communities, made explicit in Italy by the focus on 'historic linguistic minorities', also demonstrates how, in line with the European Charter for Regional or Minority Languages, these language policy mechanisms appear to be built upon the exclusion of other language practices.
Consequently, as Shohamy has highlighted, offering legal recognition to specific named languages, while seemingly more inclusive than the previous recognition of a single national language, in reality continues to act as a strategy of exclusion and rejection of speakers of other varieties who are denied such recognition (2006: 63). While acknowledging the real material political, social and economic effects that top-down classification as a language can entail, it would appear that merely extending the classification of 'language' to a wider range of speech varieties will not necessarily resolve the challenges faced by speakers of 'contested languages'. Instead it will merely perpetuate the belief that it is possible to separate and identify distinct varieties and elevate them above others, forcing speakers to either prove they meet the 'conventional and dominant criteria of "languageness"' (Jaffe 1999: 41) or to accept with resignation a subordinate and typically marginalised position for their language practices. Despite clear differences between the Sardinian and Asturian contexts, both emphasise that what is urgently needed in language policy is an expanded view of language (Shohamy 2006: 151-152;Makoni and Pennycook 2007) to prevent language classification acting as a block to providing any form of recognition for speakers whose language practices do not fit these rigid and constructed definitions.
Acknowledgements Thank you to Professor Brian Richardson and Dr Angel Smith, who supervised the Ph.D. thesis on which this article is based. Thanks also to those who generously shared their time and knowledge for the interviews which made this research possible, as well as the anonymous peer reviewers who offered valuable and constructive feedback which allowed me to further develop and articulate my argument.
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creat iveco mmons .org/licen ses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.