Navigation – Plan du site

AccueilNuméros8(1)A matter of honour: Speaking Hozo...

A matter of honour: Speaking Hozo (Ak’mo Wandi), an endangered language in western Ethiopia

Sophie Küspert-Rakotondrainy et Klaus-Christian Küspert

Résumés

Cet article concerne le hozo, une langue minoritaire parlée dans l’ouest de l’Éthiopie, et ses locuteurs. Aujourd’hui, la pratique du hozo, qui appartient à la branche mao des langues omotiques, décline rapidement. Nous cherchons à étudier les discours et les circonstances pratiques qui sous-tendent le changement et la disparition de la langue, et comment la communauté linguistique donne un sens à ses options et à ses choix. Nos méthodes de recherche étaient principalement ethnographiques, et les données sont issues à la fois d’un long travail de terrain avec des entretiens approfondis et d’observation participante, et de dix ans d’expérience pratique dans divers projets de développement linguistique dans la région, avec des locuteurs de langue maternelle hozo et des locuteurs d’autres langues mao. Les résultats montrent que bien que les locuteurs fondent leurs choix concernant l’utilisation de la langue sur des considérations pratiques, la langue elle-même a une valeur symbolique importante qui confère à la communauté un sentiment de kap’á (honneur) et témoigne d’une « pureté » ancestrale. Par conséquent, le développement de la langue n’augmente pas nécessairement les fonctions pratiques du hozo, ou n’arrête pas le processus de disparition de la langue, mais il peut encore avoir une incidence au-delà de la génération des derniers locuteurs natifs.

Haut de page

Texte intégral

1. Introduction

  • 1 The authors have known Isak, which is a pseudonym, since 2012. For several years, we worked with h (...)
  • 2 Ethnologue: ISO 639-3 code hoz, glottocode: hozo1236.

130-year-old Isak1 was born in the Kondala District (wäräda) in the Western Wollega Zone of the Oromia Region. In this area, several languages are spoken by people generically known as “Mao”, a small minority of the population. The social category Mao encompasses speakers of entirely unrelated languages such as the Omotic Hozo, Seze, and Bambasi Mao, and the Koman Gwama language. Other users of minority languages in the area such as Ganza, Komo, and some Gwama speakers do not identify as Mao. The majority language and official language of the region is the Cushitic language Oromo (afaan Oromoo). One of the above-mentioned minority languages is Isak’s mother tongue, usually called “Hozo” in the academic literature but with different local designations, such as Ak’mo Wandi meaning ‘mouth of the Mao’, which is spoken by just a few thousand people.2 Isak regards this language as his mother tongue when, in reality, he does not speak it as well as he speaks Oromo. As a teenager, he moved to the Benishangul Gumuz Regional State and later married an Oromo-speaking woman from this region. His children do not speak any Hozo, and with them hundreds of other children from Hozo-speaking parents. Yet, when asked, Isak claims their mother tongue (usually termed diya wandi lit.: ‘mouth of the father’) is Hozo because they are members of Gebsi, a Hozo-speaking clan.

  • 3 The term kap’á is used to express honour, reverence, or respect to a person or deity (e.g. Naa naa (...)

2Our findings show that many individuals from families who previously spoke Hozo are now using other languages to communicate and that the process of language shift to Oromo may be irreversible. Yet we also argue that the concept of mother tongue is a decisive factor for the self-identification of individuals who regard themselves as Mao even if the language does not necessarily play a practical role in their daily lives. Language development therefore takes on a deeper significance since it is not only about making a language usable as a means of communication or for the sake of the academic venture of documentation and description but because it is connected to local notions of honour (kap’á): Young Hozo speakers who advocate for their language regard knowledge of their ancestral language or descent from a family that speaks the language as a proof of being Ak’mo gará (‘pure Mao’), a state that makes an individual kap’adi ‘honoured’ or wak’ti ‘feared’.3

Figure 1 — Map showing the area where Hozo and other Mao languages are spoken (prepared on the basis of the maps in Küspert 2015).

Figure 1 — Map showing the area where Hozo and other Mao languages are spoken (prepared on the basis of the maps in Küspert 2015).

Language names are shown in white, administrative units in all upper case, and district capitals in the research area in bold red.

1.1 Methodology and research questions

3In this article, we present a glimpse of the fruits of a decade of linguistic and ethnographic research in the Begi-Tongo area which we have carried out — the primary author in anthropology and the secondary author in linguistics. It started by mapping the different minority languages in the area, their spread and the approximate number of speakers, published by the secondary author (Küspert 2015). In 2017, the primary author went on a one-month fieldtrip to Kondala, where she conducted 38 in-depth interviews with Hozo speakers in 16 different settlements. Additionally, in four settlements, she carried out a standard questionnaire with key informants which was based on the nine factors of the UNESCO framework for Language Vitality and Endangerment (UNESCO 2003). Although this framework is not applied to guide the analysis of the findings in the present article, the questionnaire gives us an indication of the number of speakers and proportion of people who do not use the language anymore, intergenerational transmission, and the use of the language in various domains.

  • 4 Since the security situation in western Ethiopia had drastically deteriorated by the time of the s (...)

4The primary author carried out a new period of ethnographic research during 20 weeks of intensive fieldwork, spread out over a period of 12 months from July 2020 to June 2021. In this time, she conducted 189 interviews in total; 74 with individuals from the Kondala and Babo Gambel Districts, of which 29 were Hozo mother tongue speakers. She lived with Mao families who used different Mao languages in the Oromia and Benishangul Gumuz Regions throughout the research period.4 Parallel to this, the secondary author has continuously been engaged in language development activities together with Hozo speakers in numerous workshops both in the area and in Addis Ababa, as well as, starting from January 2021, frequent sessions with mother tongue speakers for the development of a comprehensive dictionary of Hozo.

5The findings presented in this paper are largely based on evidence obtained through the ethnographic experiences in our interactions with the language community. This is because we have come to realise the importance of the context in which language change, language loss, and language development take place. With this paper, we aim at laying the ground for more specific socio-linguistic research in this area. We want to provide a deeper understanding of how local people and members of the minority language communities make sense of the changes that are happening. Instead of asking why the Hozo language is endangered and how we can “save” it, our research question has been how the processes of language shift takes place and which practical functions the different languages have for the portrayal of identities in the communities. Hence, we have approached our own engagement in language development critically, as we have realised how multi-layered and complex the notion of “Maoness” and the reasons for marginalisation and language endangerment are.

1.2 Theoretical discussion

6Many languages in the world today are endangered and may not survive the next generation, and it is minority languages in particular that are more vulnerable to extinction (Patrick 2008; UNESCO 2017; Eberhard et al. 2022). This has led to an academic and practical linguistic field that promotes preserving language ecology and promoting linguistic human rights (Mühlhäusler 1996; Skutnabb-Kangas & Phillipson 2010). The challenge with this position is that it often assumes that speakers operate as a unified group, that their languages overlap with their ethnic identities, and that individuals naturally want to continue using the minority language (May 2012). Indeed, linguistics and anthropology have in recent years been criticised for using the field of language endangerment to justify their work and “legitimize their disciplines” (Heller & Duchêne 2008: 4). Mufwene asks whom language revitalisation is done for and why, and whether it makes sense for the people who are supposed to benefit from it. He writes: “Should Native Americans be forced to remain on their reservations in order to maintain their languages, even if this would prevent their cultural assimilation and sustain their marginalization from the dominant, nonindigenous socioeconomic world order?” (Mufwene 2017: 208). He argues that it may be more beneficial to view language use and shift as a process where the speakers make conscious choices that they believe will improve their lives rather than something that happens to them.

7Instead of focussing on the endangerment of a language, Heller and Duchêne (2008) recommend focusing on its speakers, and what their stakes are in the choice between different languages. Furthermore, a language spoken by a minority does not have to be less prestigious, and language shift does not necessarily imply the death of a language (Léglise & Alby 2006; Patrick 2010). Instead of considering language endangerment as a static concept, it may therefore be more fruitful to discuss the process of “language minorisation”, meaning which conditions contribute to prestige, spread, and use of certain languages over others in specific situations (Léglise & Alby 2006; Patrick 2010). Languages and their use are never neutral in the context of a “linguistic hierarchy” (Patrick 2010: 176), especially in a situation where speakers of some languages constitute a social, political and economic “periphery” (Cornips & Rooij 2018: 5). The advantage of approaching this research from a perspective of minorisation is therefore that it puts the process of language use and language change in focus. We find that this yields a better understanding of individuals’ agencies within a situation of language change.

2. Brief background of the area

2.1 Historical and political backdrop

8Historical and archaeological research have found evidence that people whose material art, cultures, and languages were related to that of today’s Mao lived in this area long before the Macha Oromo immigration into western Ethiopia, which started in the 17th century (Fleming 1984: 32; Negaso 2001; González-Ruibal 2014: 298; Meckelburg 2018). Many of the previous inhabitants of the area were assimilated by the Macha through adoption and marriage. However, those who were “too different” from the Macha in prestige were not assimilated (Tesema 2006: 45), and many “were forced to move to [the] inhospitable and unhealthy [lowland/rural] zone after the occupation of the highland areas by the Oromo” (Negaso 2001: 31). The Macha started taking slaves from the peripheral people, who were often labelled “Mao”, and those who were not enslaved remained part of the lowest stratum of the society (Bartels 1983; Hultin 1987). By the 19th century, most of today’s Western Wollega was majorly Oromo-speaking, and those identified as Mao inhabited the “fringes” of the Oromo settlements (James 1980: 62).

  • 5 The discussion on autochthony is not peculiar to Ethiopia. Wherever there is an “indigenous” group (...)
  • 6 Oromo nationalism cannot be discussed in detail here. For more discussion, see Baxter (1994), Mere (...)

9Today, in the whole of the Oromia Region, the Oromo are regarded as indigenous and the legitimacy of their firstness and their authority over the land constitute a central part of the local political discourse. Similar discourses can, for example, be observed in the Omo valley (Bassi 2014) and at the Oromo-Somali border (Fekadu 2011).5 Official publications by the Oromia government do acknowledge the existence of an original population in the area, but these are said to have been so “unorganised” that the area should be considered as having been a “no-man’s-land” when the Oromo immigrated (Alemayehu et al. 2004: 217). This political discourse, coupled with a strong Oromo nationalist movement has led to individuals, families, clans and languages identified as Mao or hereditary close to Mao being largely excluded from political power, economic development and social opportunities.6

10In the Oromia Regional State, there is no political or institutional acknowledgement of the existence of any social and ethnic identity other than Oromo. The constitution only caters for ummanni Oromoo (‘Oromo people’), and the only official language is Oromo (Revised Constitution of the Oromia Region 2001: §39:1; Van der Beken 2015). Mao languages that are only found in the Oromia Region, therefore, are excluded from any use in the public sphere. The constitution acknowledges the existence of non-Oromo in the region, but only in terms of “peoples who made a choice to live in the Region”, and not in terms of people who have resided in this area for centuries (Revised Constitution of the Oromia Region 2001: §2.1). Hence the Mao in Western Wollega live as hidden minorities without the cultural rights enshrined in the Federal Constitution of Ethiopia (Constitution of the Federal Democratic Republic of Ethiopia 1995; Meckelburg & Küspert-Rakotondrainy 2019). The situation in the Oromia Region stands in contrast to the neighbouring Benishangul Gumuz Regional State, where two Mao languages, Gwama and Bambasi Mao (Màwés Aas’è), are officially recognised and have been introduced in school (SIL 2011; Revised Constitution of the Benishangul Gumuz Regional State 2002; Küspert-Rakotondrainy 2016).

11Despite their inferior social, economic, and political status, individuals who self-identify as Mao express a sense of pride attached to their Mao descent since it sets them apart from individuals who are said to be descendants of slaves. The main difference between so-called “pure Mao” and purported slave descendants is that the former come from families that speak a Mao language whereas the slave descendants are monolingual in Oromo, allegedly because their ancestors grew up in Oromo-speaking homes as domestic workers (Küspert-Rakotondrainy forthcoming). The term gará (‘pure’) in Hozo may also be translated with ‘clean’, as in gará haanni ‘clean water’. The term gará is often used with a negation to describe a patrilinear genealogy which is believed to include ancestors who are allegedly of non-Mao descent or who were slaves (Zaga moo gará bishu ‘This man is not pure’). The discussion below on the symbolic value of the Hozo language is related to this honour of coming from a Mao-speaking family, irrespective of whether an individual is fluent in this language or not.

2.2 The Hozo language and its speakers

  • 7 We are not aware of any meaning of ak’mo other than ‘Mao people’; more research into the origin of (...)

12The term “Hozo” or “Hozi” is usually understood by Ak’mo Wandi-speaking people to refer to a specific clan and dialect, and is not generally accepted as a language designation by all the speakers although it may be one of the most common. Other names for the same language are labels referring to specific groups, such as Maramo. Alternatively, the term Shuluyo is used, which is probably a label given to speakers of Hozo by Seze speakers (Küspert 2015). The Hozo word for “Mao” is Ak’mo and may be a designation accepted by all speakers of the language. It is a general term for all minority groups that are subsumed under the label “Mao” and not only Hozo speakers. Since the term Ak’mo Wandi is used exclusively in this specific language, we here use it as a synonym for Hozo.7 The reason why we use the term Hozo in this article is that it is the most widely used in academic writing and research (Bender 1975; Fleming 1976; Siebert et al. 2002; Getachew 2015; Ahland 2012; UNESCO 2017; Eberhard et al. 2022). However, the language informants in the ongoing dictionary building (by the secondary author through SIL Ethiopia) insist on using the term Ak’mo Wandi to make it acceptable for all speakers of this language.

13Against the backdrop of the above-mentioned social and political situation, we can note the absence of a dedicated term for the Hozo language which would set it apart from other Mao languages. It demonstrates a limited politically conditioned group formation based on linguistic criteria (Jenkins 2008: 10). To themselves as well as to others, these people are simply Mao, no matter which language they speak. Where linguists have found fundamental language differences (cf. Bender 1975 and 1976; Fleming 1976), the local community largely operates without acknowledging differences between various Mao languages. Furthermore, even users of the Oromo language may be classified as Mao if they are believed to have hereditary relations to ancestors similar to those of the Mao, or have a darker skin complexion — a characteristic associated with Mao people (González-Ruibal 2014; Meckelburg & Küspert-Rakotondrainy 2019; Küspert-Rakotondrainy forthcoming). Since all people in this area practice exogamy, marriages happen indiscriminately between any Mao clan and language (Meckelburg & Küspert-Rakotondrainy 2019). To a certain degree, this also includes Gwama-speakers in the Benishangul Gumuz Regional State, where “Mao” today is a political term (Küspert-Rakotondrainy 2021). Who is and is not included in the Mao category, therefore, bears witness of a plastic use of ethnic and linguistic labels and identities which is closely linked to social hierarchies; in most situations, Oromo is understood to be more prestigious than Mao, and less prestigious individuals tend to be labelled “Mao” — the distinction between alleged slave descendants and Mao here being blurred.

  • 8 In this article, we mainly consider the situation of Hozo with a sideways glance at Seze, both of (...)
  • 9 There is new evidence of villages with a proportion of Hozo speakers south, east and north-east of (...)

14Another language spoken by the Mao people in the Begi and Kondala Districts is Seze, closely related to Hozo and also belonging to the Omotic language family. In the Begi District of the Western Wollega Zone of Oromia and across the border into the Mao Komo Special District (liyu wäräda) of the Benishangul Gumuz Region, a considerable number of the speakers of the Koman language Gwama also self-identify as Mao (Küspert 2015; Küspert-Rakotondrainy 2021).8 The number of speakers is estimated to be 6,000 for Hozo and 13,000 for Seze (Küspert 2015). Even if this figure is conservative,9 the proportion of Hozo speakers compared with the whole population in these two districts does not exceed 5%. For Seze, it may be between 8 and 10%.

15The most recent census carried out in Ethiopia reports that the total population in the Kondala and Babo Gambel Districts is 157,227 (Central Statistical Agency 2007). The census reports 24,272 Mao in the Oromia Region as a whole but it is unknown which criteria were used to count Mao in the census (cf. Meckelburg & Küspert-Rakotondrainy 2019). For example, individuals who by the local Mao community are not categorised as “pure Mao” (i.e. descendants of slaves), may be identified by outsiders as Mao due to their appearance and purported ancestral origin (cf. Küspert-Rakotondrainy forthcoming). Furthermore, the census does not make a distinction between members of the Mao community who do not speak the language and fluent speakers. As we will discuss, many individuals who are said to speak Hozo may not be able to express themselves fluently or have a limited vocabulary in several domains. The proportion of Hozo speakers in the whole Ethiopian population (2021 estimate) is hardly bigger than 0.005%.

16Ethnologue classifies the language vitality status of Hozo as being at level 7 (“shifting”), and the same is true for Seze (Eberhard et al. 2022). Gwama appears at level 6b (“threatened”) and Komo at level 6a (“vigorous”) although the vitality may differ substantially in the different regions. Ganza (spoken in the lowland towards Sudan) is also assigned to level 6a, which seems optimistic for a language with probably less than 1000 speakers in Ethiopia. Bambasi Mao (Màwés Aas’è) is classified as level 5 (“developing”) because of recent language development measures. However, this language is only spoken by a small minority in a few villages around Bambasi town (Ahland 2012).

3. Language shift and advocacy for Hozo

3.1 Using and ceasing to use Hozo

  • 10 This was done in group discussions with elders in two different sub-settlements of Gemi Gaba by es (...)

17Almost all Mao people with whom we have spoken have been able to fluently express themselves in Oromo. Only a very few, elderly people have preferred to speak in their respective mother tongues, Seze or Hozo. When looking at who uses the Hozo language more consistently, it is apparent that the older generation does not only speak the language better and more fluently, but also more frequently. A rough estimate based on Isak’s home village in the Gemi Gaba Sub-district (qäbäle), indicates that around 25% of the children of Hozo-speaking parents at school-going age cannot speak Hozo at all.10 It is only in very rural settlements closer to the Dabus River in the east that it is common to meet small children who do not speak any language other than Hozo. However, even in such settings, a significant number of children use Oromo at home and in communication with close family.

  • 11 The work of NGOs on language documentation and the development of a preliminary Hozo orthography w (...)

18According to our findings, there are few individuals self-identifying as ethnic Oromo who speak a Mao language (less than 5%), even in the villages with a significant number of Mao speakers. On the other hand, the percentage of Hozo and Seze speakers who have good or very good Oromo skills exceeds 90%. Hozo speakers who speak Seze, or Seze speakers who speak Hozo, may amount to 15-20% in areas where both languages are spoken. As found by Küspert (2015), there are a number of villages in the western part of the Kondala District, which house Seze and Hozo speakers equally, but there is much less use of Seze further east in the district. In the Babo Gambel District, east of the Dabus River, no Seze speakers are reported. In all villages of the area, Hozo is a minority language with an Oromo-speaking majority. Only a few small and very remote settlements without road access in the Guma Gara Arba Mountains and the Dabus Swamps have a Hozo-speaking majority (see map). Neither Hozo nor Seze have an official orthography and none of the Mao languages are formally used in education or religious institutions in the Oromia Region.11

19Migration, education and marriage have an impact on children’s language use. Children who continue with their schooling beyond what is available in the local village or town, especially after grade 8, may often not develop full mother tongue competence since the language of formal education is exclusively Oromo and few other Mao students continue to secondary school. It is evident that many of the more educated young Mao do not feel comfortable using Hozo in longer or complex discussions. This may be due to a limited vocabulary and the fact that they feel their language skills have remained at the level of a child. Documenting their proficiency would require further linguistic research. Polygamy is widespread, and co-wives often speak different languages (Seze, Hozo, Gwama or Oromo). When they speak to each other, they usually communicate in Oromo. Children raised in these multilingual homes, where the lingua franca is Oromo, often end up with little or no knowledge of any of the Mao languages that their mothers may speak.

20Much of the lexicon that is considered particular to the Hozo language is related to customs like hunting, which is no longer practiced, the building of traditional houses, and farm and kitchen tools that are no longer in use. Furthermore, many concepts in the language are related to the traditional religion, performed through a spirit medium (biash moo) in a spirit house (biash kera). The majority of Hozo speakers are formally Muslim, as are most of the Oromo-speakers in this area (Meckelburg & Küspert-Rakotondrainy 2019). Although practicing the traditional religion alongside Islam is not uncommon, the prevalence and visibility of traditional prayer gatherings and spirit houses have drastically reduced over the last generations. In many villages, the last spirit medium is an old man who will not pass on his function to an heir. Hence, these practices are no longer transmitted to the next generation, and people like Isak have never brought their children to a prayer house. In such a context, not only words, but also the worldviews of a culture disappear (cf. Phillipson & Skutnabb-Kangas 1995).

21When young people talk, they are often concerned with matters related to their daily lives, at work and in their community, in the mosque or church, at school, in sports, or matters related to administrative issues in the village or district. They marry out of love, convention, or economic advantage, not by exchange, like their grandparents did, and they increasingly choose Oromo-speaking partners. This is reflected in their language. Out of habit or because it is easier for them, Oromo words and phrases are used, and whole conversations are held in Oromo. Many younger people can greet and give simple commands in Hozo, but struggle when it comes to the narration of a longer story or the elaboration of a complex argument because their vocabulary is not large enough or is limited to a few domains. They may show uncertainties in the use of grammatical forms and especially in the correct use of tones, which have both semantically and morphologically distinctive functions in the language. Elders often complain that young people are no longer able to appropriately use formal language and polite forms, and instead answer them in Oromo.

22Mao children and adults are often made fun of in school, in their villages, and in government offices. Their language is sometimes ridiculed as a “bird language” (a pejorative expression for a language that “no one can understand”), and even as a language that does not really exist because it is neither written nor officially acknowledged. According to Mufwene (2017: 211), explanations concerning minority language speakers giving up their languages because of shame or a lack of pride can be conceived as “condescending”. Nevertheless, it is a frequently recurring claim made by our interlocutors that many Hozo speakers “feel ashamed of using their language” (expressed by an elderly man from Karma Shoro in the Kondala District). Often, the mother tongue is only spoken in private, and no language other than Oromo is spoken at markets. The main reason why speakers do not use their mother tongue in public seems to be not wanting to attract unnecessary attention and to avoid being taunted, which many Mao people experience anyway because of their different appearance.

3.2 Church, state and the promotion of Hozo

23A small group of mainly young, quite well-educated Hozo speakers are concerned about the intergenerational transmission of their language and want to take action. Some of them are currently collaborating with SIL, a faith-based organisation that works together with local churches, and the Norwegian Mission Society in language development efforts such as orthography development, word collection for dictionary building, oral Bible storytelling and the writing of Christian songs with traditional melodies. All of them are male, and many live outside of the region, mainly in Asosa, the capital of the Benishangul Gumuz Regional State. These young men can be referred to as “cultural brokers” in the understanding of Eriksen (1992: 317): “those individuals mastering both the code of the dominator and that of the dominated, are simultaneously the minority members farthest removed from the traditional culture and those best equipped to serve their interests”. Many of them are, like Isak, not fully proficient in the language they are promoting, and most of them have married outside of their language community. As we will see, it appears that it is religion, rather than political activism, that these young Hozo speakers turn to in their endeavour to counteract the loss of their language.

  • 12 This information is also based on personal observations from the authors’ cooperation with the loc (...)

24Some of these eager, young Hozo speakers are Christians. Protestant Christianity often requires a strict break with traditional forms of worship (Data 2005), and the local churches use no language other than Oromo in evangelisation, preaching and regular church work. Some churches even actively discourage any cultural or linguistic expression other than that of the majority population since all deviations from this norm are considered to be non-Christian (Küspert 2020).12 Although evangelisation efforts among non-Oromo people seem to progress significantly slower than among Oromo speakers, there is today a small community of Mao people in the Ethiopian Evangelical Church Mekane Yesus (EECMY), and among them some Hozo mother tongue speakers. Since Christians often sever ties with their families when they convert, many Muslims believe that Christians have more limited mother tongue skills than followers of the traditional religion and of Islam, although no data about this exists.

25Interestingly, it is through the Mekane Yesus church that some minority language work has been done, thanks to several donor-initiated projects. Paradoxically, then, at the same time as the church is contributing to language endangerment and loss, it has also become a vehicle through which SIL and the Norwegian Mission Society seek to counteract this development. The young Christians themselves often state that they think converting to Christianity is beneficial for language development since Protestant Christianity is strongly associated with education — a prerequisite, in the opinion of most young Mao, for both language development and more social and political visibility. In accordance with Eriksen’s above-mentioned paradox, becoming a Christian and potentially working as part of the church’s language development efforts is believed by some Hozo speakers to be a means of maintaining and reviving their language. Isak briefly converted to Christianity and was quickly recruited by a language development project. He, like many other Mao, converted back to Islam when his family threatened to expel him and when his engagement in the language work came to a standstill. However, he still carries with him the eagerness to do something to stop the decline of his language and often expresses his readiness to work with language activities— through the church or other institutions.

26None of the young men who wish to promote their language are engaged politically, and their advocacy is mainly limited to their own community or towards foreigners or NGOs. The support given for language development work is also a means of income for them. Political advocacy is regarded as dangerous, especially given the extremely fragile security situation in the area. Politics in Ethiopia is to a large extent ethnicised, and the core of any political struggle is framed as the liberation of the nationalities (i.e. ethnic groups) against “oppression” from the majority groups or politically powerful elites (cf. Vaughan 2006; Aalen 2011). What the young Hozo speakers fear is that a politicisation of language development efforts — even the question of script choice which is a prerequisite for getting the informal orthography of Hozo approved by the regional government — will be perceived as a Mao uprising against the Oromia government. Initiatives that would have the potential for activism aimed at setting Mao and the Mao languages on the agenda are therefore stifled before even attempted.

4. Practical functions and symbolic values of Hozo

27The language development work of Hozo is taking place in the context of tension between the attrition of the language and the honour of the language — two seemingly irreconcilable forces. This tension can be understood by examining which functions the different languages have for the community. Just because speakers of Oromo see Hozo as unsuitable for communication and just because Hozo speakers tend to refrain from using their language in public, this does not mean that there are no advantages to knowing Hozo. Although many Hozo speakers may choose Oromo as the most practical and prestigious way of making themselves understood, the symbolic value of Hozo cannot be ignored. Below, we will seek to unpack some of the processes that are triggered by Hozo language development and how these interact with the practical choices and beliefs of our interlocutors.

  • 13 Cf. Patrick (2010: 187) on inequality between minorised languages and the dominant language due to (...)

28It is not uncommon among NGO workers as well as among young Hozo engaged in language development projects to find the opinion that all Mao should speak their mother tongue and that they do not know what is best for them if they choose not to. In principle, people like Isak agree with this idea. In practice, however, other choices correspond more with their practical needs. Isak wanted to move out of his home area to get work, and he chose to marry an Oromo-speaking woman. In order to provide his children with future opportunities, he has made sure that they speak Oromo and Amharic fluently, as these are essential skills for getting a good job. When we asked Isak whether indigenous people should remain in their “territories” for the sake of maintaining their languages (cf. Mufwene 2017: 208), his answer is that he has no wish to return to his father’s village and become a subsistence farmer just because this may prevent language loss in his family.13

  • 14 Furthermore, language development can lead to status conflicts, as the example of southern Ethiopi (...)

29The stigma of being a Mao may be reduced if Hozo gets an officially recognised orthography, although the marginalisation and vulnerability of Hozo speakers will not be removed by simply developing Hozo as a written language (cf. May 2012). A deep sense of inferiority is often ingrained in Hozo speakers vis-à-vis Oromo speakers because of their social and political exclusion and subordination. A political debate on language development which aims at placing the language rights of the “Mao nationality” on the agenda is sensitive in the region, and young Mao say they are afraid of talking about these issues, as addressed above.14 Yet there is a desire in the Mao community to be recognised as something other than Oromo: “Our nationality [i.e. ‘ethnic group’] has its own culture, its own language”, said a young man from Babo Gambel, although he acknowledges that he would not dare to express this publicly. Being proud of a language that exists on paper, that can be written, is something that many Hozo-speakers dream of.

30Establishing written proof of the existence of Mao languages, which these young men demand on behalf of the language community, does not necessarily mean that every Mao needs to speak Hozo or Seze fluently. Rather, their wish is that, if they wanted, the language would still be there for them to speak. A conversation with an elderly man from a village neighbouring Isak’s exemplifies this idea:

Old man: With our language we cannot even buy a cup of salt. So it’s useless.

Researcher: So, you think your language has no use these days?

Old man: No, we can use it. But the Oromo say: “We are those who administer you, so you need to use our language.” They even tell us that we need to call ourselves Oromo. We should not be Mao –we should be Oromo. […] But we don’t accept this.

31This idea of having a language seems to be important for this man’s self-identification as Mao and is a key recurring feature in conversations with Hozo speakers. However, it is a sentiment that does not necessarily actively contribute to the conservation or revival of the language. A young Hozo speaker who lives in Asosa said: “There is no one, no single Mao who totally forgets the Mao language. How can he forget? There are old people who still speak the language.” If for a moment we assume that Mao ethnic identity does not depend on the language (cf. Mufwene 2017: 205), the “Maoness” of the Hozo-speakers could survive without it. Nevertheless, at some point in the near future, there may no longer be a language to learn, and with the language a big portion of the community’s self-identification as Mao would vanish.

32As mentioned above, the discourse on ethnicity and language in the Mao area is largely concerned with claims of firstness and land ownership. In order to claim to be the rightful owner of the land, one has to be able to claim that one is, indeed, a descendant of the “original population”. Although speakers of the Mao languages in Oromia may be ridiculed by Oromo speakers of hereditary Oromo descent, knowing their own language is, within the Mao community, something the speakers take pride in. For many Mao, it is closely related to their claim that they are the “true” owners of the land. This relationship between language and land can also be observed globally, for example in Canada among the Aboriginals’ effort in favour of language revitalisation (Patrick 2008: 52). In such cases, language has more of a symbolic value than posing any practical advantage in terms of communication.

  • 15 The borana social class used to play an important role in Oromo tradition, as only individuals of (...)

33Language and honour are also part of the local discourse on slavery. As mentioned above, it was not uncommon for the slaves of the Macha Oromo to be identified as “Mao” (James 1980; Bartels 1983; Negaso 2001). According to oral history, slaves were incorporated into the clans of their masters and adopted their language and genealogy since they themselves had no knowledge of their biological ancestors (Küspert-Rakotondrainy forthcoming). Today in Western Wollega, this has created a social stratum of alleged slave descendants who have a lifestyle and language similar to hereditary Oromo, but who are believed not to be of “pure” (borana)15 Oromo descent. Through this narrative, a relationship between Mao and slavery is created. Anyone in the community who is known to be non-Oromo or who shows characteristics believed to be “non-Oromo”, such as a darker complexion, could therefore potentially be a slave descendant. As such, many families feel the need to prove that they are not descended from slaves and one of the easiest ways to do this is to speak a Mao language. In terms of local hierarchies, it is regarded as much more honourable to be a “pure” Mao than a “slave”, which arguably is the most stigmatised category in the area and means that an individual is not worthy of receiving honour¸ as expressed in Hozo: Nuu gamt’abella kap’awi ‘we don’t give honour to slaves’.

34The day that Isak sent elders to the family of the girl he intended to marry, one of the first things her father asked the elders was if Isak was a “real” Mao or a slave descendant. When they had made sure that he was, indeed, a Mao, they agreed to the marriage. Isak did not have to prove that he spoke Hozo. However, had the community in his village not known that he comes from a family of Hozo speakers, he would have struggled significantly more to prove his non-slave descendency. We have knowledge of some individuals who are, allegedly, descendants of slaves, and who have later learned a Mao language. In some situations, this skill enables them to pass as Mao and avoid stigma based on slavery. However, the local community, who “knows” of their purported slave descent, regard this as cheating and disapprove the behaviour of “falsely” claiming “pure” descent. Isak is therefore interested in keeping at least a minimum memory of his language alive, and this is typical of the entire Hozo-speaking community. Isak’s children do not have to speak Hozo fluently to keep their honour as “true” Mao. As long as they know their own descent line and some greetings, Isak believes that no one could question their “purity”.

5. Conclusion

35In this article, we have sought to not only document the language use among speakers of the Mao-language Hozo (Ak’mo Wandi) but also the underlying discourses that influence the choices they make. As we have seen, Hozo language use is declining, and the intergenerational transmission of the language is incomplete. Discussions in domains of modern and public life are often held entirely in the dominant language, Oromo. Furthermore, as more young people become educated and move out of their home villages, they increasingly lack full mother tongue competence. Increased prevalence of intermarriage across language borders may lead to families choosing to speak Oromo because this is the only language that all family members speak fluently. Nevertheless, although Hozo is used less frequently, in fewer domains, and by fewer speakers, the language still plays an important role in self-identification as Mao for many individuals from hereditary Hozo-speaking families. Hence, the practical choices they make to learn and speak other languages stand in contrast to the sense of honour attached to having a mother tongue, whether or not they speak it fluently.

36As workers in language development, we need to ask: Does the development of the Hozo language make sense? It is ultimately the language community that decides whether they want to make choices that contribute to its continued use. We cannot, as outsiders, judge whether they are making the right choices. Nevertheless, we argue that the development of Hozo as a written language makes sense for the community even though the language loss may be irreversible. Affirming that Hozo “exists” and is worth writing is among the main aims of mother tongue language development workers, not only vis-à-vis Oromo speakers but also in order to prove that the Mao people are different from slave descendants, although this could mean that language development simultaneously cements the marginalisation of slave descendants. The ancestral “purity” of a family is essential for their honour but very difficult to establish without proof that the family is, formally, Mao-speaking. Having ties to an ancestral language also plays an important role in discourses on autochthony and land rights.

37Development efforts for Hozo are probably not sustainable in terms of ensuring its long-term, active use as a means of communication for all domains. However, they may enable the community to better make their own choices regarding using or ceasing to use the language. Documentation in the form of a dictionary or grammar may also enhance choices for individuals who would like to revitalise their language. The choices that individuals make regarding language use are also largely dependent on the political situation in the area and the lack of opportunities to use their language publicly. Therefore, a (temporary) process of language revival may still be possible, as we can see in the neighbouring region, Benishangul Gumuz, where the minority language Gwama is currently experiencing a slight upsurge. In the end, even if the Hozo language should die, it would die with the honour of having existed.

Acknowledgments

38We would like to thank two anonymous reviewers and the editorial board of the journal for their useful feedback on this paper. We are grateful for the experience we gained through projects supported by the Norwegian Mission Society (NMS) and SIL Ethiopia. We also thank all our Hozo-speaking colleagues and research assistants.

Haut de page

Bibliographie

Aalen, Lovise. 2011. The politics of ethnicity in Ethiopia: Actors, power and mobilization under ethnic federalism. Leiden: Brill.

Ahland, Michael Bryan. 2012. A grammar of Northern Mao (Màwés Aas’è). Eugene, OR: University of Oregon (PhD dissertation).

Alemayehu Haile, Boshi Gonfa, Daniel Deressa, Sanbato Busha & Umar Nure. 2004. History of the Oromo to the sixteenth century. 1st ed. Adama: Oromia Culture and Tourism Bureau.

Bartels, Lambert. 1983. Oromo religion. Myths and rites of the western Oromo of Ethiopia: An attempt to understand. Berlin: Dietrich Reimer.

Bassi, Marco. 2014. Federalism and ethnic minorities in Ethiopia: Ideology, territoriality, human rights, policy. DADA Rivista di Antropologia post-globale 4(1). 45-74.

Baxter, P.T.W. 1994. The creation & constitution of Oromo nationality. In Katsuyoshi Fukui & John Markakis (eds.), Ethnicity & conflict in the Horn of Africa, 167-186. London: James Currey.

Bender, M. Lionel. 1975. The beginnings of ethnohistory in Western Wellegga: The Mao problem. In Robert K. Herbert (ed.), Patterns in language, culture, and society: Sub-Saharan Africa, Proceedings of the Symposium on African Language, Culture, and Society, held at the Ohio State University, Columbus, 11 April 1975 (Working Papers in Linguistics 19), 125-141. Columbus, OH: Department of Linguistics, The Ohio State University.

Bender, M. Lionel. 1976. Introduction. In M. Lionel Bender (ed.), The non-Semitic languages of Ethiopia, 1-24. East Lansing, MI: African Studies Center, Michigan State University.

Central Statistical Agency. 2007. Summary and statistical report of the 2007 population and housing census. Population size by age and sex. Addis Ababa: Federal Democratic Republic of Ethiopia Population Census Commission (accessed 16 June 2021).

Constitution of the Federal Democratic Republic of Ethiopia. 1995. Federal Negarit Gazeta. Proclamation No. 1/1995 (accessed 5 April 2021).

Cornips, Leonie & Vincent A. Rooij. 2018. Introduction: Belonging through linguistic place-making in center-periphery constellations. In Leonie Cornips & Vincent A. Rooij (eds.), The sociolinguistics of place and belonging: Perspectives from the margins, 1-14. Amsterdam: John Benjamins.

Data Dea. 2005. Christianity and spirit mediums: Experiencing post-socialist religious freedom in southern Ethiopia (Working Paper 75). Halle/Saale: Max Planck Institute for Social Anthropology.

Donham, Donald L. 2000. On being “first”: Making history by two’s in southern Ethiopia. Northeast African Studies 7(3). 21-33.

Eberhard, David M., Gary F. Simons & Charles D. Fennig (eds.). 2022. Ethnologue: Languages of the world. 25th ed. Dallas, TX: SIL International (accessed 4 April 2022).

Eriksen, Thomas Hylland. 1992. Linguistic hegemony and minority resistance. Journal of Peace Research 29(3). 313-332.

Fekadu Adugna. 2011. Overlapping nationalist projects and contested spaces: The Oromo-Somali borderlands in southern Ethiopia. Journal of Eastern African Studies 5(4). 773-787.

Fleming, Harold C. 1976. Omotic overview. In M. Lionel Bender (ed.), The non-Semitic Languages of Ethiopia, 299-323. East Lansing, MI: African Studies Center, Michigan State University.

Fleming, Harold C. 1984. The importance of Mao in Ethiopian history. In Sven Rubenson (ed.), Proceedings of the Seventh International Conference of Ethiopian Studies, University of Lund, 26-29 April 1982, 31-38. Uppsala/Addis Ababa: Scandinavian Institute of African Studies/Institute of Ethiopian Studies.

Geschiere, Peter & Francis Nyamnjoh. 2000. Capitalism and autochthony: The seesaw of mobility and belonging. Public Culture 12(2). 423-452.

Getachew Kassa Bahiru. 2015. A grammar of Hoozo. Addis Ababa: Addis Ababa University (PhD dissertation).

González-Ruibal, Alfredo. 2014. An archaeology of resistance: Materiality and time in an African borderland. Lanham, MD: Rowman & Littlefield.

Heller, Monica & Alexandre Duchêne. 2008. Discourses of endangerment: Sociolinguistics, globalization and social order. In Alexandre Duchêne & Monica Heller (eds.), Discourses of endangerment: Ideology and interest in the defence of languages, 1-13. London: Continuum.

Hultin, Jan. 1987. ‘Sons of slaves’ or ‘sons of boys’: On the premise of rank among the Macha Oromo. In Jan Hultin (ed.), The long journey. Essays on history, descent and land among the Macha Oromo, 1-19. Uppsala: Uppsala University.

James, Wendy. 1980. From aboriginal to frontier society in western Ethiopia. (Ed.) Donald L. Donham & Wendy James. Cambridge: African Studies Center.

Jenkins, Richard. 2008. Rethinking ethnicity. 2nd ed. Los Angeles, CA: Sage.

Jeylan Wolyie Hussein. 2016. Examining the nature of defensively situated politics of difference, identity and essentialism in Ethiopia: A critical engagement. International Area Studies Review 19(4). 340-354.

Küspert, Klaus-Christian. 2015. The Mao and Komo languages in the Begi-Tongo area in western Ethiopia: Classification, designations, distribution. Linguistic Discovery 13(1). 1-63.

Küspert, Klaus-Christian. 2020. Språk og etnisitet. NMS misjonen blant Gumuz i Etiopia 1970-2020. Norsk Tidsskrift for Misjon 74(1). 21-40.

Küspert-Rakotondrainy, Sophie. 2016. Language policy and identity in a diverse society. The impact of language of instruction on social group identity. A comparative case study from Ethiopia. Saarbrücken: Lambert Academic Publishing.

Küspert-Rakotondrainy, Sophie. 2021. Who are the ‘true’ Mao? A contested identity in Mao Komo Liyu Wäräda. Journal of Ethiopian Studies 54(1). 37-68.

Küspert-Rakotondrainy, Sophie. Forthcoming. The power of the past: Slavery, stigma and social stratification in Western Wellega. In Ahmed Hassen, Giulia Bonacci, Alexander Meckelburg & Serena Tolino (eds.), Slavery and the slave trade in Ethiopia revisited. Wiesbaden: Harrassowitz.

Léglise, Isabelle & Sophie Alby. 2006. Minorization and the process of (de)minoritization: The case of Kali’na in French Guiana. International Journal of the Sociology of Language 182. 67–85.

May, Stephen. 2012. Language and minority rights: Ethnicity, nationalism and the politics of language. 1st ed. New York: Routledge.

Meckelburg, Alexander. 2018. Minority integration and citizenship expansion. Observations from the Mao and Komo groups in western Ethiopia. In Susanne Epple (ed.), The state of status groups in Ethiopia: Minorities between marginalization and integration, 173-192. Berlin: Dietrich Reimer.

Meckelburg, Alexander & Sophie Küspert-Rakotondrainy. 2019. Culture summary: Mao. Human Relations Area Files (accessed 15 June 2021).

Merera Gudina. 2003. Ethiopia: Competing ethnic nationalisms and the quest for democracy 1960-2000. Maastricht: Shaker.

Messay Kebede. 2006. The ethnicization of Ethiopian politics: Origins and significance. Ethiopian Review 30 (November). 1-20.

Mufwene, Salikoko S. 2017. Language vitality: The weak theoretical underpinnings of what can be an exciting research area. Language 93(4). 202-223 (accessed 7 October 2022).

Mühlhäusler, Peter. 1996. Linguistic ecology: Language change and linguistic imperialism in the Pacific region. London: Routledge.

Negaso Gidada. 2001. History of the Sayyoo Oromo of southwestern Wallaga, Ethiopia, from about 1730 to 1886. Addis Ababa: Mega.

Patrick, Donna. 2008. Indigenous language endangerment and the unfinished business of nation states. In Alexandre Duchêne & M. Monica Heller (eds.), Discourses of endangerment: Ideology and interest in the defence of languages, 35-56. London: Continuum.

Patrick, Donna. 2010. Language dominance and minorization. In Jürgen Jaspers, Jan-Ola Östman & Jef Verschueren (eds.), Society and language use, 176-191. Amsterdam: John Benjamins.

Phillipson, Robert & Tove Skutnabb-Kangas. 1995. Linguistic rights and wrongs. Applied Linguistics 16(4). 483-504.

Revised Constitution of the Benishangul Gumuz Regional State. 2002. Proclamation No. 31/2002. Asosa.

Revised Constitution of the Oromia Region. 2001. Megeleta Oromia. Proclamation No. 46/2001. Finfinnee.

Siebert, Ralph, Kati Siebert & Klaus Wedekind. 2002. Sociolinguistic survey report on languages of the Asosa-Begi-Komosha area: Part I. Dallas, TX: SIL International (accessed 23 February 2022).

SIL. 2011. Basic literacy and numeracy baseline assessment of grades 1, 2 and 3 students in Benshangul Gumuz Region. Draft. Addis Ababa (accessed 18 May 2021).

Skutnabb-Kangas, Tove & Robert Phillipson (eds.). 2010. Linguistic human rights: Overcoming linguistic discrimination. Berlin: De Gruyter Mouton.

Tesema Ta’a. 2006. The political economy of an African society in transformation: The case of Macca Oromo (Ethiopia). Wiesbaden: Harrassowitz.

UNESCO. 2003. Language vitality and endangerment. UNESCO Ad Hoc Expert Group on Endangered Languages. Document submitted to the International Expert Meeting on UNESCO Programme Safeguarding of Endangered Languages (accessed 3 April 2021).

UNESCO. 2017. Atlas of the world’s languages in danger (accessed 3 April 2021).

Van der Beken, Christophe. 2015. Federalism, local government and minority protection in Ethiopia: Opportunities and challenges. Journal of African Law 59(1). 150-177.

Vaughan, Sarah. 2006. Responses to ethnic federalism in Ethiopia’s southern region. In David Turton (ed.), Ethnic federalism: The Ethiopian experience in comparative perspective, 181-207. Oxford: James Currey.

Haut de page

Notes

1 The authors have known Isak, which is a pseudonym, since 2012. For several years, we worked with him in activities for language development, and he has also acted as a guide and translator on fieldtrips. The primary author has lived with his family in the countryside. We still keep in contact with Isak and his family although we no longer have a professional relationship. His story is used with consent.

2 Ethnologue: ISO 639-3 code hoz, glottocode: hozo1236.

3 The term kap’á is used to express honour, reverence, or respect to a person or deity (e.g. Naa naa diya kap’adi ‘I am honouring my father’); wak’i means ‘fear’, which in many cases is interchangeable with ‘honour’.

4 Since the security situation in western Ethiopia had drastically deteriorated by the time of the second period of fieldwork, some interviews had to be carried out in the towns Begi and Kondala or among the Hozo-speaking diaspora in the neighbouring Benishangul Gumuz Region, either with members of the diaspora or their relatives who had come to visit.

5 The discussion on autochthony is not peculiar to Ethiopia. Wherever there is an “indigenous” group, there is usually a sub-group that is “more indigenous” (Mufwene 2017: 212). For more discussion concerning the challenges related to discourses on autochthony see, for example, Donham (2000) and Geschiere & Nyamnjoh (2000).

6 Oromo nationalism cannot be discussed in detail here. For more discussion, see Baxter (1994), Merera (2003), Messay (2006), and Jeylan (2016).

7 We are not aware of any meaning of ak’mo other than ‘Mao people’; more research into the origin of the term is needed.

8 In this article, we mainly consider the situation of Hozo with a sideways glance at Seze, both of which are spoken in an area that does not overlap much with Gwama.

9 There is new evidence of villages with a proportion of Hozo speakers south, east and north-east of the Dabus River that could increase the number of speakers by a few thousand.

10 This was done in group discussions with elders in two different sub-settlements of Gemi Gaba by estimating the number of children and then counting the number of families where the parents speak Hozo but not the children.

11 The work of NGOs on language documentation and the development of a preliminary Hozo orthography will be addressed below.

12 This information is also based on personal observations from the authors’ cooperation with the local churches as well as interviews with current and former church leaders, and former Norwegian missionaries, most notably: 18 May 2019 (former church leader), 25-26 September 2020 (current church leaders) and 21 April 2021 (former missionary).

13 Cf. Patrick (2010: 187) on inequality between minorised languages and the dominant language due to internal migration.

14 Furthermore, language development can lead to status conflicts, as the example of southern Ethiopia shows (Cohen, 2000).

15 The borana social class used to play an important role in Oromo tradition, as only individuals of borana, i.e. “pure”, descent were fit to take the roles of spiritual leaders (Bartels 1983). “Pure”, in this context means of non-slave descent and descent from men who went through the social gadaa system with its eight-year cycles in the prescribed way (Hultin 1987). The term borana has today lost its relationship to religion but is sometimes still used as a local term for someone who is of “pure” descent.

Haut de page

Table des illustrations

Titre Figure 1 — Map showing the area where Hozo and other Mao languages are spoken (prepared on the basis of the maps in Küspert 2015).
Légende Language names are shown in white, administrative units in all upper case, and district capitals in the research area in bold red.
URL http://journals.openedition.org/lla/docannexe/image/2075/img-1.png
Fichier image/png, 131k
Haut de page

Pour citer cet article

Référence électronique

Sophie Küspert-Rakotondrainy et Klaus-Christian Küspert, « A matter of honour: Speaking Hozo (Ak’mo Wandi), an endangered language in western Ethiopia »Linguistique et langues africaines [En ligne], 8(1) | 2022, mis en ligne le 30 juin 2022, consulté le 21 mai 2026. URL : http://journals.openedition.org/lla/2075 ; DOI : https://doi.org/10.4000/lla.2075

Haut de page

Auteurs

Sophie Küspert-Rakotondrainy

University of Birmingham

Klaus-Christian Küspert

Norwegian Mission Society

Haut de page

Droits d’auteur

CC-BY-SA-4.0

Le texte seul est utilisable sous licence CC BY-SA 4.0. Les autres éléments (illustrations, fichiers annexes importés) sont susceptibles d’être soumis à des autorisations d’usage spécifiques.

Haut de page
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search