Abstract
Introduction
Not so long ago, we seemed to be making progress on the question of how to incorporate meaningfully the cultural and linguistic backgrounds of our increasingly diverse school student population. In the 1990s, multicultural and bilingual educational approaches were becoming commonplace in modern liberal democracies (Banks & Banks, 2004; May, 1999, 2009). Similarly, multiculturalism appeared to be gaining widespread acceptance as a public policy response to the burgeoning diversity of state populations in an era of rapid globalization and related transmigration (Kymlicka, 2001, 2007; Parekh, 2000). Both developments built on a history of nearly 50 years of advocacy of multicultural and bilingual education, and wider state policies of inclusion for minority groups, which had its genesis in the U.S. Civil Rights movement but had extended to other Western countries, including Canada, Britain, Australia, and New Zealand. Even critics of multiculturalism conceded its impact on public policy at that time, particularly within education—a wearied resignation most notably captured in Nathan Glazer’s (1998) phrase, “We are all multiculturalists now.” Multiculturalism, at least in Glazer’s view, had finally “won” because the issue of greater public representation for minority groups was increasingly commonplace in discussions of democracy and representation in the civic realm—including, centrally, within schools (see, e.g., Goldberg, 1994; Kymlicka, 1995; Taylor, 1994).
How times have changed. Over the past decade, and particularly post-9/11, we have seen a rapid and significant retrenchment of multiculturalism as public policy, particularly within education. In the United States, a burgeoning standards and testing movement, spearheaded by the No Child Left Behind (NCLB) Act of 2001, has replaced earlier attention to racial and ethnic diversity (May & Sleeter, 2010). Decades of affirmative action and related civil rights advances for African Americans have been dismantled, most notably in relation to access to higher education (Kellough, 2006). The related provision of bilingual education, particularly for Latino/Latina Americans, has also been severely circumscribed, and in some U.S. states actually proscribed, by legislation promoting a monolingual English language philosophy as a prerequisite for U.S. citizenship (Crawford, 2008; May, 2012, chap. 6). Meanwhile, across Europe, multiculturalism as public policy is in apparent full retreat, as European states increasingly assert that minority groups “integrate” or accept dominant social, cultural, linguistic, and (especially) religious mores as the price of ongoing citizenship (Modood, 2007). Again, education has been a key focus, with bilingual and multicultural education programs facing significant retrenchment across Europe as a result (Modood & May, 2001).
Language, Cosmopolitanism, and Mobility
This apparent retrenchment of multiculturalism as public policy has been bolstered by parallel arguments for a more “cosmopolitan” approach to education within an increasingly globalized world (Archibugi, 2005; Barry, 2001; Waldron, 1995). In these accounts, globalization is sometimes viewed as primarily a geocultural consequence of late capitalism (e.g., Cox, 1996) or, alternatively, as a geopolitical phenomenon, linked to previous forms of social and political organization, most notably the nationalism of the previous few centuries (Held, McGrew, Goldblatt, & Perraton, 1999; Hobsbawm, 2008; Robertson, 1992, 1995). 1 However, in both accounts, as the sociologist Craig Calhoun (2003, 2007) argues, globalization is linked with new global (and hybrid) forms of identity that allow one to “escape” the “confines” of more localized (and, by extension, fixed) identities, including linguistic ones. In this view, globalization constitutes both the basis of individual transformation—the ability to adopt hybrid, cosmopolitan forms of identity that transcend both local and national borders—and, more broadly, is seen as the next stage of the modernization process. Martha Nussbaum’s (1997) notion of “citizen of the world,” a global form of citizenship that is no longer rooted in, or confined to, local, ethnic, or national identities, but specifically transcends them, highlights this clearly.
Closely allied with this generalist cosmopolitan view is an advocacy of languages of wider communication, and particularly English as the current world language, as the new means of global interchange and the basis of social mobility. A related argument is that the ongoing use of “local languages,” via language rights or the broader politics of multiculturalism, simply entrenches social, cultural, and political isolationism, as well as socioeconomic disadvantage. A recent example of this position can be seen in the work of the Italian political theorist, Daniele Archibugi (2005). He argues that the answer to the “problem” of increasing linguistic diversity is not the multiculturalist recognition of language rights but rather a global cosmopolitanism based on a language of wider communication. This is because, for Archibugi, a democratic politics requires “the willingness of all players to make an effort to understand each other” and thus a “willingness to overcome the barriers of mutual understanding, including linguistic ones” (p. 537). Following from this, he maintains that “linguistic diversity is an obstacle [italics added] to equality and participation” (p. 549; see also Blommaert, Peppännen, & Spotti, 2012). Although he uses the metaphor of the artificial language, Esperanto, to illustrate his normative arguments, and perhaps also to create some cover for their implications, it is nonetheless still abundantly clear that the “common language” he has in mind here is English (Ives, 2010). We see this in the case study examples he uses to illustrate his position: In one—a state school in an increasingly mixed Anglo/Latino neighborhood in California—Archibugi (2005) outlines a hypothetical scenario of increasing tension between the two groups with respect to the school’s future direction: The Hispanic students do not speak English well and their parents speak it even worse. School parents-students meetings end in pandemonium, with the Anglos complaining that their children are starting to make spelling mistakes and the Hispanics protesting because their children are bullied. At the end of a stormy meeting, an Anglo father, citing Samuel Huntingdon, invites the Hispanic community to dream in English. In return, an outraged Mexican slaps him in the face. (p. 547)
Meanwhile, Archibugi (2005) also assumes in his scenario that the Anglo parents are middle-class and that most of the Latino parents are “cleaners” but with aspirations “to enable their children to live in conditions that will avoid perpetuating the [existing] class division based on different ethnic groups” (p. 548). In offering potential solutions going forward, he contrasts a multiculturalist response of parallel English and Spanish instruction within the school for the respective groups—bilingual or dual language education, in effect—with, in his view, a clearly preferable cosmopolitan solution of English language instruction for all. This cosmopolitan solution is predicated on the basis that “American citizens with a good knowledge of English have (1) higher incomes; (2) less risk of being unemployed; (3), less risk of being imprisoned and (4) better hopes for a longer life” (p. 548). As a salve to the Latino population, however, the cosmopolitan proposal also includes compulsory courses in Spanish language and culture for all, while encouraging Latino parents to learn English in night school and Anglo parents “salsa and other Latin American dances” (p. 548).
The underlying presumptions about language and mobility in Archibugi’s (2005) scenario are also clearly echoed in comparable accounts. Another prominent political theorist, Thomas Pogge (2003), for example, begins his discussion on the language rights attributable to Latinos in the United States—or, more accurately, in his view, the lack thereof—by asserting unequivocally that English is the predominant language of the United States and has been so for much of its (colonial) history.
2
He then proceeds to observe, in close accord with Archibugi’s (2005) hypothetical example, that many Latinos “do not speak English well” (Pogge, 2003, p. 105). Accordingly, to best serve the educational interests of Latino children, Pogge specifically endorses an English “immersion”
3
educational approach so as to ensure that Latino students gain the necessary fluency in English to succeed in the wider society. He argues, The choice of English as the universal language of instruction is justified by reference to the best interests of children with other native languages, for whom speaking good English (in addition to their native language) will be an enormous advantage in their future social and professional lives. (p. 120)
4
But Pogge (2003) does not end his arguments there. He also suggests that those (Latino) parents who opt instead for bilingual education may well be “perpetuating a cultural community irrespective of whether this benefits the children concerned” (p. 116). For him, this amounts to an illiberal “chosen inequality” for those children because it “consigns” them to an educational approach that, in maintaining Spanish (or other languages), willfully delimits their longer term mobility in U.S. society. This position is made even starker by Pogge’s intimation that such a choice could possibly warrant the same constraints applied to parents as other child protection laws; equating bilingual education, in effect, with child abuse, a trope much used by the wider “English-only” movement in the United States (Crawford, 2008; May, 2012, chap. 6; see below for further discussion).
Two other political theorists, David Laitin and Rob Reich (2003), argue much the same position, asserting that “forcing” bilingual education on children will curtail “their opportunities to learn the language of some broader societal culture” (p. 92). Relatedly, they fret that these “individuals have no influence over the language of their parents, yet their parents’ language if it is a minority one . . . constrains social mobility.” As a result, “those who speak a minority (or dominated) language are more likely to stand permanently [italics added] on the lower-rungs of the socio-economic ladder” (p. 92). Indeed, they proceed to observe that if minority individuals are foolish enough to perpetuate the speaking of a minority language, then they can simply be regarded as “happy slaves,” having no one else to blame but themselves for their subsequent limited social mobility.
Setting aside, if one can, the clearly racialized paternalism that pervades all these accounts, a key additional problem with them is that they are substantively wrong. This is so both with respect to the links between English and social mobility and in terms of the apparent efficacy of English-only language instruction for learning English as an additional language. I will return, in due course, to the latter issue of the most effective pedagogies for bilingual students. But first let me, briefly, address the former presumption that knowledge of English automatically leads to increased educational, social, and economic mobility for linguistic-minority students. Quite simply, this is not nearly so often the case as advocates of a cosmopolitan education in English might presuppose. Why this is so is because what is often (willfully) overlooked in such cosmopolitan accounts are the ongoing structural inequalities that necessarily delimit access and opportunities—particularly for minoritized populations. For example, in many postcolonial countries, small English-speaking elites have continued the same policies as their former colonizers to ensure that (limited) access to English language education acts as a crucial distributor of social prestige and wealth (Heugh, 2008; Ives, 2010). Pattanayak (1969, 1985, 1990) and Dasgupta (1993) describe exactly this pattern in relation to India, where English remained the preserve of a small high-caste elite until at least the 1990s. The impact of globalization has changed this somewhat in India since then, particularly with the increasing use by multinational companies of business process outsourcing and information technology outsourcing requiring English language expertise (Graddol, 2007). Examples here include call centers and publishing, both of which India has benefited directly from in the past decade. However, these developments also highlight the significant differentials and inequalities in pay and conditions for workers in India and other comparable contexts when compared with “source” countries—indeed, these conditions are the principal raison d’être for the outsourcing in the first place. Meanwhile, the necessary English language expertise is still closely related to existing social class and related educational hierarchies in India, as elsewhere (Morgan & Ramanathan, 2009; Sonntag, 2009).
A similar scenario is evident in Africa, where, despite English being an official or co-official language in as many as 15 postcolonial African states, the actual percentage of English speakers in each of these states never exceeds 20% (Heugh, 2008; Ngũgĩ, 1993; Schmied, 1991). Indeed, Alexandre (1972) has gone as far as to suggest that in postcolonial Africa, social class can be distinguished more clearly on linguistic than economic lines. Although this observation willfully understates the coterminous nature of linguistic and social class stratification—in Africa, as elsewhere—it does usefully underscore how these class/linguistic distinctions can extend to the types of English language varieties (often agglomerated in cosmopolitan arguments for access to English) also used in these contexts. For example, the English acquired by urban Africans may offer them considerable purchase and prestige for their middle-class identities in African towns, where it often acts as a trade lingua franca, but the same variety of English may well be treated quite differently if they moved to London, identifying them as stigmatized, migrants, and from the lower class. Blommaert (2010) describes the latter as context-specific, “low-mobility” forms of English (p. 195) that do little to change social and economic trajectories in contexts of transmigration.
Returning to the question of Latinos learning Spanish in the United States, canvassed earlier, we see a similar disconnection with empiricism in relation to the leitmotif that the obverse—the maintenance of Spanish—entrenches ghettoization and/or immobility. First, there is the question of context. If Spanish is demonstrably a language of social prestige and mobility in other contexts (as in Spain or Latin America), why cannot it also be so in the United States? And then there is the inconvenient fact of demographics. After all, African Americans have been speaking English for 200 years in the United States and yet many still find themselves relegated to urban ghettos (Macedo, 1994). Racism and discrimination are far more salient factors here than language use. Likewise, English is almost as inoperative with respect to Latino social mobility in the United States as it is with respect to Black social mobility. Twenty-five percent of Latinos currently live at or below the poverty line, a rate that is at least twice as high as the proportion of Latinos who are not English speaking (García, 1995; San Miguel & Valencia, 1998).
All this points to the fact that the lofty advocacy of English as the basis for a new global cosmopolitanism is significantly overstated. Rather, it is existing elites who benefit most from English—or, more accurately, those prestigious varieties of English to which they have preferential access (high-status varieties with normative accents and standardized orthographies). For the majority of other linguistic-minority speakers, the wider structural disadvantages they consistently face, not least poverty, racism, and discrimination, limit, even foreclose, any beneficial effects (Blommaert, Muyllaert, Huysmans, & Dyers, 2006). Acquiring English is thus more often a palliative than a cure, masking rather than redressing deeper structural inequalities. As Peter Ives (2010) concludes, Learning English, or any dominant language, is not inherently detrimental in the abstract, but the context in which it occurs often means that it helps to reinforce psychological, social and cultural fragmentation. Thus a “global language” like English can never fulfill the role cosmopolitanism sets for it, that of helping those marginalized and oppressed by “globalization” to be heard. (p. 530)
If such accounts are demonstrably wrong factually, why do they still hold such sway in discussions of language and education for minority students? The answer lies in the wider sociohistorical and sociopolitical conceptions that still frame what it means to be a citizen in modern liberal democracies. Ironically, this also returns us to the significance of the social and political organization of the modern nation-state system and the role of public education therein. This is ironic because of the almost de rigueur assumption by advocates of cosmopolitanism that the nation-state is in permanent decline and/or already irremediably passé: that we have moved on, in effect, to the next level of social and political organization. 5 Of course, nation-states are clearly subject to wider trends toward greater economic and political interdependence in this age of late capitalism. To suggest otherwise would be foolish. But the idea that this is an entirely new phenomenon, the recent product of globalization in effect, would also appear to be misplaced. After all, nation-states have always been subject to wider economic and political forces of one form or another. As Hinsley (1986) observes, the idea of complete economic and political sovereignty is a “situation to which many states may have often aspired, but have never in fact enjoyed” (p. 226). Meanwhile, for the majority of us, the nation-state model remains the primary social, political, and linguistic frame of reference for our everyday public lives including, crucially, within education (see R. Bauman & Briggs, 2003; Blommaert et al., 2012). In light of this, it behooves us to examine more closely the underlying presumptions that still so influence the more specific debates on language and education before we turn to an examination of possible alternatives.
The Nation-State and the “Pluralist Dilemma”
Debates over citizenship in modern liberal democracies have often focused on the significance of language to both national identity and state citizenship. These debates have addressed, in particular, two key issues:
Whether speaking the state-mandated or national language—that is, the majority or dominant language of the state—is, or should be, a requirement of national citizenship and a demonstration of both political and social integration by its members (especially for those who speak other languages as a first language [L1])
Whether this requirement should be at the expense of, or in addition to the maintenance of other languages—minority, or nondominant languages, in effect—within the state: or to put it another way, whether public monolingualism in the state-mandated language should be enforced on an often-multilingual population or whether some degree of public as well as private multilingualism can be supported (see May, 2008b for further discussion)
Needless to say, how these two issues are addressed has significant implications for the development of language policy and the provision of language education in modern nation-states. In particular, they require modern nation-states to address the balance between social cohesion, a key concern of all such states, and the recognition (or lack thereof) of cultural and linguistic pluralism. This often-difficult balancing act between maintaining cohesion on the one hand and recognizing pluralism on the other within modern nation-states has been termed by Brian Bullivant (1981) as the pluralist dilemma. As Bullivant observes, it is “the problem of reconciling the diverse political claims of constituent groups and individuals in a pluralist society with the claims of the nation-state as a whole [italics added]” (p. x); what he elsewhere describes as the competing aims of “civism” and “pluralism.” In an earlier analysis, Schermerhorn (1970) described these countervailing social and cultural forces as centripetal and centrifugal tendencies: Centripetal tendencies refer both to cultural trends such as acceptance of common values, styles of life etc., as well as structural features like increased participation in a common set of groups, associations, and institutions . . . Conversely, centrifugal tendencies among subordinate groups are those that foster separation from the dominant group or from societal bonds in one way or another. Culturally this most frequently means retention and presentation of the group’s distinctive tradition in spheres like language, religion, recreation etc. (p. 81)
How then can the tensions arising from the pluralist dilemma best be resolved in the social and political arena? Drawing on political theory, two contrasting approaches have been adopted in response to this central question, which Gordon (1978, 1981) has described as “liberal pluralism” and “corporate pluralism.” Liberal pluralism, exemplified in the seminal contribution of John Rawls (1971), is characterized by the absence, even prohibition, of any national or ethnic minority group 6 possessing separate standing before the law or government. Its central tenets can be traced back to the French Revolution and Rousseau’s conception of the modern polity as comprising three inseparable features: freedom (nondomination), the absence of differentiated roles, and a very tight common purpose. On this view, the margin for recognizing ethnic, cultural, and linguistic differences within the modern nation-state is very small (Taylor, 1994). In contrast, corporate pluralism or multiculturalism, as it is now more commonly described, involves the recognition of minority groups as legally constituted entities, on the basis of which, and depending on their size and influence, economic, social, and political awards are allocated.
It is clear, however, that for most commentators the merits of liberal pluralism significantly outweigh those of a group rights or multiculturalist approach. In effect, the answer to the pluralist dilemma has been consistently to favor civism over pluralism. On this basis, the “claims of the nation-state as a whole”—emphasizing the apparently inextricable interconnections between social cohesion and national homogeneity—have invariably won the day over more pluralist conceptions of the nation-state where ethnic, linguistic, and cultural differences between different groups are accorded some degree of formal recognition. As such, formal differentiation within the modern nation-state on the grounds of minority group association is rejected as inimical to the individualistic and meritocratic tenets of liberal democracy. Where countenanced at all, alternative ethnic, cultural, or linguistic affiliations should be restricted solely to the private domain since the formal recognition of such alternative collective identities is viewed as undermining personal and political autonomy and fostering social and political fragmentation. As the political philosopher Will Kymlicka (1989) observes of this, “The near-universal response by liberals has been one of active hostility to minority rights . . . schemes which single out minority cultures for special measures . . . appear irremediably unjust, a disguise for creating or maintaining . . . ethnic privilege” (p. 4). Any deviation from the strict principles of universal political citizenship and individual rights is seen as the first step down the road to apartheid. Or so it seems. The resulting liberal consensus is well illustrated by Brian Bullivant (1981) himself: Certain common institutions essential for the well-being and smooth functioning of the nation-state as a whole must be maintained: common language, common political system, common economic market system and so on. Cultural pluralism can operate at the level of the private, rather than public, concerns such as use of ethnic [sic] language in the home . . . But, the idea that maintaining these aspects of ethnic life and encouraging the maintenance of ethnic groups almost in the sense of ethnic enclaves will assist their ability to cope with the political realities of the nation-state is manifestly absurd. (p. 232)
These emphases on civism at the expense of pluralism help to explain the fragility of the apparent gains for multicultural and bilingual education initiatives up until the 1990s and the significant retrenchment they have since faced internationally. They also undergird many of the presumptions that inform the arguments against the maintenance of minority languages via public education on the basis of social mobility (and integration), as outlined in my earlier discussion of cosmopolitanism. How, then, can one respond effectively (if at all) to this confluence of opinion, and associated ideological and structural imperatives, arraigned against language education provision for linguistic-minority students? In two ways: via the promotion of language rights, particularly in relation to both international and national law, and by an associated exposition and defense of the efficacy of minority language educational approaches for linguistic-minority students. Moreover, these need to be deployed in conjunction if they are to have any immediate—let alone, lasting—effect in shifting the locus of debate in favor of such provision. Let me outline each of these crucial, interconnected, elements in turn.
The Argument for Language Rights
International Law
Much of the post–Second World War era has been largely antipathetic to the notion of language rights for linguistic minorities. This is because the prevailing emphasis in international law, most clearly encapsulated by the 1948 United Nations (UN) Universal Declaration of Human Rights, has been on a view of human rights as primarily, even exclusively, individual rights. In similar vein to the notion of civism, discussed above, what follows from this is a view of rights that addresses the person only as a political being with rights and duties attached to his or her status as a citizen. Such a position does not countenance private identity, including a person’s communal membership, as something warranting similar recognition. These latter dimensions are excluded from the public realm because their inevitable diversity would lead to the complicated business of the state mediating between different conceptions of “the good life” (Dworkin, 1978, 2000; Rawls, 1971, 1999). On this basis, personal autonomy—based on the political rights attributable to citizenship—always takes precedence over personal (and collective) identity and the widely differing ways of life that constitute the latter. In effect, personal and political participation in liberal democracies, as it has come to be constructed post–Second World War, ends up denying group difference and posits all persons as interchangeable from a moral and political point of view (Young, 1993, 2000).
This poses immediate difficulties for the advocacy of language rights because the right to the maintenance of a minority language within any given context presupposes that the particular language in question constitutes a collective or communally shared good of a particular linguistic community (de Varennes, 1996, 2001; Thornberry, 1991). Or, put more simply, arguments for maintaining a language necessarily require a wider language community with which to interact in the first instance. An additional challenge is that language rights are often linked directly, as we have seen above, with the (unnecessary) promotion of ethnic particularism at the perceived expense of wider social and political cohesion. The prominent sociolinguist Joshua Fishman (1991) ably summarizes this view: Unlike “human rights” which strike Western and Westernized intellectuals as fostering wider participation in general societal benefits and interactions, “language rights” still are widely interpreted as “regressive” since they would, most probably, prolong the existence of ethnolinguistic differences. The value of such differences and the right to value such differences have not yet generally been recognised by the modern Western sense of justice. (p. 72)
Even so, there is a nascent consensus within international law on the validity of minority language and education rights. This is predicated on the basis that the protection of minority languages does fall within generalist principles of human rights (May, 2011; see also Kymlicka, 2001, 2007). Following from this, there is a growing acceptance of differentiated linguistic and educational provision for minority groups within some national contexts. These developments, which I will discuss in more detail shortly, can first be usefully referenced by a distinction that another prominent sociolinguist, Heinz Kloss (1971, 1977), has made between what he terms tolerance-oriented rights and promotion-oriented rights (see also Macías, 1979). 7
Tolerance-oriented rights ensure the right to preserve one’s language in the private, nongovernmental sphere of national life. These rights may be narrowly or broadly defined. They include the right of individuals to use their L1 at home and in public; freedom of assembly and organization; the right to establish private cultural, economic, and social institutions wherein the L1 may be used; and the right to foster one’s L1 in private schools. The key principle of such rights is that the state does “not interfere with efforts on the parts of the minority to make use of [their language] in the private domain’” (Kloss, 1977, p. 2).
In contrast, promotion-oriented rights regulate the extent to which minority rights are recognized within the public domain, or civic realm of the nation-state. As such, they involve “public authorities [in] trying to promote a minority [language] by having it used in public institutions—legislative, administrative and educational, including the public schools” (Kloss, 1977, p. 2). Again, such rights may be narrowly or widely applied. At their narrowest, promotion-oriented rights might simply involve the publishing of public documents in minority languages. At their broadest, promotion-oriented rights could involve recognition of a minority language in all formal domains within the nation-state, thus allowing the minority language group “to care for its internal affairs through its own public organs, which amounts to the [state] allowing self government for the minority group” (p. 24). The latter position would also necessarily require the provision of state-funded minority language education as of right.
It is this notion of promotion-oriented rights that most concerns us here since it impinges directly on the issue of the provision of minority language education. And there are increasing examples in international law that support this position, albeit with the proviso that individual nation-states may still choose to ignore such laws, or apply them in only limited ways. One of the most significant of these developments is the UN Declaration on the Rights of Persons Belonging to National or Ethnic or Religious Minorities, adopted in December 1992. This UN Declaration recognizes that the promotion and protection of the rights of persons belonging to minorities actually contributes to the political and social stability of the states in which they live (Preamble). Consequently, the Declaration asserts, Persons belonging to national or ethnic, religious and linguistic minorities . . . have the right [italics added] to enjoy their own culture, to profess and practise their own religion, and to use their own language, in private and in public [italics added], freely and without interference or any form of discrimination. (Article 2.1)
In addition, and significantly, Article 2.1 recognizes that minority languages may be spoken in the public as well as the private domains, without fear of discrimination—a clear, promotion-oriented language right. That said, the 1992 UN Declaration remains a recommendation and not a binding covenant—in the end, it is up to nation-states to decide if they wish to comply with its precepts. In a similar vein, the actual article that deals with minority language education (Article 4.3) qualifies the more general positive intent of Article 2.1 considerably: “States should take appropriate measures so that, wherever possible, persons belonging to minorities have adequate opportunities to learn their mother tongue or to have instruction in their mother tongue [italics added]” (see Skutnabb-Kangas, 2000, pp. 533–535, for an extended discussion).
Another example where exactly the same question applies can be found in the 2007 UN Declaration on the Rights of Indigenous Peoples (UNDRIP). The UNDRIP was formulated over a 25-year period. This included the development over more than 10 years of the 1993 Draft Declaration by the Working Group on Indigenous Populations, in turn a part of the UN’s Sub-Commission on the Prevention of Discrimination and Protection of Minorities. The merits of the Draft Declaration were subsequently debated for nearly 15 years, with many UN member states raising substantive and repeated objections to its promotion of greater self-determination for indigenous peoples (see Xanthaki, 2007, for a useful overview). Despite these objections, UNDRIP retained its strong assertion of indigenous rights, including specific promotion-oriented language and education rights. Article 14.1 states, for example, “Indigenous peoples have the right to establish and control their educational systems and institutions providing education in their own languages, in a manner appropriate to their cultural methods of teaching and learning.” The key question remains though: Given many nation-states’ reservations about the document, what likelihood is there of its meaningful implementation within those states?
Other developments in pan-European law also reflect these competing tensions between, on the one hand, a greater accommodation of promotion-oriented minority language and education rights, and on the other, the ongoing reticence of nation-states to accept such a view. The 1992 European Charter for Regional or Minority Languages is another such example. It provides a sliding scale of educational provision for national and regional minority languages (but not immigrant languages), which ranges from a minimal entitlement for smaller groups—preschool provision only, for example—through to more generous rights for larger minority groups such as elementary and secondary/high school language education. Again, however, European nation-states have discretion in what they provide, on the basis of both local considerations and the size of the group concerned. These nation-states also retain considerable scope and flexibility over which articles of the Charter they actually choose to accept in the first place. In this respect, they are only required to accede to 35 out of 68 articles, although 3 of the 35 articles must refer to education. The process here is twofold. A state must first sign the Charter, symbolically recognizing its commitment to the Charter’s values and principles. Following this, states can ratify the treaty—formally recognizing, in this case, which particular regional or minority languages within the state are to be recognized under the treaty’s auspices. On this basis, 33 European states have since signed the Charter, although only 24 of these have actually since ratified it (Council of Europe, 2008; Grin, 2003; Nic Craith, 2006). 8
A similar pattern can be detected in the 1994 Framework Convention for the Protection of National Minorities, which was adopted by the Council of Europe in November 1994 and finally came into force in February 1998. The Framework Convention allows for a wide range of tolerance-based rights toward national minorities, including language and education rights. It also asserts at a more general level that contributing states should “promote the conditions necessary for persons belonging to national minorities to maintain and develop their culture, and to preserve the essential elements of their identity, namely their religion, language, traditions and cultural heritage” (Article 2.1). That said, the specific provisions for language and education remain sufficiently qualified for most states to avoid them if they so choose (Gilbert, 1996; Thornberry, 1997; Trenz, 2007; Troebst, 1998).
Territorial Language Principle
Despite the ongoing potential for noncompliance at the level of the nation-state, the increasing endorsement of promotion language rights in international law has also been reflected in some national and regional contexts. Where this has occurred, it has usually been based on one of two organizing principles. The first is the “territorial language principle,” which grants language rights that are limited to a particular territory in order to ensure the maintenance of a particular language in that area (see Williams, 2008, for a useful overview). The most prominent examples of this principle can be found in Catalonia, Wales, and Québec, as well as in Belgium and Switzerland. Taking one of these examples, Catalonia has since 1979 used the territorial language principle to (re)establish Catalan as the working language of regional government and of education. Prior to this time, Catalan had been publically banned by the 40-year dictatorship of Franco’s Spain (1936–1975) in favor of Castilian Spanish—this despite (or perhaps because of) Catalan boasting over 9 million speakers in Catalonia and throughout Europe (see May, 2012, chap. 7, for further discussion). With the advent, post-Franco, of the 1978 Spanish Constitution, however, a far greater emphasis was placed on the recognition of Spain’s cultural and linguistic pluralism—granting specific cultural and linguistic rights, as well as a considerable degree of political autonomy, to the different national minorities within the Spanish state. The result was the establishment of 17 “autonomias,” or autonomous regions within Spain, of which Catalonia was the first, as formalized by the 1979 Catalan Statute of Autonomy. The 1979 statute asserts that Catalan is “la llengua pròpia de Catalunya” (Catalonia’s own language) and that “Catalan is the official language of Catalonia as Castilian is the official language of the whole of the Spanish state” (Article 3.2; see Artigal, 1997, p. 135; Guibernau, 1997, p. 96). The statute also stipulates a specific plan of action: The Generalitat [Catalan regional government] . . . will guarantee the normal and official use of both languages, adopt whatever measures are deemed necessary to ensure both languages are known, and create suitable conditions so that full equality between the two can be achieved as far as the rights and duties of the citizens of Catalonia are concerned. (Article 3.3; cited in Strubell, 1998, p. 163)
The reestablishment of Catalan as a civic language since 1979 has been achieved by embarking on an extensive language policy program within Catalonia itself. The principal instrument of this program, at least initially, was the 1983 Llei de Normalització Lingüística (Law of Linguistic Normalization), also known as the “Charter of the Catalan Language.” Linguistic normalization in the Catalonian context was first described by the Congress of Catalan Culture (1975–1977) as “a process during which a language gradually recovers the formal functions it [has] lost and at the same time works its way into those social sectors, within its own territory, where it was not spoken before” (cited in Torres, 1984, p. 59). In this light, and following Fishman (1991), the process of linguistic normalization subsequently embarked on in Catalonia can be described as having three broad initial aims:
To achieve the symbolic promotion and functional institutionalization of Catalan in all key public and private language domains, including, crucially, within schools
To redress illiteracy in Catalan, and any remaining sense of inferiority attached to Catalan, both legacies of the Franco years
Via a “policy of persuasion” (Woolard, 1989), to gain the commitment of first-language Spanish speakers to Catalan, while at the same time countering any hostility toward Catalan as a perceived “threat” to Spanish as the language of the Spanish state
Over the course of the past 30 years, Catalonia has been largely successful in achieving the first two objectives, although the third remains contested, with conservative elements elsewhere in Spain still viewing Catalan autonomy as a potential threat to the Spanish state (DiGiacomo, 1999; May, 2011). Nonetheless, as a result of these language policy developments, today Catalonian citizens have the right to use Catalan on all (public and private) occasions, while virtually all written and oral work in the Generalitat and local authorities is now undertaken in Catalan. More pertinently for our purposes, Catalan is also now the language of education. The latter was crucially facilitated by the 1998 Catalan Linguistic Policy Act, which supported the legal consolidation of Catalan language policies in schools and the wider civil service. The former was achieved by fully implementing unified Catalan immersion education, the latter by further strengthening formal Catalan language requirements for civil servants working in the Catalonian Generalitat and in local authorities. A second objective of the 1998 Act was to increase the presence of Catalan in the media and commerce fields (in which Castilian Spanish remains dominant), principally via the introduction of minimum Catalan language quota systems in the media and the requirement of bilingual service provision in the commercial sector (Pujolar, 2007). In this latter respect, the Act specifically called for private companies to implement programs and measures in support of the further use of Catalan at work. The third objective of the Act was a more broad-based one, to achieve full equality or comparability between Catalan and Spanish in all formal language domains. This included not only the devolved areas of administration regarded as the responsibility of the Catalonian Generalitat but also those areas that still remained under the jurisdiction of the Spanish central government, notably the judicial system, law and order, and tax administration (Costa, 2003).
More recently, an updated (2006) Catalan Statute of Autonomy further emphasizes the separate regional identity of Catalonia, as distinct from the wider Spanish state, and Catalan as its territorial language. As Colino (2009) summarizes it, Following the notion that Catalan is Catalonia’s only “proper” language, considering Spanish as simply the official state language, the new statute establishes Catalan as the language of preferential use in public administration bodies, in the public media and as the language of normal use for instruction in the education system, now extending this to university education . . . [it requires citizens] to know the regional language . . . thus equalizing its status with that of Spanish in the constitution. It also introduces the so-called obligation of linguistic availability, which imposes the obligation for businesses and establishments of answering its users or consumers in the language of their choice. (p. 275)
In addition, the 2006 statute extends promotion-oriented language rights within the region to the Aranese variant of Occitan, spoken in the subregion of Val d’Aran, and to Catalan sign language, both of which are elevated to co-official languages alongside Catalan and Castilian. The new statute also directly addresses the rights of other language speakers, particularly the growing number of Arabic and Urdu speakers who have migrated to the region in increasing numbers over the past decade. Specific tolerance-oriented language rights are now provided so that other language speakers can maintain these languages privately, if they so choose, as well as ensuring access to key services and opportunities to learn the official languages of the region (París, 2007).
Personality Language Principle
The second approach to the promotion of minority language rights is predicated on the “personality language principle,” which attaches language rights to individuals, irrespective of their geographical position. This provides greater flexibility than the territorial language principle in the apportionment of group-based language rights, although it also has its strictures. The most notable of these is the criterion “where numbers warrant”—that is, language rights may be granted only when there are deemed to be a sufficient number of particular language speakers to warrant active language protection and the related use of such languages in the public domain. Canada adopts the personality language principle, where numbers warrant, in relation to French speakers outside of Quebec, via the 1982 Canadian Charter of Rights and Freedoms. A similar approach is adopted in Finland with respect to first-language Swedish speakers living there. Swedish speakers can use their language in the public domain in those local municipalities where there are a sufficient number of Swedish speakers (currently, at least 8%) for these municipalities to be deemed officially bilingual. With over 200 language varieties spoken across 30 states and five Union territories, India though, provides perhaps the best example of this principle in operation. On one hand, we have seen in India the long-standing promotion of English, and more recently Hindi, as the state’s elite, pan-Indian, languages. On the other hand, there are 18 languages recognized in India as “principal-medium languages,” which, in addition to English and Hindi, include 16 official state languages. The division of India’s states along largely linguistic grounds means that local linguistic communities have control over their public schools and other educational institutions. This, in turn, ensures that the primary language of the area is used as a medium of instruction in state schools (see Daswani, 2001; Schiffman, 1996). Indeed, dominant regional language schools account for 88% of all elementary schools in India (Khubchandani, 2008). But not only that, the Constitution of India (Article 350A) directs every state, and every local authority within that state, to provide “adequate” educational facilities for instruction in the L1 of linguistic minorities, at least at elementary school level. As a result, over 80 minority languages are employed as mediums of instruction in elementary schools throughout India.
Recent developments in both international and national law are, then, at once both encouraging and disappointing. The principle of separate minority recognition in language and education is legally enshrined at least as a minimal tolerance-oriented right—that is, unhindered ongoing language use when restricted to the private domain—in almost all modern nation-states 9 and is reinforced by clear intent in instruments of international law. However, more liberal interpretations of tolerance-oriented rights (involving some state support where numbers warrant), and certainly more promotion-oriented rights, are less secure and remain largely dependent on the largesse of individual nation-states in their interpretation of international (and national) law with respect to minorities. As a result, there are as yet no watertight legal guarantees for the recognition and funding of minority language and education rights, although there is a clearly articulated basis for them.
At the very least though, there is an increasing recognition within both international law, and in a growing number of national contexts, that significant minorities within the nation-state have a reasonable expectation to some form of state support (Carens, 2000; de Varennes, 1996). In other words, although it would be unreasonable for nation-states to be required to fund language and education services for all minorities, it is increasingly accepted that where a language is spoken by a significant number within the nation-state, it would also be unreasonable not to provide some level of state services and activity in that language. In combination, this provides an ever-strengthening basis for minority language education rights.
The Case for (and Against) Bilingual Education
Having addressed the basis for language rights in both international and national law, let me now turn to the key educational tenets underpinning a more linguistically diverse approach to language learning within schools. This is crucial, because contra to the oft-stated position that the best means of learning English is via English-only instruction, the best means of learning English (or any additional language) is actually via one’s L1. Indeed, it has long been demonstrated that English-only education programs are the least effective means of successfully acquiring (bi)literacy for bilingual learners (May, 2010; see also below). The reason for this, of course, is that such programs demonstrably fail to build on the existing linguistic repertoires of bi/multilingual learners—instead, willfully ignoring and/or excluding them from the teaching and learning process (May, 2014). Moreover, their principal aim is usually for learners eventually to become monolingual in English at the specific expense of their bi/multilingualism—as the earlier accounts of Archibugi (2005), Pogge (2003), and the like would clearly seem to suggest.
In contrast, extensive international research on bilingual education consistently finds that long-term or “late-exit” bilingual programs are the most effective educationally for bilingual students, followed by short-term or “early-exit” bilingual education programs. The academic literature supporting these conclusions is voluminous and I am not able to explore it in detail here (see C. Baker, 2011; García, 2009; May, 2008a, for useful recent overviews). But I will discuss briefly, by way of example, two major studies, both conducted in the United States, which clearly support these general conclusions.
Ramírez, Yuen, and Ramey (1991) compared English-only programs with early-exit and late-exit bilingual programs, following 2,352 Spanish-speaking students over 4 years. Their findings clearly demonstrated that the greatest growth in mathematics, English language skills, and English reading was among students in late-exit bilingual programs where students had been taught predominantly in Spanish (the students’ L1). For example, students in two late-exit sites that continued L1 instruction through to Grade 6 made significantly better academic progress than those who were transferred early into all-English instruction. Ramírez et al. conclude, Students who were provided with a substantial and consistent primary language development program learned mathematics, English language, and English reading skills as fast or faster than the norming population in this study. As their growth in these academic skills is atypical of disadvantaged youth, it provides support for the efficacy of primary language development facilitating the acquisition of English language skills. (pp. 38–39)
In contrast, the Ramírez et al. (1991) study also confirmed that minority language students who receive most of their education in English rather than their L1 are more likely to fall behind and drop out of school. In fact, it is important to note here that the English-only programs used for comparison in the Ramírez study were not typical to the extent that although the teachers taught in English, they nonetheless understood Spanish. This suggests that in the far more common situation where the teacher does not understand the students’ L1, the trends described here are likely to be further accentuated.
In the largest study conducted to date, Thomas and Collier (2002) came to broadly the same conclusions. Between 1996 and 2001 Thomas and Collier analyzed the education services provided for over 210,000 language minority students in U.S. public schools and the resulting long-term academic achievement of these students. As with the Ramírez et al. (1991) study, one of Thomas and Collier’s (2002) principal research findings was that the most effective programs—“feature-rich” programs as they called them—resulted in achievement gains for bilingual students that were above the level of their monolingual peers in English-only classes. Another key conclusion was that these gains, in both L1 and second language (L2), were most evident in those programs where the students’ L1 was a language of instruction for an extended period of time. In other words, Thomas and Collier found that the strongest predictor of student achievement in L2 was the amount of formal L1 schooling they experienced: “The more L1 grade-level schooling, the higher L2 achievement” (p. 7). Thomas and Collier also found that students in English-only classes performed far less well than their peers in late-exit bilingual programs, as well as dropping out of school in greater numbers. Students in early-exit, transitional, bilingual programs demonstrated better academic performance over time but not to the extent of late-exit bilingual programs. In both these major studies, then, length of L1 education turned out to be more influential than any other factor in predicting the educational success of bilingual students, including socioeconomic status.
There are a wide range of other studies, both from the United States and from other national contexts, that broadly corroborate these findings. 10 Suffice it to say, these findings, when taken cumulatively, unequivocally support the efficacy of late-exit bilingual education 11 and thus stand in direct contradiction to Archibugi (2005), Pogge (2003), and Laitin and Reich’s (2003) conclusions about bilingual education for Latino students, discussed earlier. English-only instruction ensures neither “fluency in English” (Pogge, 2003) nor participation in “broader societal culture” (Laitin & Reich, 2003). Indeed, it is most often the cause of, rather than the solution to, the educational failure of bilingual students. Why then, given the widely attested efficacy of bilingual education, does it still so seldom translate into policy and practice in the language education provision for linguistic-minority students? Again, the answer is primarily a sociopolitical/sociohistorical one, rather than an educational one per se. Returning to the pluralist dilemma, the imperative of civism, which, in this case, requires the linguistic assimilation of minority students as the “price” of citizenship, outweighs any arguments for the educational efficacy of bilingual education and/or the educational and wider benefits that would result for those students. Or, to put it another way, research evidence that appears to contradict the dominance of English and/or the related ideology of public monolingualism is not a welcome feature for those wishing to buttress or ensure them. As Thomas Ricento (1996) observes of the U.S. context, for example, in spite of an impressive amount of both qualitative and quantitative research now available on the merits of bilingual education, “The public debate (to the extent that there is one) [still] tends to focus on perceptions and not on facts” (p. 142). Or as Joshua Fishman (1992) despairingly asks of the same context, “Why are facts so useless in this discussion?” (p. 167). This laissez-faire approach to the facts is no more starkly demonstrated than in the ideology and political advocacy of the so-called official English movement in the United States—or, as I prefer to term it, the English-only movement.
The English-Only Movement
Since its inception in the early 1980s, the English-only movement has been preoccupied with achieving an English Language Amendment (ELA) to the U.S. constitution that would recognize English as the official language of the United States.
12
The aim of an ELA is to ensure that English is the only language of government activity (see Crawford, 2000, 2008) and to “require” immigrants to the United States—and here Latinos are particularly targeted—to learn English (with an implicit, sometimes explicit, agenda that this should be at the specific expense of their L1). To date, all ELA proposals at the federal level have failed. However, the English-only movement has increasingly turned to the state level to enact its political aims. As Daniels (1990) observes of the latter, “The overall strategy [here] seems to be to get some official-English law on the books of a majority of states and to continually fan public resentment over schooling policies that ‘degrade English’ and ‘cater’ to immigrants” (p. 8). This state-level approach has also included specifically targeting existing bilingual education programs. For example, in a U.S. English advertisement in 1998, the tagline read, Deprive a child of an education. Handicap a young life outside the classroom. Restrict social mobility. If it came at the hand of a parent it would be called child abuse. At the hand of our schools . . . it’s called “bilingual education” (cited in Dicker, 2000, p. 53).
This has also been the leitmotif of Ron Unz, a prominent software entrepreneur and conservative public figure, and his organization, English for the Children. Targeting four key U.S. states that allow for citizens-initiated referenda, Unz has successfully promoted public propositions delimiting or dismantling bilingual education in three of these states. The most prominent of these is California’s Proposition 227 (1998), which saw 61% support the measure overall, including 37% of Latino voters. However, Unz was also successful in sponsoring similar measures in Arizona (2000; 63%) and Massachusetts (2002; 68%). Only in Colorado (2002; 44%) was a comparable measure defeated, although, controversially, this was because those opposing the measure played on the fears of Anglo parents that this would result in the reintroduction of too many Latino students into mainstream classrooms (Crawford, 2007). Suffice it to say, in none of the referenda did research evidence in support of the educational efficacy of bilingual education inform wider public debate. Prior to the Arizona referendum (Proposition 203), for example, the New York Times ran a front-page story in August 2000 strongly supporting the apparently incontrovertible educational merits of its precursor, Proposition 227. Subsequently, it was revealed that the claims made in the article simply repeated key talking points in Unz’s campaign, although, not surprisingly, it still had a powerful effect on strengthening public support for Proposition 203 (Wiley & Wright, 2004; Wright, 2005).
This pattern of deliberate misrepresentation and obfuscation by proponents of English-only continues to the present largely unabated. For example, George W. Bush’s flagship NCLB (2001) saw the revocation of the 1968 Bilingual Education Act on the premise that English language instruction and assessment is best for all students. 13 Subsequent advocacy of NCLB has touted its positive effects for bilingual students. However, research evidence again suggests the opposite is true. Menken (2008) has documented the clearly negative educational effects that NCLB testing policies are having on bilingual students. She found in New York City, for example, that, as a result of NCLB’s requirement to be assessed in English, with no recourse to one’s L1, bilingual students ranged from 20 to 50 percentage points below native English speakers.
Insisting on the merits of English-only education for bilingual students in the face of research evidence that clearly suggests otherwise demonstrates a remarkable degree of cynicism. But this cynicism becomes breathtaking when measured against another variable: the funding of English-only programs. In this respect, the English-only movement is again clearly found wanting. Many who supported the establishment of official state-level English policies, for example, including the Unz-led referenda against bilingual education, logically assumed that a principal concern of the legislation was to expand the opportunities for immigrants to learn English. However, logical or not, this has proved not to be the case. Although the English-only movement spent lavishly to get ELA measures on the ballot, in 1988 it declined to support legislation creating a modestly funded federal program for adult learners of English (Crawford, 1992). As a result of public criticism, the English-only movement did make some subsequent effort to fund similar ventures, but these efforts have remained largely desultory and continue to constitute only the barest minimum of their total funding efforts (Dicker, 2000).
Conclusion
The example of the English-only movement in the United States, and its ongoing political effectiveness in delimiting bilingual education there, should remind us why there has been such an effective retrenchment of more plurilingual approaches to language education over the past two decades, as discussed at the outset of this article. Indeed, my aim in (necessarily) covering so much ground in this account is precisely to highlight how difficult it is to make the case for minority language rights, despite the positive developments in international law and (some) national contexts. This is because of a confluence of ongoing forces arraigned against such rights being recognized and enacted—the predilection toward civism (and a related public monolingualism) at the level of the nation-state, as well as the valorization of English as the new language of cosmopolitanism and of social mobility in an increasingly globalized world. Add to this a lingering antipathy to minority linguistic identification that is, more often than not, constructed by opponents as antediluvian and/or as an attempt at willful ghettoization by minorities, and one can see just what is at stake. These difficulties also explain why bilingual education approaches, despite their attested educational efficacy, are potentially so fragile and easily dismantled as a result of the wider social and political vicissitudes described above.
Even so, as this account also makes clear, much too has been achieved since the advent of multiculturalism in the 1960s/70s, and not all these accomplishments have been retrenched over the past 20 years. Indeed, there is a nascent consensus that suggests that the long-held practice of making no accommodations to linguistic-minority demands is not so readily defensible in today’s social and political climate. Ignoring such demands is also unlikely to quell or abate the question of minority rights, as it might once have done. Indeed, it is much more likely to escalate them, not least because minority groups are far less quiescent about the injustices attendant on their long-standing cultural and linguistic marginalization as the price of civic inclusion in the (monolingual) nation-state (May, 2012). Under these circumstances, as de Varennes (1996) outlines, Any policy favouring a single language to the exclusion of all others can be extremely risky . . . because it is then a factor promoting division rather than unification. Instead of integration, an ill-advised and inappropriate state language policy may have the opposite effect and cause a levée de bouclier. (p. 91)
As for bilingual education, so often the frontline in these wider debates, there is the related responsibility of those well versed in the research literature to articulate clearly and consistently the benefits of more plurilingual approaches to the education of bi/multilingual students—in both academic and the wider public domains. As McGroarty (2006) argues, It is the job of [those] interested in policies that include attention to bilingualism to keep the value of bilingualism in the public consciousness, to continue to demonstrate that bilingual approaches to education are not only feasible but, in fact, actually exist. (2006, p. 5)
In order to achieve this, McGroarty (2006) continues, “Advocates for positive language and education policies must constantly articulate the value of bilingualism, and to be able to do so in varied terms that respond to a protean environment of public discussion” (pp. 5–6). A failure to do so will mean that the clearly attested educational effectiveness of bilingual/immersion programs will continue to be (conveniently) overlooked in these wider public debates, to the inevitable cost of bilingual students. More broadly, it is only from an informed research base, which in turn directly influences language educational policy development, that further progress can be made on realigning the predilection to monolingualism that still so dominates public policy on language education—particularly, but not exclusively, when the language in question is the new global juggernaut of English.