Linguistic disenfranchisement, minority resistance and language revitalisation: The contributions of ethnolinguistic online communities in Zimbabwe

Zimbabwe is a multicultural nation with multiple ethnolinguistic groups. Since 1980, when the country attained its independence from colonial rule, the government has struggled to fit all linguistic and cultural representations in the public domain. This resulted in disenfranchisement and disillusionment of linguistic minorities. However, the subject of ethnicity is sensitive and is to some extent associated with undertones of regionalism and division. Against this background, the disenfranchised linguistic minorities have taken advantage of the liberative potential of social media and are regrouping in the virtual space; in the process forming vibrant ethnolinguistic online communities. This study critically analyses the motivations and purpose of ethnolinguistic online communities on Facebook. This is a netnographic study of 10 purposively selected Facebook groups of linguistic minorities in Zimbabwe. The study is grounded in the concepts such as hegemony, hidden transcripts and alternative media. The findings of the study show that ethnolinguistic online communities provide the disenfranchised linguistic minorities

Abstract: Zimbabwe is a multicultural nation with multiple ethnolinguistic groups. Since 1980, when the country attained its independence from colonial rule, the government has struggled to fit all linguistic and cultural representations in the public domain. This resulted in disenfranchisement and disillusionment of linguistic minorities. However, the subject of ethnicity is sensitive and is to some extent associated with undertones of regionalism and division. Against this background, the disenfranchised linguistic minorities have taken advantage of the liberative potential of social media and are regrouping in the virtual space; in the process forming vibrant ethnolinguistic online communities. This study critically analyses the motivations and purpose of ethnolinguistic online communities on Facebook. This is a netnographic study of 10 purposively selected Facebook groups of linguistic minorities in Zimbabwe. The study is grounded in the concepts such as hegemony, hidden transcripts and alternative media. The findings of the study show that ethnolinguistic online communities provide the disenfranchised linguistic minorities ABOUT THE AUTHORS Phillip Mpofu is a Postdoctoral Research Fellow in the Indigenous Language Media in Africa research entity at North West University, South Africa. He is also a senior lecturer at Midlands State University, Zimbabwe. His research combines sociolinguistics, language policy and planning, language and identity politics and indigenous language media.
Abiodun Salawu is Professor of Journalism, Communication and Media Studies and Director of the research entity, Indigenous Language Media in Africa at the North-West University, South Africa. He has to his credit, scores of scholarly publications in academic journals and books. He has also edited/co-edited six books and authored one. He is a co-vice chair of the journalism section of IAMCR and a member of editorial/advisory boards of a number of journals. He is rated by the NRF as an established researcher and he is a member of the Codesria's College of Senior Academic Mentors. This article consolidates their research interest and engagements in indigenous language print, broadcast and digital media.

PUBLIC INTEREST STATEMENT
Politics of language and belonging are contentious issues in post-independence Zimbabwe. The country is characterised by a complex language ecology which comprises multiple ethnolinguistic identities. Discounting English, language use in the public domain exhibits Shona and Ndebele hegemony. Consequently, the marginalised linguistic minorities are disenchanted. However, debates or studies on ethnicity are often perceived as retrogressive to national integration. Therefore, this study is a contribution to the meagre scholarship on the resistance of linguistic minorities to domination in the context of digital and social media. The study analyses ethnolinguistic online communities on Facebook which consists of geographically dispersed but linguistically and culturally bound members. These online communities function as subaltern counterpublics that contain rejoinders and subversions to the dominant linguistic and cultural representations in the public domain.
with space and agency for protesting against marginalisation and language shift, in the process promoting the use of their languages. This study is a significant contribution to the growing body of scholarship on the nexus between language politics and social media in Zimbabwe.

Introduction
Linguistic identity is largely a political matter and languages are flags of allegiance (Rajagopalan, 2001). Therefore, the presence of a common language is instrumental to the formation of social networks and connections. This means that languages are not entirely linguistic constructs or simply means of communication; rather, they are in actual fact political objects. This is so because a language is a vital component of ethnic identity, and the existence of an ethnic group is always connected to a particular language. However, researching ethnicity has been portrayed as contentious and retrogressive to national integration (Makoni, Dube, & Mashiri, 2006).
Zimbabwe is characteristically a multilingual and multicultural nation which comprises several ethnolinguistic groups which include the Ndebele, Tonga of Mudzi, Nambya, Tswana, Shangani, Sotho, Dombe, Xhosa, Venda, Tonga, Tshwawo, Kalanga, Chibarwe, Sena, Doma, Chikunda, Chewa and Shona (Ndhlovu, 2006). These groups contest for representation in the public spaces such as the media, education and politics. However, language use patterns and cultural representations in these domains demonstrate the supremacy of English, Shona and Ndebele and the marginalisation and exclusion of ethnolinguistic minorities (see Mpofu & Mutasa, 2014;Ndhlovu, 2007). This is shown by what Suarez (2002) calls daily forms of linguistic hegemony which comprise among others language use in the media, institutions and social relationships that associate linguistic minorities with inferiority, low self-respect and belittlement.
With specific reference to Zimbabwe, there is a hierarchical organisation of language use in public spaces. Disregarding the dominance of English, language use patterns and cultural representations in the public domain reflect the dominance of Shona and Ndebele at the expense of the minority languages (Ndhlovu, 2007). In Ndlovu-Gatsheni's (2009 terms, Zimbabwe is plurilinguistic society which is typified by a bimodal ethnicity of the Ndebele and Shona. On another level, though Zimbabwe is generally imagined as divided into Ndebele and Shona identities, Shona enjoys supremacy as a result of socio-political power and cultural domination (see Hachipola, 1998;Ndlovu-Gatsheni, 2013). At the lowest level, Shona and Ndebele ethnolinguistic groups are characteristically marked by heterogeneity. Shona is a collective noun comprising of spatially and linguistically dispersed people with distinguishable sub-ethnolinguistic identities which are the Karanga, Zezuru, Korekore, Manyika and Ndau (Mlambo, 2013). However, the Zezuru subethnolinguistic identity enjoys dominance (Mpofu & Mutasa, 2014). Likewise, the Ndebele group is also a heterogeneous construction which encompasses multiple identities that include the Ndebele of Nguni, Sotho, Tswana, Kalanga, Venda, Tonga, Kalanga, Nyubi, Venda, Birwa and Lozwi and Shona (Mazarire, 2009).
The foregoing indicates that the cultural identity of the Zimbabwean nation is complex since it consists of multiple ethnolinguistic groups and is characterised by the hegemony of other ethnolinguistic representations and the disenfranchisement of ethnolinguistic minorities. For that reason, members of ethnolinguistic minorities are emotionally disenchanted and always feel marginalised and excluded in the public domain. In most cases, individuals belonging to linguistic minorities are forced to speak other people's languages, in the process assuming the dominant linguistic and cultural identities. However, openly discussing ethnicity is criminalised and labelled as tribalism which is retrogressive and divisive to an apparently united Zimbabwean nation (Mhlanga, 2013). Therefore, it is problematic for the marginalised and disillusioned ethnolinguistic minorities to overtly discuss about their situation. Rather, they only do so in cagey and classified settings among themselves.
Be that as it may, the emergence of social media supported by new digital media technologies and internet provides platforms and spaces with liberative potential for free participation (Chibuwe & Ureke, 2016). For that reason, there is remarkable proliferation of lively online-based ethnolinguistic groups on Facebook which are made up of members that share the same language, history and cultural identity. These groups include the Tonga, Sotho, Venda, Karanga, Ndau and Korekore, among others.
Against this background, this study explores ethnolinguistic groups on Facebook-an online social networking site and their use by linguistic minorities in resisting disenfranchisement and language shift. The argument which is advanced in this study is that ethnolinguistic online communities provide the disenfranchised with space and agency for resisting marginalisation and language shift, in the process promoting their languages and cultures. To some extent, ethnolinguistic online groups on Facebook extend the work of language and culture associations of linguistic minorities which for a long time defended the promotion of minority languages. However, in other instances, the groups have degenerated into platforms for political activism, tribalism and prejudiced exclusion.

Linguistic and cultural disenfranchisement in Zimbabwe
Generally, language and language differences are highly poignant subjects. For that reason, the question of "who we are" lie at the core of most language policy conflicts (Schmidt, 2006). In Zimbabwe, the exclusion and marginalisation of citizens on the basis of language and culture are equally emotive issues. Where there is conflict over language policy, the core issue is on how to achieve equality among ethnolinguistic groups. In the Zimbabwean context, the people are rooted in smaller communities and their first loyalties are to an ethnic group and region, such that language differences and regionalism are serious problems that militate against national unity and are basic factors in ethnic conflict. Thus, any attempt to impose a particular language on other ethnic groups often invites conflict or protest.
The pervasive existence of multilingualism and multiculturalism in Zimbabwe is officially recognised in the 2013 Constitution of Zimbabwe Amendment No. 20 Act. Therefore, there is a universal need for multilingual/multicultural policies to ensure that all members of different ethnolinguistic groups have access to public information and participate in national affairs without discrimination. Typical of many African countries whose national boundaries do not follow linguistic boundaries, Zimbabwe consists of more than 18 ethnolinguistic groups (Ndhlovu, 2006). Sixteen languages are listed as "officially recognised" in Section 6 of the Constitution in a list which also includes English, Khoisan, Ndau (previously a dialect of Shona) and sign language. Though Dombe, Sena, Chikunda, Tshwawo and Tonga of Mudzi are excluded, communities which speak these languages do exist. This is an indication that Zimbabwe is a multi-ethnic society, and hence bound to have multiple language and ethnic identities.
However, despite the complicated language ecology, Section 63 of the Constitution of Zimbabwe Amendment No. 20 Act (2013, p. 31) declares that every person has got the right to use "the language of their choice, and to participate in the cultural life of their choice". Nevertheless, identity making is a political process that is mediated through imperatives of inclusion and exclusion (Ndlovu-Gatsheni, 2013); therefore, common linguistic and cultural goals are not followed in multicultural Zimbabwe as other groups are often marginalised. To explain this occurrence, Ndhlovu (2007) stresses that Zimbabwe is characterised by microsocial forms of language-based exclusions among the diverse ethnolinguistic polities. These exclusions manifest in a number of ways that include biased language use patterns in the public domain, negative perceptions and stereotypes about ethnolinguistic minorities and forced assimilation of minority language speakers into majority language groups (Ndhlovu, 2007).
As a result, multi-layered ethnolinguistic hegemonies are visible in the Zimbabwean public spaces (Mpofu & Mutasa, 2014). Shona and Ndebele are the preponderant languages which from colonial times have been recognised as regional lingua francas, languages of the media and education as well as languages of upward socio-political and economic mobility (Ndhlovu, 2007). The colonial regime demarcated the then Rhodesia into the Shona-and Ndebele-speaking regions. In the post-independent period, Shona and Ndebele continued to be dominant. Prior to the introduction of minority languages such as Sotho, Kalanga and Tonga in schools, minority language speakers were compelled to learn Ndebele as a home language. More so, mainstream print and broadcast media have failed to fairly represent the all languages in their content.
While the common denominator of existing literature on language policy in Zimbabwe revolves around the protection of indigenous languages and the negative effects of English supremacy in the public domains, Ndhlovu (2009) exonerates English from that denunciation as a killer language in Zimbabwe. Instead, he regards Shona and Ndebele as "killer languages" which creep up the survival of other indigenous languages. This means that Shona and Ndebele are threat to linguistic diversity in Zimbabwe. For instance, speakers of Ndau, Barwe, Shangani and Hwesa are spoken in predominantly Shona-speaking regions, whilst Sotho, Tonga, Venda, Kalanga and Tswana, among others, are spoken in evidently Ndebele-speaking regions. Thus, the political influence of the Shona-and Ndebele-speaking people strangles the ethnolinguistic minorities, threatening their very existence. Muzondidya and Ndlovu-Gatsheni (2007) also concur that many ethnic groups, particularly the minority language speakers were put into ethnicised administrative entities and their real identities are overlooked.
The above exposes what Kembo-Sure (2000) call conflict linguistics where two or more languages or varieties of languages that are in contact are also in a state of rivalry, resulting in tensions where one language dominates or threatens the existence of other languages. As demonstrated in existing literature, there is inherent competition among Zimbabwean local languages especially between the languages of the majority ethnic groups and languages spoken by minorities. At micro-level, there is a relative competition among sub-varieties of the dominant languages where other language varieties occupy respected positions. Generally, ethnolinguistic minorities are neglected. This study shows how this politics of language manifest on social media.
The foregoing has shown that Zimbabwe is a multilingual society, but is projected as a basically Ndebele and Shona nation. As a result, there is a growing body of literature on the widespread marginalisation of minority languages (see Makoni, Makoni, & Nyika, 2008;Mpofu & Mutasa, 2014, Ndhlovu, 2006, Nyika, 2008. In real life, one unavoidable effect of language competition is tension among speakers because speakers of non-dominant languages feel marginalised (Mustapha, 2013). Paradoxically, it is the much disparaged language English which in turn plays the neutral function as lingua franca that brings members of different ethnolinguistic groups together. The choice of an indigenous language as lingua franca is therefore not a conceivable decision.
Though Shona and Ndebele are often considered to be dominant indigenous languages, Shona is comparatively more dominant than Ndebele (Hachipola, 1998;Ndlovu-Gatsheni, 2013). With reference to broadcast media, Mpofu and Mutasa (2014) argue that Shona is more prominent as shown by the number of programmes and its use in programmes of national significance, in most cases with no Ndebele parallel programmes. This validates unconcealed Shona hegemony over Ndebele. Hence, the propagation of the politics of Ndebele particularism, an existing determination for the re-establishment of a sovereign Ndebele nation separate from the current Shona-centred territorial nationalism of Zimbabwe (Ndlovu-Gatsheni, 2013). Though, the Ndebele of Zimbabwe are basically an ethnic group, promoters of Ndebele particularism prefer the term nation or state instead of ethnic group or community. They endeavour to disconnect themselves from the Shona socio-historical experiences and envisage revivification of the Ndebele Empire. This is a reflection of the subtle language based struggles of those who experience marginalisation. This study demonstrates how promoters of Ndebele particularism establish their ambitions through the formation of Ndebele particularistic groups on Facebook.
The concepts of language and dialect are not exclusively linguistic constructs. The primary conditions for a language to be recognised as such are political. A language, as it has often been remarked with a rather dismissive shrug, is a dialect with an army and a navy (Rajagopalan, 2001). With regard to the Shona language, the possibility of complete mutual comprehension fails to produce satisfactory results when we confront practically a continuum of subvarieties of Karanga, Zezuru, Korekore and Manyika. The fact that prior to 2013 Ndau was a variety of Shona and was raised to the status of an officially recognised language in the new Constitution shows that Shona is a socio-political construct. It is not a monolithic language group as it is comprised spatially and linguistically separate sub-ethnic groups (Mazarire, 2009). Besides Zezuru which was used as the central dialect in the standardisation of Shona, the speakers of other dialects (Karanga, Korekore and Manyika) are unreservedly disgruntled for being forced to write, not the way they speak, or being belittled in the media and everyday jokes. This study shows that in the era of digital and social media, there are ethnolinguistic online communities on Facebook which protest against the marginalisation of other language varieties of Shona and challenge Zezuru supremacy. There are also other language-based language groups of other officially recognised languages such as Hwesa and Ndau which challenge Shona hegemony in the Shonaspeaking region.
Similarly, Ndebele is a socio-political creation with various identities, where members of the "royal class" fight to retain their position and the co-opted fight for their emancipation. Thus, despite the existence of the politics of Ndebele particularism, the concept of Ndebele identity itself is not a self-evident concept. For that reason, it is practically contestable with questions being raised on who qualifies as proto-Ndebele. The Ndebele of Zimbabwe today are a community which is characterised by assimilation and incorporation of people from different ethnic groups (Mazarire, 2009). According to Ndlovu-Gatsheni's (2013) right from the founding of the Ndebele kingdom by Mzilikazi, people of Shona origin were also assimilated into the Ndebele state. The Ndebele state were classified into categories according to their origin, with the "abeZansi" (those from the South) forming the aristocratic group, the "abeNhla" (those from the North) constituting those assimilated from the Rozvi state. Ultimately, Ndebele ethnolinguistic group in is a combination of identities of multiple origins that were ordered in a hierarchy of belonging including Ndebele of Nguni, Sotho, Tswana, Kalanga, Venda, Tonga, Kalanga, Nyubi, Venda, Birwa and Shona. As a result, some members of the Ndebele of today consider themselves as "proto-Ndebeles" who belong to "abeZansi" (those from the South) class. Furthermore, this study observes that there are individuals who feel that they were "forcibly" co-opted into the Ndebele ethnic identity by historical, educational or geographical circumstances, and yet they have distinct ethnolinguistic identities. These language groups include the Sotho, Tonga, Kalanga, Nambya and Venda who converge on Facebook and nostalgically reflect on their real identities using their native languages.

Minority resistance to linguistic and cultural disenfranchisement
Thinking about hegemony should provoke not only consideration of structures constraining agency, but also the agency of countering structures of dominance (Sonntag, 2003, p. 7). Existing scholarship on post-independence Zimbabwe's language policy and practices has largely focused on the effects of English, Shona and Ndebele on minority languages, without considering the forms of resistance adopted by the subjugated linguistic minorities. This study focuses on the agency of the disenfranchised ethnolinguistic groups towards re-asserting and revitalising their linguistic and cultural identities using the wired space. While we acknowledge that resistance of minorities can manifest in various ways, this study reflects on how new digital technologies and social media provide them with space to exercise agency to counterattack.
One fascinating strategy of resistance which is described by Eriksen (1992) is cultural assimilation where proficiency in the dominating language is regarded as a successful strategy of resistance. He contends that unexpectedly "cultural minorities may have to assimilate culturally in important respects in order to present their case effectively and thereby retain their minority identity" (p. 313). This means that the paradox of linguistic hegemony is that one must accept it in order to resist it. This is the case with ethnolinguistic minorities in Zimbabwe since the majority are proficient in at least one of the dominant indigenous languages.
However, the need to resist remains concealed and manifests with each and every chance that comes. As this study shows, the emergence of social media such as Facebook provides ethnolinguistic minorities with opportunity to resist domination by making use of the Facebook group function to create online communities. This study further shows the impact of online communities towards the revitalisation of previously marginalised languages.
Resistance can be motivated by the need to fight against language death and endangerment which can be necessitated by what Fishman (1991) calls language shift. That is the loss of critical structures which sustains the survival of a spoken language which include the loss of speakers and domains of use. As witnessed in Zimbabwe, the dominated languages such as Sotho, Kalanga, Tonga and Shangani are restricted in use as the dominant language always encroaches in their territories. Minority languages exist in diglossic circumstances where they are in danger of extinction. Therefore, resistance against marginalisation by ethnolinguistic minorities is justified as a fight against loss of linguistic diversity.
The other reason for resisting linguistic disenfranchisement is based on the linguistic human rights perspective, where disenfranchising languages is likened to excluding people who speak those languages. The notion of linguistic human rights is based on the observation that dominant languages of nation states penetrate into, transform and undermine the minority communities' ability to maintain their languages, cultures and identities. Language is one of the most important human attributes on the basis of which people are not supposed to be prejudiced. Hence, Skutnabb-Kangas (2006) has referred to language shift occurring within the context of forced assimilation as linguistic genocide. Thus, the observation of linguistic human rights is one way of preventing linguistic genocide, promoting integration and defending people resist assimilation and promoting linguistic diversity. Observing linguistic rights suggest that every ethnic group have right to use their language in schools and other public domains. This is because linguistic interruption for a particular community of speakers rarely occurs in isolation from sociocultural and socioeconomic dislocation (Fishman, 1991).
Because of the political and emotive nature of the subject of language, especially language differences, the marginalised minorities are stirred to resist. They may adopt conventional or unconventional means to voice their displeasure and lobby for the promotion and inclusion of ethnolinguistic identities in the public domain. The formation of language associations representing different ethnolinguistic minorities in Zimbabwe is one traditional strategy that has been adopted to resist hegemonic languages. These include Tonga Language and Cultural Committee, Kalanga Language Committee, Nambya Language Committee, Shangani Language and Cultural Committee, Sotho Language and Literature Committee and Venda Language Committee. These committees are affiliates to Indigenous Languages Promotion Association which was formed in 2001 to campaign for the revitalisation of indigenous languages (Nyika, 2008). This demonstrates that linguistic minorities, through their associations, have a role to play in making sure that their languages get recognition and are developed.
Academic voices have also expressed concern about the disenfranchisement of ethnolinguistic minorities of Zimbabwe. This is demonstrated by how the dominance and marginalisation of other ethnolinguistic identities have been subjected to academic enquiry. They have acted as language advocates who are committed to defending the status of minority languages in the academic sphere. Ndlovu-Gatsheni (2009), Ndhlovu (2007Ndhlovu ( , 2009 and Nyika (2008), among many others, have decried the hegemony of Shona and Ndebele over ethnolinguistic minorities. On the other hand, Ncube and Munoriyarwa (2017) demonstrate the manifestation of language politics in football fandom, especially during matches involving Dynamos FC and Highlanders FC. Innovatively, this study shows how such discourses permeate the online communities on Facebook.
Through research activities, the African Languages Research Institute (ALRI), a research entity at the University of Zimbabwe has played a pivotal role in the promotion and revitalisation of minority languages. ALRI is an interdisciplinary research unit which is dedicated to the development of all indigenous languages. Its goals are to "research, document and develop the Zimbabwean indigenous languages in order to promote and expand their use in all spheres" (Chabata, 2007, p. 281). Civic organisations such as Catholic Commission for Justice and Peace, Silveira House and Save the Children have been also instrumental in promoting minority languages (Nyika, 2008).
It is also worth noting that resistance to linguistic and cultural disenfranchisement has also manifested in political movements and action. For instance, there are Ndebele particularistic movements that have emerged which contest the notion of the Zimbabwean nation. For instance, the online community United Mthwakazi Republic advocates the restoration of pre-colonial Ndebele nation in the mould of Swaziland and Lesotho (Ndlovu-Gatsheni, 2008). Likewise, it is also difficult to separate ZANU Ndonga's dominance in Manicaland town Chipinge from the Ndau particularistic and cultural nationalism. This study demonstrates how such ethnic-based actions have pervaded social media through ethnolinguistic groups on Facebook which are littered with discourses of resistance to domination. This section has demonstrated the expediency of Stroud's (2001, p. 339) metaphor "language as a site of struggle". Thus, language politics and ethnolinguistic disenfranchisement are pervasive. However, because of the political and emotive nature of the subject of language and language differences, the disenfranchised minorities always resist with every chance that comes. However, due to the volatility of the subject of ethnicity which is often criminalised and is labelled as tribalism and considered retrogressive and divisive to an apparently united nation (Mhlanga, 2013), members of the disenfranchised ethnolinguistic groups find it difficult to overtly speak about it, or publicly voice their marginalisation. As shown in this section, they use other counter spheres to challenge linguistic and cultural representations in the public domain. Against this background, this study demonstrates how Facebook's group function has provided the disenfranchised with space for regrouping and resisting marginalisation and language shift, thereby promoting language and culture revitalisation. It shows how the ethnolinguistic online communities build on the traditional forms of minority resistance, of course with limitations.

Conceptual underpinnings: Hegemony, hidden transcripts and alternative media
The study is grounded in the concepts of hegemony, hidden transcripts and alternative media. Since the study discusses linguistic marginalisation and resistance of ethnolinguistic minorities against dominance by major languages in Zimbabwe, it is imperative to set a conceptual grounding of hegemony and resistance. The framework depicts how dominant languages assume dominance and thrive in a polity and also provides insights on how the disenfranchised linguistic minorities resist domination. According to Suarez (2002), the Gramscian concept of hegemony relates to the power relations between dominant and minority groups, particularly the means by which the dominant group secures its position. In particular, linguistic hegemony is identified as what is achieved when dominant groups create a consensus by convincing others to accept their language norms and usage as standard or paradigmatic. A situation of linguistic hegemony is demonstrated by language use patterns in the public domain where some languages are dominant and others are side-lined. As a result, the marginalised languages assume inferior positions and the speakers perceptibly feel denigrated and disenfranchised in situations where they are required to access public information using the dominant languages.
However, this study recognises the agency of ethnolinguistic minorities to resist domination. Like all acts, resistance is situated in certain time, space and relations and engages with different types of actors, techniques and discourses (Vinthagen & Johansson, 2013). Since the minorities' capabilities and chances of contesting dominance and exercising resistance in public spaces are constricted; this study considers the existence of ethnolinguistic minorities' online communities as hidden transcripts and subversions to the dominant ethnic groups. Therefore, the ethnolinguistic online communities are part and parcel of what Scott (1985) calls "everyday resistance", a theoretical concept which covers a different form of resistance that is not as dramatic and visible as rebellions, riots, demonstrations and other forms of confrontational articulations of resistance, but rather which is discrete, disguised or otherwise seemingly invisible. These are tactics that are exploited by people in order to both survive and undermine repressive domination, especially in contexts when rebellion is too risky (ibid).
Thus, since language and ethnic politics in Zimbabwe are volatile issues (Mhlanga, 2013), this study considers ethnolinguistic online communities as part of everyday forms of resistance to the dominant groups. This framework enhances the understanding of the entrenched linguistic and cultural disenfranchisement in post-independence Zimbabwe and methods that are employed by linguistic minorities to resist domination. In this case, members belonging to minority and disenfranchised communities resort to what Fuchs (2010) calls alternative or critical media. Alternative media interrogates domination and articulates the standpoints of the oppressed and dominated groups and people, in the process arguing for the existence of an egalitarian society. It is the new digital media technologies that offer alternative communicative platforms which function as online public spheres (Pappacharissi, 2002). The online spheres now include the previously excluded participants in the offline sphere. Digital media enable voices of previously excluded from dominant discourses to form what Dahlberg (2011) calls counter-publics which are littered with counter-discourses from the marginalised and oppressed groups. Therefore, ethnolinguistic communities on Facebook function as typical subaltern counter-publics and spaces of withdrawal and reorganisation against the dominant. They emerge in response to exclusions of ethnolinguistic minorities' identities. To use Scott's (1990) term, online ethnolinguistic groups contain "hidden transcripts" which are rejoinders to the "public transcript" which are constructions of dominant linguistic and cultural representations. In this context, Facebook groups are the "offstage" which consists of offstage posts, discussions or opinions which challenge dominant ethnolinguistic identities in the offline public domain. Facebook as a social networking site presents opportunities for free participation by the marginalised to speak and radically resist dominant ethnolinguistic identities and, in the process, resist language shift and promote their languages.

Methodology
In this qualitative inquiry, a netnographic study of online communities of Zimbabwean ethnolinguistic groups on Facebook was conducted. Contrary to traditional communities whose membership is defined by geography, online communities use networked technology to establish association across geographical barriers and time zones (Johnson, 2001). In other words, the community is located on the Internet or the World Wide Web. This study was focused on the online communities on Facebook which comprise people bound together by common language and culture. Netnography is a research design which refers to the online adaptation of ethnographyan innovative technique of conducting systematic ethnographic investigations on online communications networks which combines archival work, participation and observation (Kozinets, 2015). It is a combination of research methods that studies communities and cultures formed through computer-mediated social interaction.
For the purposes of this study, we studied 10 out of 23 identified ethnolinguistic groups on Facebook which were drawn from the Ndebele (3), Tonga (2), Hwesa (1), Ndau (1), Sotho (2) and Kalanga (1). We observed and analysed the content on the Facebook pages, that is, the posts and the subsequent comments for six successive months.
Though netnography utilises computer-based methods of data collection, participant observation still remains a critical element of the ethnographic process (Miller & Slater, 2001). For that reason, we made use of acquaintances from the selected ethnolinguistic groups to post and lead discussions on our research questions. Thereafter, they translated the meanings of the posts and comments from the indigenous languages in question to English. This allowed us to "connect" with members of geographically dispersed communities via the online platforms. Content analysis was done using the readily available and obtainable data from the 10 Facebook ethnolinguistic group pages representing the selected six languages spoken in Zimbabwe. In this study, we selected netnography because it is useful in revealing interaction styles, personal and group narratives, networked communities and is an archive of life and window into hidden worlds (Kozinets, 2015). Therefore, using thematic analysis, this study critically analyses the manifestations of linguistic disenfranchisement, minority resistance and language revitalisation in ethnolinguistic groups on Facebook.

Findings
This study explored the use of Facebook by the disenfranchised ethnolinguistic minorities to resist disenfranchisement and promote the use of their languages. This section presents the key findings of this study which were exposed by content analysis of posts and comments on Facebook group pages of the Ndebele, Tonga, Hwesa, Ndau, Sotho and Kalanga.
The first finding of this study shows that ethnolinguistic minorities engage in self-naming and self-definition discourses, in the process reclaiming their presumed proto-identities. The post and comments show high consciousness by members of online ethnolinguistic minority groups of their own proto-identities. The tags of the groups show collective pride of members in naming and defining the self. The groups include BaTonga KwaBinga (the Tonga of Binga), BaTonga People Worldwide, BaTonga baKasambezi (the Tonga of Zambezi), BaSotho BaJahunda (the Sotho of Jahunda), Bana bebu Kalanga (the Kalanga descendants), KwaBulawayo (bring back Bulawayo), The Ndebele State, Mthwakazi republic, Ndebele Qho (the proto-Ndebele), Ndau legacy Association, Rekete ChiNdau: Leave a Legacy (speak Ndau and leave a legacy) and Hwesa Nepakati PaMhokore (Hwesa via Mhokore mountain).
This study observed that, while the majority of language groups are present on Facebook, no Shona or Zezuru groups were observed. This is understandable, since the discourses in most ethnolinguistic Facebook groups challenge the hegemony of Shona language and Zezuru dialect depending with context. While existing literature show that Ndebele dominates other languages spoken in the Matebeleland region, discourses on Ndebele particularistic Facebook groups contest the prominence of Shona identity in public domains. These groups include Ndebele Qho and Not Everyone is Ndebele or Shona. The posts and comments on these pages display detest and resistance against the hegemony of Shona. This study also shows that there are discourses that are characterised by banter and invectives in some ethnolinguistic online communities which are directed to the dominant groups. For example, posts and comments on the pages exhibit detestation of the Shona. For example, in the group Not everyone is Ndebele or Shona, the page has insolent posts and comments that expose and magnify deep-seated resistance of the Ndebele against Shona hegemony. For instance, Ubutshapa akusobami, ngingayenza ubutshapa amaShona wona abesesenzani) (I am not dirty, if I am to develop that habit what will the Shona do?).
Sicela lingasibhaleli ngesiShona egroupin, asiso maShona singaba koMthwakazi thina (Please no posts or comments in Shona language, we are not Shonas, we are proud descendants of Mthwakazi) The study also show that while the existence of ethnolinguistic online communities exhibit the politics of language and belonging in Zimbabwe; the discourses in these groups also show that at times the politics of language degenerate into debates on politics of the nation of Zimbabwe. For example, the Facebook group uMthwakazi in particular displays discourses of devolution which are also observable in post-independence national political discourse. For instance, Kube sisangaphansi kombuso wamaNdebele koMthwakazi ngabe asidubeki ngezomnotho (if only I was in Mthwakazi republic, I would not be in this economic challenges) On the other hand, the dominant position of the Shona identity in Zimbabwe's public domains such as education, media and political administration has disenchanted the Ndau and Hwesa. This is demonstrated by the posts and comments in the Facebook groups Rekete ChiNdau: Leave a Legacy and Hwesa Nepakati PaMhokore.
The comments above show that these ethnolinguistic online communities challenge Shona as an invention of the standardisation process which elevated the Zezuru dialect at the expense of other mutually intelligible varieties of the language. The Korekore, Karanga, Manyika dialects of Shona and Ndau language (originally a dialect of Shona) have been sub-merged in the so-called unified Shona identity for a long time. In actual fact, in studying Shona, the speakers have been forced to write the way they do not speak. Therefore, these ethnolinguistic online communities challenge the homogeneity of standard Shona.
While the foregoing has shown that the prominent discourses in Ndebele particularistic Facebook groups revolve around objection of Shona hegemony in Zimbabwe, this study also detected that there are other ethnolinguistic online communities which protest against Ndebele language and culture. These ethnolinguistic online communities include the Sotho, Kalanga, Venda and Tonga. Posts and comments demonstrate resentment on Ndebele hegemony due to disenfranchisement of their languages in the public domain. The groups include BaSotho BaJahunda, BaSotho baGwanda, Bana bebu Kalanga and BaTonga baKasambezi. In the studied ethnolinguistic online communities, it is forbidden to use either Shona or Ndebele which are considered the delinquent languages. Such comments include the following: Sehlopha saFB se ke sa baSotho fela. Arebatle se tebele (This group is strictly Sotho, no Ndebele please) Atitohaka nhu unolebeleka tjiTebele mugubungano, iswi timaKalanga ntja (We don't want anyone speaking Ndebele in this group, we are Kalanga to the core).
The post and comments on the webpages of the studied groups show that they are preoccupied with defending their languages and cultures, vilifying the Ndebele cultural identity and revising their own misconstrued identities.
Another finding of this study is that there is a connection between Facebook groups and the undertakings of language associations of studied linguistic minorities in Zimbabwe. For instance, the members of the Kalanga and Ndau language associations are networked by Facebook groups, in the process forming vibrant online communities. For instance, an existing language association, such as the Kalanga Language and Cultural Development Association, utilises a Facebook group with the same nomenclature to support activities that bolster the development of the Kalanga language. On the other hand, Rekete Ndau: Leave a legacy is a vibrant Facebook platform whose activities are synchronised with the Ndau Language Association.
This study also recognised that posts and comments on linguistic minorities' Facebook group pages demonstrate members' commitment towards the revitalisation and maintenance of previously marginalised languages. Members feel that their languages are at nascent stages of corpus and status development. In this regard, the Ndau, Tonga, Venda and Kalanga online communities demonstrate real action and commitment towards developing the respective languages. For instance, in the Facebook group BaTonga KwaBinga, members of the group always share information and literature on Tonga people and culture that cover an array of issues including chieftainship, marriage, family issues, new books and history, among others. There is real enthusiasm towards increasing knowledge and visibility of the marginalised languages in the virtual space. In Rekete Ndau: Leave a legacy, authors and other members promote reading material written in the Ndau language. In these arenas, the marginalised relive and (re)construct their ethnicities, histories and aspirations.
Though the previous finding showed linguistic minorities' commitment towards language revitalisation, this study noted that no matter how much members advocate the use of their languages, some posts and comments exhibit codeswitching and mixing which involve Shona, Ndebele or English depending with the Facebook group. For example, in the groups BaTonga People Worldwide, BaSotho BaJahunda (the Sotho of Jahunda) and Bana bebu Kalanga (The Kalanga progenies), posts and comments show code-switching and mixing from their languages to either Ndebele or English. In some extreme cases, some comments are totally in English or Ndebele. For example, "ayisikho la besikhuluma ngesticking together wena sungena nge angle" (you were talking about unity, but now you have changed goal posts). On the other hand, posts and comments on the pages of Rekete ChiNdau: Leave a Legacy show traces of codeswitching and codemixing from Ndau to either Shona or English. Similarly, in some extreme cases, comments are entirely in Shona or English.

Discussion
The presence of ethnolinguistic online communities on Facebook belonging to the Ndebele, Tonga, Hwesa, Ndau, Sotho and Kalanga invalidates the suggestion that ethnic communities exist in specific geographical regions. Social media allow members of ethnolinguistic communities to interconnect with others displaced in time and space. These online communities endeavour to reinvent "offline" autochthonous ethnic communities which are bound by a specific language. This implies that in a country with a complex linguistic ecology characterised by linguistic disenfranchisement, members of ethnolinguistic minorities still attach importance to their languages in the interest of culture and identity maintenance.
While use of major indigenous languages in Zimbabwe's public domain appear to be pragmatic, members of ethnolinguistic minorities are disillusioned by the disenfranchisement of their languages. However, although public channels of communication are limiting, the disillusioned ethnolinguistic communities resist marginalisation and subordination of their languages using alternative and cloistered spaces. Social media allow people who share the same language to interface, cooperate and vent out opinions on the marginal position of their languages in the public domains.
Against the background that the "offline" public domain projects the dominant linguistic and cultural representations, this study shows that Facebook is an alternative medium and a counter-public sphere, where people's history, languages and cultural identities are negotiated, produced and disseminated. The overarching argument of this study is that ethnolinguistic online communities exist as alternative or counter-public spheres for challenging dominant identities which are often deliberately or by default imposed on the minorities through language use patterns in the public domain. This study shows that online communities are simply collective forms of self-identification and resistance against domination and exclusion in the offline life. Facebook is a liberating space that allows the formation of communities -the convergence and interaction of the disenfranchised using their native languages without fear of restriction or punishment because social media are characterised by "the death of the gatekeeper" (Dominick, 2005, p. 19).
The existence of Ndebele ethnolinguistic online communities that principally challenge Shona hegemony and envision the restoration of a separate Ndebele state concurs with arguments in existing studies on Ndebele particularism in Zimbabwe (see Ndlovu-Gatsheni, 2008).
However, there are also other ethnolinguistic online communities that challenge Ndebele hegemony in the Matebeleland regions of Zimbabwe. This finding augments existing studies on the complexity of the Ndebele identity since it involves sub-ethnic groups that originally did not call themselves Ndebele (see Mazarire, 2009). Because Ndebele has functioned as the de facto lingua franca of the Matabeleland region, the disenfranchised ethnolinguistic minorities are perturbed and they remain conscious of their proto-identities. The flip-side shows Ndebele ethnolinguistic online communities with members resist "contamination" by the so-called non-Ndebele who were assimilated and "Ndebelelised". Such Ndebele particularistic groups on Facebook such as "Ndebele Qho" (proto Ndebele) and 'Mthwakazi kandaba abesintu are sustained by pride and self-importance of "royalised" Ndebele identity that exclude others as not genuine Ndebeles. This demonstrates fascination about going back to the "Nguni roots" in South Africa, in the process deconstructing the Ndebele identity and (re)constructing of an imaginary "original" identity which is somewhat utopian. Likewise, ethnolinguistic online communities which challenge the existence of a unitary Shona identity endeavour to deconstruct the "Shona" identity which they consider to be a creation that benefits the Zezuru at the expense of other subsumed language varieties. This unearthing augments prominent arguments in critical African linguistics literature which regards the so-called indigenous African languages as colonial inventions (see Zeleza, 2006). Ndebele and Shona are not self-evident concepts, since the language groups are heterogeneous and therefore contested.
Language use on Facebook pages of the studied ethnolinguistic online communities shows members' commitment to the revival, maintenance and legitimation of their languages and cultures. The study shows that the online communities provide alternative means of a counter-public sphere for the disenfranchised to reunite and speak against their domination in their own languages and in the process promote the revitalisation of their languages. Marginalised languages that include Venda, Ndau, Sotho and Tonga, among others, are at nascent stages of development; therefore, ethnolinguistic online communities augment the activities of revitalising, documenting and increasing the visibility of these languages.
In this regard, the study shows that language associations of linguistic minorities have embraced new communication and networking technologies and applications in their activities. Facebook ethnolinguistic online communities have emerged as active platforms for organised language and culture activism. This is what Ricento (2006) refers to as the new-found vigour to use a language as an attempt to stop language shift or death. This confirms the arguments that the most successful revitalisation efforts are not by the government but by the linguistic minorities themselves (ibid). The groups fervently advocate the recognition and use of the respective minority languages. In this particular case, social media has been appropriated into undertakings of language activism. However, though there is commitment to promote linguistic minorities on the online sphere, that is derailed by the colonial and post-independence language policy and practices that shaped language use and practices in the "offline communities". Colonial and post-independence language policy and practices in Zimbabwe promoted English, Ndebele and Shona (Makoni et al., 2006). As a result, Shona, English or Ndebele have become part of the complex sociolinguistic tapestry of the linguistic minorities. This explains code switching and code mixing in ethnolinguistic online communities Facebook. This study shows that online communities provide an alternative means of a counter-public sphere for the disenfranchised to reunite and speak against domination in their own languages and in the process promote the revitalisation of their languages.
However, this study show that instead of concentrating on resisting language shift and promoting language revitalisation, discourses on ethnolinguistic online communities degenerate into ethnopolitics and politics of the Zimbabwean state and nation. While the studied ethnolinguistic Facebook communities replicate Zimbabwean language politics enunciated in existing literature (see Mpofu & Mutasa, 2014;Ndhlovu, 2009), they also replicate political contestations in postindependence Zimbabwe (see Ndlovu-Gatsheni, 2013). To use Scott's (1990) terms, though the "offline" public transcript in the public domain projects a united nation with one vision, the hidden transcript in ethnolinguistic online communities exhibit national dislocation and disillusionment of the marginalised linguistic minorities.

Conclusion
This study is a significant contribution to the growing body of scholarship on the politics of language, ethnicity and identity in Zimbabwe in particular and Africa in general. Though there is substantial existing literature on the politics of language, particularly the marginalisation and the exclusion of ethnolinguistic minorities in the public domain, no scholarship exists that addresses the appropriation of social media by ethnolinguistic minorities in resisting marginalisation. This study therefore informs and extends debates on the everyday forms of resistance by the disenfranchised ethnolinguistic minorities, especially in the era of digital and social media. This study demonstrates that the various groups on Facebook which are formed on the basis of individuals sharing the same language and cultural identity are new platforms in which resistance to hegemony is exercised. We conclude that to some extent ethnolinguistic online communities provide an alternative social space for the disenfranchised minorities to exercise their agency by re-asserting their identities and resisting marginalisation, in the process contributing to the promotion and revitalisation of endangered languages and cultures. However, in other instances, these platforms have degenerated into platforms of political activism, regionalism and prejudiced exclusion.