The Mao and Komo Languages in the Begi–Tongo area in Western Ethiopia: Classification, Designations, Distribution

The people who are usually referred to as 'Mao' include the following: (1) Anfillo, [...] not Nilo-Saharan speakers (at least not now); (2) The Komo and Kwama of west and south of Begi [...] (3) The Hozo and Sezo clans of the Begi area; (4) The Maramo clans of the Bambeshi area.[...]. Of these only in Hozo (ma^w) and Sezo (ma:y), is 'Mao' or a variant of it, the word for 'people'. However, Bambeshi seems to be fairly closely related to Hozo and Sezo, and I shall henceforth refer to Hozo, Sezo, and Bambeshi as the Mao languages. ^[2]

Village names: Blue: Mainly Koman speakers; Green: Mainly Omotic speakers
Village points: Light blue: Komo; Dark blue: Gwama/Koman Mao (cf. 4.1.2); Light green: Seze; Dark green: Hozo/Shuluyo (cf. 4.2.3)

This statement by Wendy James corresponds well with the findings in the interviews. On question 6 in the questionnaire: “What do other people call you” (appendix 1), the vast majority of interviewees in the Western Wollega zone of Oromia answered “Mao”. This can be considered as an official designation and is widely accepted by the different groups of indigenous people in the area. At this level, the language differences Koman vs. Omotic is not of importance. In Mao Komo Special Woreda, the picture is more complicated. While the term Mao is most common both as outside and self-designation of the indigenous people around Tongo and in the north-eastern part of the Woreda (cf. appendix 2), the indigenous people of the lowland areas around Lak’e, Yangu, Keser and Zebsher are called Komo. The term Komo here functions as an official designation used by the government. Also in this case the language differences (Komo vs. Gwama) does not play a decisive role.

In the transition area west and north of Tongo you will find people living together in the same village and speaking dialects of the same language, but attributing themselves either to Mao or to Komo, as well as people being summarised as Komo in spite of speaking different languages. The border between Oromia (Western Wollega zone) and Benishangul Gumuz (Mao Komo Special Woreda) does not function as a boundary between ethnic groups or ethnic designations; the boundary between the ethnic groups seems to rather go between lowlands and highlands – or in their “distance to the Oromo culture.”^[23]

3.3 The Mao: One ethnicity, different dialects

James (1981:29) states that Mao can be considered a social term that identifies indigenous people with a close relationship to the dominating Oromo population, not as equal partners but as servants or farm workers depending on the land owners. The highlands – lowlands opposition may reinforce this fact since the Oromo and thus the Mao population traditionally is limited to the areas with higher elevation. Mao can be speakers of Omotic languages, yet also speakers of a Koman language (cf. Bender 1975:125). However, on a deeper level we see some differences. When asked for self-designation, speakers of Omotic languages call themselves Mao without any hesitation, while it turns out that the term Mao is not equally accepted by speakers of Koman Mao:

The word Mao^[24] is Omotic and means ‘man; people’, occurring as [ma:ɪ] in Seze and as [mɔ:] in Hozo^[25]. In the Koman Mao dialects the meaning of a word <mao> is opaque. It would, however, give a wrong picture to limit the term Mao to the speakers of Omotic languages.

In figure 2 the villages with mainly speakers of Koman Mao/Gwama and Komo are arranged in clusters according to the self-designation of the people (P) and their languages (L). The arrangement of the villages is based on the answers of the questions 3.What do you call your people? (P) and 4.What do you call your language? (L). Neighbouring villages with for the most part identical answers are combined in an oval.

Regarding the ethnic self-designation (P), we see a significant number of Mao entries all over the central area there, while Komo and Gwama is dominating in the western lowlands and the Kuro and Sit Shwala^[26] mainly in the northern and eastern parts. It appears that Mao and Sit Shwala here often are used synonymously with a predominance of Sit Shwala when asked for ethnicity. The use of Sit Shwala is possibly a demarcation against the Oromos who prefer or even impose the term Mao. Mao may probably feel “closer to Oromo” than it Shwala.

In the villages around Shoshor Butuji east of Tongo and in Tongo town, Mao is the most common self-designation and Sit Shwala recedes. Here it seems that particularly the prominent Warra Seta clan^[27] uses Mao both for ethnicity and language. A further investigation of this clan and its history and political influence might give some more information.

Conclusively we may state that Mao is frequently used as an ethnic term, and that there is a kind of Mao identity across the language differences. The fact that there is a considerable group of people speaking a Koman language who consider themselves and their language Mao makes it impossible to restrict this designation to speakers of an Omotic language.

3.4 Mao and Gwama: one language, different ethnicity

Somi Ambeyu is an old Mao man from Kokeb. The village is located about 15 km northwest of Tongo at 1230m altitudes, halfway down to the lowlands towards the border of South Sudan. When he was asked for the difference between Mao and Gwama in Kokeb, he recounted:

Here Komo and Mao are used as both ethnic and linguistic terms, while Gwama seems to function only as a language term. Some Komo people may speak Komo, others Gwama.

In a group discussion with Gwama speakers in Zebsher, a few kilometres north-west of Kokeb and still further down in the lowlands, we recorded the following quotation:

This area of transition between the highlands and the lowlands (cf. map 3), seems to be the area of some dynamic movements. Not only Oromo people move down as office workers, workers for land investors and farmers in search of fertile land, but also Mao from the highlands speaking a language very similar to Gwama, but with a different ethnic identity.

Gwama is, however, not merely the name for a language. As shown in figure 2, in the village cluster north-east of Tongo where amongst others Kokeb and S’ulgolo are grouped, we find Gwama and Mao as ethnic self-designations side by side. These terms do not refer to the same ethnic group as it is the case with Mao and Sit Shwala. The ethnic Gwama – called Komo by others - will not call themselves Mao while ethnic Mao do not accept the term Gwama nor Komo as a name for themselves. This seems to depend on a mutual dislike: The Mao with their Oromo-influenced way of living may consider the Gwama as “far from civilisation”, while the Gwama look at the Mao as “mixed” and “oromised”.^[28]

3.5 Gwama: More than a linguistic term

The following statement gives the impression that Gwama is merely a linguistic term:

As mentioned in 3.2, Komo is used by others as an ethnic term for speakers of Komo and Gwama. This may be because of political reasons.^[29] In Benishangul Gumuz Regional State, Komo is – besides Mao - established as an ethnic term while Gwama is not (cf. the name Mao Komo Special Woreda); thus Komo is used for Komo and Gwama speakers by the authorities.

On a lower level, the picture will be more nuanced:

Gwama is obviously used as an ethnic term as well, not only as a linguistic term. There are e.g. Gwama houses^[30] in Ishgogo and in other areas, beehive huts that play an important role in the culture and believe of the Gwama people.^[31] This indicates a cultural Gwama identity. Figure 2 shows further that Gwama is used as an ethnic term in most of the villages west of Tongo when people were asked for the ethnic self-designation (cf. question 3, appendix 1).

We can conclude that there is a correspondence on the level of ethnic self-designation of the speakers of Komo and Gwama in the lowland areas west of Tongo. The Gwama speakers do have an ethnic Gwama identity, but are classified as Komo by the government and would rather call themselves Komo than Mao in spite of the language differences. The Komo always call themselves Komo, but they are aware that there are others called Komo speaking a different language than they do (Bryant et al. 2007:460).

3.6 Ethnic identity and clan affiliation

A deeper analysis of the relationship between clan affiliation, ethnic self-identification and language is beyond the scope of this study. A more detailed investigation of the raw data listed in appendix 2 might also give increased knowledge and insight. So far, my findings show that, to some extent, there is a cross linguistic clan identity; this clan affiliation seems to be stronger than the language

Obviously there is a Mao identity beyond the borders of the language families, e.g. Koman Mao and Omotic Mao speakers may feel common clan loyalties which, amongst other things, result in intermarriage, while Gwama speakers have affiliations to the Komo in spite of language differences. At this point, it can only be assumed that this has to do with language change and voluntary or involuntary displacement of entire ethnic groups due to political and social conditions. Further investigations could possibly shed some light on language change and migration in this area. Slavery may have had a huge impact on the societies in the region - coupled with common exogamy in marriage practices, and may have caused creolisation or change of languages or cross linguistic ethnic affiliations.

4. Language and nomenclature

In spite of the frequent use of terms like “the Mao language” or “the Mao Komo language” even by officials in Tongo and Asosa who actually speak one of the concerned languages and are well aware that these are different and mutually unintelligible, it is now generally accepted that the languages subject to this survey belong to different language families: the Koman and the Omotic languages.

In the eastern part of the survey area we find the Omotic languages Seze and Hozo^[32], while the related Northern Mao is spoken around Bambasi north of our survey area and in the Didessa valley. The Omotic language Ganza is spoken by small minorities in some Koman speaking villages around Zebsher. In the western part of the area and continuing to the other side of the border to Sudan/South Sudan, the indigenous peoples Komo and Gwama speak Koman languages. Mao is not the ethnic or linguistic self-designation of these peoples and is rarely used by others.

Between those distinct eastern and western groups we find a considerable number of speakers of a Koman language on both sides of the border between Benishangul Gumuz and Oromia region who would accept to be called Mao people speaking a Mao language and partly use Mao as ethnic and linguistic self designation (compare maps 3 and 4). This group is often overlooked, e.g. when the term “Mao languages” is reserved to the Omotic Hozo, Seze and Northern Mao, or when the Koman languages in the area are limited to Gwama and Komo.

As shown in map 4, the border between Omotic and Koman languages goes slightly south and east of the road Bambasi – Begi, i.e. most of the speakers of Omotic Mao in this area do not live in Begi Woreda, but in K’ondala or in Babo Gambel Woredas. Along this road there are minorities of Omotic Mao speakers in villages with a relative Koman Mao majority^[33] (cf. appendix 3 and map 2). Except these meeting places there are very few villages with both Koman and Omotic speakers.^[34] The border between the regional states, i.e. the eastern border of Mao Komo Special Woreda neither occurs as a language border nor as a boundary between ethnic groups.

4.1 The Koman varieties

This survey focuses on the Koman languages Komo and Gwama without any further discussion of the internal relationship within the Koman languages or the relationship between the Koman languages and other Nilo-Saharan languages like Berta and Gumuz.^[35] Opo is, together with Berta, mentioned in several villages as an additional language to Gwama and Komo.^[36]

The relationship between Gwama and the closely related language varieties on both sides of the border between Mao Komo Special Woreda and Begi Woreda in Oromia - here referred to as Koman Mao - will be given special attention. This may give a satisfying answer to the question whether these Koman Mao dialects constitute one or more separate languages, or if they - together with Gwama - are varieties of one and the same language.

4.1.1 Komo and Gwama

In figure 2 above all villages with a Koman language speaking majority, which are part of this survey, are arranged in clusters according to the self-designation of the people groups regarding their ethnicity (P) and language (L). The findings are derived from the answers to question 3: “What do you call your people” and question 4 “What do you call your language” in the questionnaire (appendix 1). When systemising the answers of the questionnaires some decisions had to be made. This was particularly the case with the term Kwama that in some cases may be a synonym for Gwama, while in others – where it is mentioned side by side with Gwama and obviously refers to a different group of speakers - it may be taken as the Gwama term for the Komo language.^[37]

The word list interviews and recordings (appendix 3) give a clear understanding that there are two different Koman languages in the area – disregarding Opo and possibly also Uduk: Komo and Gwama.^[38] The word list in appendix 4 shows that there are around 30 % cognates between Gwama (with related varieties) and Komo.^[39] This indicates that Gwama and Komo are related but not mutually understandable languages and corresponds with the findings in Davis et al. (2011:19), where a Kwama cluster and a Komo cluster are discovered: “The results of this survey have confirmed the existence of a separate Komo language but have also shown that the Komo people are commonly bilingual in Kwama.”

The existence of two separate Koman languages in the area and the bilingualism of Komo speakers is confirmed in several field interviews. According to Abdulalem, a teacher in Penshuba and a Gwama speaker, Komo is spoken in Yangu, Lak’e and Kawi Shumate and is not understood by Gwama speakers. Abdulalem calls the language “Kwama Dini, which is the same as Komo” (Penshuba, 10/2012).^[40] In a group interview in Zebsher it was mentioned that “they who live in Keser may also understand Komo. All Komo speakers understand Gwama. The Gwama speakers cannot understand Komo.” (Zebsher 10/2012).

The same information was given about Yangu and Lak’e:

See also Bryant et al. (2007:460)

Map 5 shows the distribution between Komo and Koman Mao/Gwama in our survey area. Komo was found to be spoken only in the westernmost part of the Koman speaking area – in the lowlands towards the border of Sudan.^[42] There is a Komo speaking community in the Woreda town Tongo, which may be due to rather recent immigration. As shown in map 2, the Komo and Gwama speakers live together in the same village, which makes it difficult to draw language maps.^[43]

According to interviews, the villages of Yiwa, Met’i and Kawi Shumate, which may have had a Komo speaking majority, were abandoned and the population has been moved to Yangu, Lak’e and Keser because of governmental resettlement programmes from 2009 to 2011. There they live as minorities in mainly Gwama speaking villages.^[44] There are no or only few Komo speakers in villages with a majority of Koman Mao speakers. It may be repeated that the Gwama speakers living together with the Komo speakers prefer to be called Komo rather than Mao in spite of the fact that their language and the Koman Mao dialects appear to be very similar while the Komo language is not understandable for them (cf. 3.6 and 4.1.2).

Outside our survey area, some recordings indicate a few settlements with Komo and Gwama speakers further south along the border to South Sudan in Gidami Woreda in Oromia. A Komo speaking population is found in the Yabus area in Sudan.^[45] Theis (1995:13) estimates the number of Komo speakers in Sudan to be around 5000.^[46] In Gambela Regional State, Pokung Special K’ebele on the road to Itang, 28 km from Gambela town has mainly Komo speakers. There are small Komo speaking groups in Gambela town (Batur/Doser Lami) and in Bonga on the road to Bure in Oromia.

4.1.2 T’wa Gwama, T’wa Sit Shwala or Afaan Mao?

In the Koman speaking area east of the road Bambasi – Tongo there were 3 main answers to question 4 in the questionnaire (cf. appendix 1) “What do you call your language”: T’wa Gwama, T’wa Sit Shwala and Afaan Mao^[47]. It seems that these terms designate closely related languages or varieties of the same language in spite of different ethnic affiliations of the speakers (cf. section 3).^[48] The following section endeavours to shed some light on the use of these terms as linguistic designations and thus in the relationship between the language varieties.

The reference to “[o]ne of the dialects of K[wama], called Mao” and “Afan Mao” as a synonym for Gwama/Kwama are of particular interest. Kievit et al. (2011:14/15) give the following information:

The findings of this survey do confirm that the language called “Mao [...] spoken in the Begi and Tongo areas” is a Koman language very similar to Gwama, spoken in a wide area in Begi Woreda and the eastern part of Mao Komo Special Woreda. This language or language variety is mainly called T’wa Sit Shwala or – oromised – Afaan Mao by the speakers.

Out of the results of the interviews and the word lists (cf. appendix 3) and on the basis of the data provided in map 2, 4 and 5 as well as in the figures 1 and 2 we can draw the following conclusions:

(1) Gwama and the languages varieties mainly called T’wa Sit Shwalaby the speakers (Koman Mao) show a very high degree of cognates.

On the long word list (78 items in English; entries based on own voice recordings; cf. appendix 3) as well as on the short wordlist based on written recordings by the survey teams there are only few clear lexical differences between Gwama and the Koman Mao varieties T’wa Sit Shwala and Kuro. The cognates between Gwama and T’wa Sit Shwala are more than 90 %. Table 2 shows some of the few lexical differences. The forms for Gwama and Sit Shwala were cross checked with speakers from different villages; the Kuro-forms are taken from the word list and need further investigation.^[49]

In several interviews with Gwama speakers it was mentioned that the Mao (i.e. Koman Mao) speakers use a lot of different words, and that they are partly oromised. This may be the case in some areas; however, it is not clearly reflected in the limited material of this survey.

(2) There are some regular sound correspondences between Gwama and the Koman Mao languages varieties mainly called T’wa Sit Shwala by the speakers.

The most prominent of these sound differences is [j] vs. [l] as in [kaja] ‘sun’ in Gwama vs. [kala] ‘sun’ in all Koman Mao varieties. This feature is well known by the speakers.All other sound differences reflect tendencies, like e.g. the [b] and [mb] in (4). The Kuro variety normally shows the forms with [mb]. Together with some other Kuro features such as [s^wɑnt’] vs. [sɔnt’] ‘foot’, the retaining of ending vowels as in [ant’i] vs. [ant’] or [bit’i] vs. [bit’] and the lexical differences as shown in table 2, the dialects in the north-eastern part of the area seem to stand out in several ways. Our material is, however, not sufficient to give answers on theories regarding origin, migration and language change of these people groups.

None of the differences listed in table 2 and 3 should hinder the communication and enhance the theory that Gwama and the varieties of Koman Mao are mutual understandable.

(3) Gwama and the languages varieties of Koman Mao are mutually understandable.

In order to check statements like this gained through the interviews and the conclusion drawn from the word lists, we did two short experiments during a workshop in Asosa in July 2013 with speakers from different dialect areas.^[55] In one case a T’wa Sit Shwala (Koman Mao) speaker told a story based on a series of pictures in their mother tongue and without the help of additional explanations in Afaan Oromo, and then the Gwama speakers would retell this story in Gwama. In the second case a Gwama speaker described a picture in Gwama and the T’wa Sit Shwala speakers would draw a picture based on what they had been told. In both cases, the listeners were able to reproduce the story respectively the images correctly.

The experiments confirmed the mutual intelligibility of the varieties represented at the workshop. These findings were strengthened by the fact that the Gwama speakers easily functioned as interpreters by using their mother tongue for ‘Koman Mao’ speakers who did not speak Amharic. In the cases where the communication between the Gwama speakers from the lowlands and the Koman Mao speakers mainly from Begi Woreda in Oromia Regional State was difficult or failed, the main problem seemed not to be a linguistic one. In almost all cases the Koman Mao speakers showed a significantly poorer language competence than the Gwama speakers from the lowlands, and in many cases they preferred to use Afaan Oromo instead of Gwama/Koman Mao.

(4) In spite of mutual intelligibility and a high degree of cognates, there is only little sense of unity between the speakers of Gwama and the varieties of Koman Mao.

In figure 2 the information of the interviews are systematised and the villages grouped according to the ethnic and linguistic self-designation of the people. The figure shows a clear differentiation between an eastern, northern and central area (Begi Woreda and north-eastern Mao Komo Special Woreda) on one side and the West (western part of Mao Komo Special Woreda) on the other side. In the North, the East and around Tongo the people call themselves Sit Shwala or Mao and their language T’wa Sit Shwala or Afaan Mao.^[56] Mao occurs more often as an ethnic designation than as a language name. Gwama is in very few cases mentioned as a language name only.

In the villages in the western and south-western part of the survey area the people call themselves Komo or Gwama and their language Komo for “proper Komo” and Gwama. Gwama and Komo speakers live normally together in the same village. The Gwama speakers would never call their language Afaan Mao or T’wa Sit Shwala, but they are aware of the existence of a people group commonly called Mao, which speaks a similar language as they do.

In the central western villages of Kokeb, S’ulgolo, Bobis Ishkaba and Mimi Akobo on the slope from the Tongo Highland down to the Sudan plains, both people groups meet and define themselves as either Mao/Sit Shwala, speaking Afan Mao/T’wa Sit Shwala or as Gwama, speaking T’wa Gwama. In an interview conducted in Kokeb we got an immediate impression that there is a considerable difference between those people groups and their languages. While the Mao here occur as immigrants from the highlands looking for fertile soil, the Gwama are said to be the indigenous population with roots to the lowlands in the west and closely related to the Komo people. If we, however, take a look at the linguistic findings and compare the word list interviews with Somi Ambeya who calls himself and his language Mao and Abdulalem Atair who calls himself and his language Gwama, the difference is small and qualifies the forms as varieties of the same language.^[57] Thus the difference between these two people groups which is expressed so clearly in the interviews seems to have other reasons than linguistic.

4.1.3 Gwama or Koman Mao - a language with no name

Map 6 shows the main self-designations for the Koman and Omotic languages subject to this survey. On the basis of linguistic findings, the Koman languages can be divided into Komo (light blue) and Gwama/Koman Mao (dark blue). It has been proved that Gwama and the Koman Mao varieties are mutually intelligible and thus can be considered to be varieties of the same language (cf. 4.1.2). The answers to question 10 in the questionnaire (cf. appendix 1): “Where else (in which villages) do people speak your language?” confirm this view. As seen in appendix 2, table 1, column 9 in many cases typical Gwama speaking villages are mentioned by Koman Mao speakers in Begi Woreda as references where they speak “the same language”. There are few or no examples where it is the other way round. In typical Gwama speaking villages in the lowlands of Mao Komo Special Woreda, highland villages where the people call their language T’wa Sit Shwala, T’wa Kuro or T’wa Kirin are not mentioned as references where people might speak the same language.

As explained earlier the term Mao is arbitrary and would not be accepted by the Gwama people to designate neither their language nor their ethnicity. The – artificial – term Koman Mao (cf. map 5) used in this paper is a somewhat more precise term than just Mao and excludes the Omotic Mao languages. It would, however, not include Gwama.^[58]

The term Sit Shwala for the people and T’wa Sit Shwala for the language is widely used as self-designations by the people groups called Mao by others, and may fit as a synonym of Koman Mao and as a generic term for the clan related terms Kuro and Kirin.^[59] It would not include Gwama and is refused by the lowland Gwama speakers. This was made clear during the workshops on the Gwama language in Asosa in July and October2013. Adadik Habte, a young Gwama from Ya’a Baldigis in Mao Komo Special Woreda clearly refused the use of T’wa Sit Shwala in spite of the fact that his language and the T’wa Sit Shwala language of other participants were easily mutually understandable. According to Asadik the term Sit Shwala simply refers to “any kind of black people” and is not distinct enough for being used for a particular language; thus also speakers of an Omotic Mao language could be called Sit Shwala – ‘black people’. The discussion gave the impression that the Gwama speakers disliked this term and that Sit Shwala might have something to do with the complicated and asymmetric relationship between Mao and Oromo people particularly in Begi Woreda. The Sit Shwala participants in the same workshop insisted that this term did not have any pejorative connotation, but they were more inclined to accept Gwama as a positive common designation than the Gwama people were to accept (T’wa) Sit Shwala.

There are Gwama people speaking the Gwama language on the other side of the border in Sudan.^[60] The terms Mao and Sit Shwala do not seem to be in use in Sudan and would probably not be accepted. The Ethiopian Gwama are aware of the existence of Gwama in Sudan and they have an understanding of a common origin. The Mao – if Koman or Omotic speaking –have a rather vague idea of “coming from the highlands” and “belonging to the same family” in spite of different languages.

All this suggests that Gwama might be the more useful and even more prestigious term compared with T’wa Sit Shwala, which is ambiguous and potentially pejorative, and Koman Mao, which is an artificial term. Gwama is already widely accepted in literature, but it still excludes thousands of speakers of very similar dialects.

4.1.4 Kirin and Kuro

In some of the interviews Kirin (Kring; Kiring ክሪንግ) and Kuro were mentioned either as ethnic terms for people groups and/or clans, or as language self-designations. Both terms seem to refer to clans or sub groups of ‘Koman Mao’ and to their language.^[61]

Kirin is mainly reported from some villages around Tongo, such as Bobos Ishkaba and Wanga Git’en, as well as from Tongo town (cf. appendix 2 and map 6). There is one isolated reporting from Lak’e in the lowlands. In all cases, Kirin as a linguistic term stands in contrast to Gwama and Komo. The information that Gwama and Kirin are mutually understandable indicates that Kirin here designates a variety of Koman Mao.^[62]

Kirin may also designate a particular group of Mao with Arabic origin and a light complexion that forms a kind of political “Mao elite” (Alexander Meckelburg, p.c.). According to oral traditions, they learned Koman Mao only three generations ago when they immigrated to or were placed in the Tongo area. Nowadays some Kirin people (Tongo Mao) may call their language Gwama, no longer Mao because of political or status reasons. This avoids confusion with the Omotic Mao languages east of Begi.

A discussion of this complex issue exceeds the boundaries of this paper, and a more thorough investigation of this form of Koman Mao and their speakers might be fruitful, as well as of the reported use of Kirin for the languages of the Seze and Hozo people. So far, the meaning of the term Kirin does not really seem clear to me.

Kuro as an ethnic and linguistic term is mainly reported from some villages in the north-eastern part of the survey area, such as Kongilo Gara Kelo, Giba Gulanza, Shombo Bayida and Kobor Chandi (cf. figure 2, appendix 2 and map 6). Differing from Kirin, Kuro often occurs as the only term both for the language and the people in the villages of this defined area.

According to other Koman Mao speakers, Kuro is a clan name, and as a language term it is simply “a dialect of T’wa Sit Shwala”.^[63] As mentioned in 4.1.2 (cf. table 2), the Kuro speakers use a significant number of different words compared with all other Koman Mao and Gwama varieties, which could give a reason to consider Kuro a distinct dialect of Koman Mao. A confirmation of this assumption would require further investigation. It should be noted that an Omotic speaking minority (Mao Seze or Mao Hozo; cf. appendix 2) is reported from the Kuro villages mentioned above; Kuro thus appears at the interface between Koman Mao and Omotic Mao.

In the villages Gemi Gaba and Laliftu Lop'i with a Hozo or Seze speaking majority, Kuro is mentioned as an additional language. This might suggest that there is a Koman Mao speaking minority in these villages. The information we got from Shagga that there were some old people in this village, who speak “an old and traditional language called Kuro” could indicate a similar situation. According to Kasim Hassan, Kuro is a “dying language" in this area that is different from all other languages; young people would not understand it.^[64]

In Shera Kama, also on the boundary between Koman Mao and Omotic Mao, Kuro is mentioned as the name of a clan (cf. appendix 2). Some of the Kuro people in Shera Kama appear to speak Koman Mao and others an Omotic language - Seze or Hozo. It is noteworthy that some of the speakers of Koman Mao in Shera Kama acknowledge the people in Shagga and Kabache, villages that have clearly Omotic speaking populations, as “the same people”.

Tadesse Fayisa from Ego Girmos, a Koman Mao speaking village, explains that the majority of the people in his village belong to the clan of Kur. They traditionally intermarry with Seze and Hozo. He states:

This information could be connected to the notice from Shagga, Kabache’s neighbouring village, about the rests of a Kuro speaking population and would possibly also explain why there is a relationship between the Koman Mao speakers in Shera Kama and the people in Shagga and Kabache.

It seems obvious that the clan membership is more important for the people than belonging to a particular language group. This is also the case for people from the Hozo clan, who may speak Hozo in one place and Seze in the other (cf. appendix 2). These findings lead to a couple of additional questions: Do the remains of Koman Mao speakers called Kuro in Shagga on one side and the Omotic speaking Kuro in Shera Kama on the other side indicate that t here is an on-going language change from Koman Mao to Omotic Mao in this area? Or is the opposite the case – that the Kuro in the north-east of the survey area as well as the Kur in Ego Girmos, originating from Kabache, spoke an Omotic language while today they are Koman Mao speakers? If some radical language change has been going on in recent times in this area, could this so help us to explain the origin and the migration of the Mao? Could this give some answers to the question why some groups who speak a Koman language closely related Gwama today consider themselves to be ‘Mao’ and look at Seze or Hozo speaking clan as the same people, but who do not look at Gwama people from the lowlands this way?

The enigmatic term Kuro at the interface between “Mao and Mao” would be an interesting starting point for f urther studies, preferably together with historians and anthropologists.

4.1.5 What is Kwama?

The use of term Kwama has caused some confusion. What does it mean? Is it a separate language or a dialect variety of Gwama or even Komo, or is it merely a synonym of Gwama?

In much of the literature such as Bender (1975; 1983); Siebert et al. (1993; 1994) and Davis et al. (2011) Kwama is used for the language or at least for a part of a dialect cluster, which is here called Gwama.^[65] In The Ethnologue (Lewis 2009) we find an entry on Kwama, where Gwama is mentioned as one of the synonyms and as the term used by Komo speakers.

All relevant entries in Encyclopaedia Aethiopica (cf. Bryant et al 2007; Bender 2007; Fleming 2007) follow this terminology. In other publications such as Hellenthal (2005), Zelalem Layew (2005) and Kievit et al. (2011), as well as in this survey, Gwama is used to refer to this language or to a part of this dialect cluster. All this indicates that Gwama and Kwama can simply be considered to be synonyms, addressing the same Koman dialects spoken in the western part of our survey area (cf. map 6). This fits with several statements made in the interviews with Gwama people who mentioned that the most correct designation of their language would be T’wa Kwama.

This opinion is confirmed by Asadik Habte from Ya’a in the transition area between highland and lowland. According to him, the Gwama speakers would use [t’^wa k^wama] or [t’^wa g^wama] for the language and [k^wama] or [g^wama] for the people - with strikingly fronted [a]’s. Whether it is pronounced with an initial [g] or [k] does not seem to be of relevance.

There is, however, other evidence that makes it difficult to believe that Kwama merely is a synonym of Gwama. In our material Kwama sometimes was noted besides Gwama, particularly in villages such as Lak’e, Shumate and Bangatarko with a mixed Gwama and Komo speaking population. Here it was reported that there are people who speak Gwama while there are others who speak Kwama.

From Yiwa, most likely a purely Komo speaking village west of Yangu^[66], we received the notice that the people call themselves Komo while others call them or their language Kwama. This again fits with what is explained in Kievit et al. (2011:13)^[67] and what is mentioned by Theis (1995). Otero (p.c.) and Hellenthal (p.c.) suggest that T’wa Kwama might be the term Gwama speakers use when they refer to the Komo language. Accordingly, Kwama here could be considered to be the Gwama word that is used when referring to the Komo language.

This opinion was confirmed in interviews in the villages around Penshuba, a multilingual area in Mao Komo Special Woreda (cf. appendix 2 and map 2). Here the Gwama speakers clearly state that there are people who speak a language that is not understood by them - the same languages that is spoken in Yangu and Lak’e. Gwama speakers call it Kwama Dini. It is obvious that they here refer to Komo. So Kwama Dini and T’wa Kwama Dini is at least in this area the Gwama designation for the Komo people and their language. The Komo of the area would call them themselves [kɒm] and there language [t’a kɒm]^[68]; [g^wama] and [t’a g^wama] is the Komo designation for Gwama.^[69] Krell (2011:11) presents a similar suggestion:

In summary we can say that Kwama may either - as frequently used in literature – be a synonym for Gwama or the Gwama word for the Komo language.^[70] There is no evidence proving the existence of a separate Kwama language or even a Kwama dialect of either Gwama or Komo. In order to avoid misunderstandings and terminological confusion, I would recommend sticking to the terms Gwama and Komo both for the people and for their language.

4.2 The Omotic Mao languages

4.2.1 Designations and relationships

The scientific discussion with regard to the genetic position of the Mao languages is thoroughly documented in Ahland (2012:13-18). While it is generally accepted today that Omotic is one of the primary branches of the Afro-Asiatic family, the position of the four languages Hozo, Seze^[71], Ganza and Northern Mao is still being discussed. Ahland, (2012:15f) refers to Bender (2003) who suggests a 2-way split from Proto-Omotic into Mao and all other Omotic languages^[72], and to Hayward (2000) who classifies the Mao languages as one member of a 3-way split from Proto-Omotic into Mao, South Omotic and North Omotic.^[73] Ahland further refers to Bender (2000) who states that the Mao languages are the most divergent in Omotic and supports a 2-way split solution lumping Omotic Mao in one group and all other Omotic languages in another branch. Finally, he concludes:

This discussion regarding the genetic classification of Mao only addresses the Omotic Mao languages. We cannot overlook that a substantial number of speakers in the Begi-Tongo area at least in certain circumstances call themselves and their Koman language Mao, and for the Oromo, the entire indigenous population of the area indiscriminately speaks what is referred to as “Afaan Mao”.

This survey has no intention of interfering with the discussion about the internal classification of the Omotic languages. It merely intends to suggest the use of terms that are as clear and unambiguous as possible. I would therefore like to make the following statements:

(1) We do not use the term Mao as an linguistic term unless it is well defined and specified in detail through its combination with another term.

As shown in figure 2, Mao is used as a designation for a Koman language – often synonym to T’wa Sit Shwala but different from Gwama (cf. 4.1.2). The term “Begi Mao” is ambiguous and misleading.

(2) Seze, Hozo, Northern Mao and Ganza are Omotic languages and may be called Omotic Mao.^[74] Anfillo is a Northern Omotic Kefoid language, not belonging to the Mao branch of the Omotic languages though its speakers call themselves and their language Mao.

This is regardless of whether Omotic Mao is classified as an independent second or third branch of the Omotic languages (Bender 2003; Hayward 2000) or as a branch of the North Omotic languages. The degree of affiliation between Ganza and this group is not yet clear (cf. 4.2.4). Northern Mao here designates the Omotic language, which is spoken around Bambasi and in the Didessa valley and may form the “eastern branch” of the Omotic Mao languages.^[75] The closely related languages Hozo and Seze may so form the “western branch” of this language group. There is no common name for these two languages/dialects; here I will use Seze/Hozo even if the status of Hozo and related dialects is not completely clear yet, or just Omotic Mao – well knowing that this term normally also includes Northern Mao and Ganza.

4.2.1 The Seze – Hozo language area

Seze, Hozo and closely related dialects are spoken in the eastern part of the survey area with their centre in K’ondala Woreda. There are also recordings from villages with Seze and Hozo speakers east and north of the Dabus River in the western part of Babo Gambel Woreda and the southernmost part of Mana Sibu Woreda (cf. maps 1 and 2). In the eastern part of Begi Woreda and along the road from Begi to Bambasi there are villages with at least Seze or Hozo minorities. It is not known if there are villages with speakers of Hozo, Seze or related dialects in the northern part of Jima Horro Woreda, but that would not be unlikely. ^[76]

The Omotic Mao languages Seze and Hozo form a compact area in Oromia Regional State surrounded by Oromo speakers on three sides (cf. map 4). Today there may be a majority of Oromo speakers within the whole area. These as well as immigrant groups from southern and north-eastern Ethiopia live mainly along the Begi –K’ondala – Babo Gambel road, while there are still villages with almost exclusively Omotic Mao speaking inhabitants in the remote parts of the Guma Gara Arba mountains and in the Dabus Swamps.

In the West close to the road from Gidami via Begi to Bambasi there may be some overlap between the Omotic Mao and Koman Mao areas. In some of the mainly Koman Mao speaking villages such as Kobor Chandi and Kongilo Gara Kelo, there are reported Hozo or Seze minorities (cf. appendix 3), as well as in the Girmos-area south of the Begi – Tongo road. Investigations in this area have shown that those “minorities” in some cases are women moved there by marriage, something that again confirms the close relationship between the Omotic Mao and the Koman Mao with a common ethnic Mao identity in spite of the languages differences. Regarding the role of the Kuro clan and the obviously bilingual area around Shera Kama cf. 4.1.4.^[77]

As we can see from this figure, Seze is spoken in the central and northern part of this Omotic Mao area, while Hozo is mainly spoken in the south and southwest (cf. also map 6 and ap-pendix 2). Shuluyo here considered closely related to Hozo or as a part of a Hozo dialect cluster is reported as the only Omotic Mao language in the east. In the southern and western part both Seze and Hozo speakers are reported from most of the villages. This may amongst other reasons be because of the existence of minorities due to intermarriage or the confusion arising from the fact that a significant portion of the members of the Hozo clan speaks Seze. The H behind the name of villages on figure 3 indicates that the majority of the speakers in this village belong to the Hozo clan – regardless if they speak Seze or Hozo. The K indicates that there is reported a “Kuro speaking minority”.

The only occurrence of Seze/Hozo in Mao Komo Special Woreda seems to be in the village of Mimi Akobo (cf. appendix 2 and map 6; not included in figure 3). The interview with Atherpha Dina (10/2012) confirms the findings of survey group 2 that there is a significant Omotic speaking population in this Gwama/Koman Mao speaking village.^[79]The word list recording shows that these people speak Seze without any obvious differences to other Seze speakers. The Seze speakers in Mimi Akobo belong to the typical Omotic Mao clan of Madego and call their language Dashen or Fargashi. They claim that they came from Hofa Fargashi in K’ondala Woreda down to the lowlands looking for fertile land. Curiously the people in Hofa Fargashi all seem to be Hozo speakers and belong to the Gadatso, a mainly Hozo speaking clan (cf. appendix 2).

4.2.2 Seze – Hozo – Shuluyo

There seems to be some disagreement amongst the scientists^[80] as well as amongst the speakers in how far Seze and Hozo can be classified as two different languages or as dialects of the same language. Furthermore, in the interviews and wordlist recordings, three language self-designations occur; besides Hozo and Seze there is a considerable group of speakers who consistently call their language Shuluyo. When discussing the question how many different Omotic Mao languages are spoken in the area with the speakers, I got the following statements:

Those interview statements may give the impression that there is only one Omotic Mao language in this area, and that Seze, Hozo and Shuluyo are just different varieties of the same language. Hika and Ramadan would even establish a hierarchy between these “dialects”:

There may be several reasons for this statement. Seze has a well defined distribution area with a centre in Guma Gara Arba K’ebele. This area in K’ondala Woreda on both sides of the road to Babo Gambel still seems to have a rather homogenous and vivid Seze speaking community that is well aware and conscious of their language. Another reason may be that Seze as a linguistic and ethnic term is well known and established, and is used indiscriminately by all speakers for their language. Dialect differences in Seze are not reported.

This is not the case for Hozo and Shuluyo. The Hozo area appears to be scattered, and there are only a few villages where Hozo is reported as the only Omotic Mao language (cf. figure 3). Additionally, Hozo is used as a clan name for both Hozo and Seze speaking people in the area. The main difference, however, seems to be that Hozo is by far as known and generally accepted as Seze. In our interviews in the south-eastern and eastern part of the area, in Boji Gara Arba k’ebele^[83], in the remote Guma Gara Arba Mountains, in the Dabus Swamps, on the eastern bank of the Dabus River and in the swamps on the K’ondala side, we recorded word lists, which were more or less identical with the Hozo recordings in the south-western part of the area and with the findings for Hozo in the literature (cf. appendix 3). The speakers, however, did not accept the term Hozo for the people or for the language. In these remote areas they only use Mao Shuluyo. This is also the term that Seze speakers in the more central areas of Guma Gara Arba use to refer to them. When the Shuluyo speakers mention villages where they speak “the same language” (cf. appendix 1, question 10) they only refer to villages in the close surroundings. There are no recordings of connections to the Hozo language that is spoken in e.g. Hofa Fargashi in the south-west of the Omotic Mao area (cf. map 6; appendix 2).

The informants on the eastern side of the Dabus River, in Bishawo Dabus and Malka Ebicha did not even have a specific name for their language. They called themselves and their language [mɔ:]^[84], ‘Mao’, and supposed that their language was not spoken any other place. However, when we talked about marriage they did admit that they took wives from Boji Gara Arba on the K’ondala side “where they speak the same language”. Any relationship to the language or clan called Hozo was denied.

The linguistic facts, however, show a different picture. The cognates between Seze and Hozo in my material are not more than 70 %, while I could not find any systematic differences between what was called Hozo and Shuluyo by the speakers.^[85] In table 4 a few examples of systematic sound contrast with common roots are shown:

In about 30 % of our –very limited - material, Seze and Hozo/Shuluyo show different roots.

These findings correspond well with the data in Bender (2007:85): “The lexical resemblance between Hozo and Sezo is about 65 % of basic lexicon, implying a low degree of mutual intelligibility”

In contrast to what was previously stated, there are a couple of reports that would deny that Hozo and Seze are mutually intelligible, as the informants from the very remote place Tullu Berc’uma prove. These Shuluyo speakers clearly assured that they prefer Oromo when communicating with Seze speakers. They admitted that there were many similar words, but the mutual intelligibility was not guaranteed or at least very arduous.

These findings may lead to the conclusion that Seze and Hozo/Shuluyo should be considered two different languages rather than a dialect cluster. The statements mentioned above indicating the opposite can thus be taken as wishful thinking, as expressions of an ideal “Mao unity” in a situation where all Mao languages, at least in Oromia region, are endangered and extremely pressed by Oromo.

With regard to Shuluyo, our material does not give any evidence for a distinguished third Omotic Mao language in the area. There are only few differences between Hozo and Shuluyo, which might even be due to different words randomly used for the same concept or interference from Seze or Oromo. Hozo and Shuluyo may thus be considered to be the same language (cf. appendix 3). However, since the speakers in the north-eastern part of the survey area – on both sides of the Dabus River – do not accept the term Hozo for their language and consequently use Mao Shuluyo as the linguistic self-designation, we may face a terminological problem in case of a development and further description of this language.

4.2.3 The Ganza mystery

According to The Ethnologue (Lewis 2009), Ganza is a language spoken in the Yabus area in Sudan belonging to the same branch of the Omotic languages as Hozo, Seze and Northern Mao. However, geographically it is rather distant and isolated from the three other Omotic Mao languages. Ganza is supposed to be closer related to Hozo^[86] or to the Seze/Hozo cluster than Northern Mao.^[87] These statements are not substantiated and may be taken only as assumptions until the degree of relatedness between Ganza and the other Omotic Mao languages is investigated more closely.

Ahland (2010:2) questions the idea of Fleming (1986) and Bender (2003) that Ganza only is spoken in the Yabus area of Sudan.^[88] He refers to speakers of Northern Mao who suppose that Ganza speakers also live on the Ethiopian side of the border west of Bambasi town, and that their language is somewhat similar to their own language.

The Ganza in Penshuba and Ya’a are supposed to come from the Duga Bele Mountain in Sudan; they did not live in Ethiopia at the time of Almahdi. ^[93]

The Ganza in Penshuba and Ya’a Baldigis call their language Gwami; Ganza as a language name is only used by others. The self-designation of the people seems to be Ganza (Ganzo), while Gwami is limited to the language. They are divided into several clans as e.g. Ganza deshis, Ganza dokanuku, Ganza puní/ɸuní, Ganza walagimbu and Ganza gwunzi.

All Ganza in Penshuba and Ya’a Baldigis seem to be bilingual and use Gwama/Koman Mao as lingua franca. There is no or only limited knowledge in Komo, Oromo and Amharic, but they may know some Berta. The Ganza community occurs rather closed where Ganza only intermarry with Ganza and children do not attend any school. This contributes to the conservation of the language in spite of the small number of speakers.^[94] An exchange with Ganza from Yabbus area in Sudan is not reported. On a social level the Ganza seem to have lower status compared to the Gwama, and use their own language only inside their houses or in other ‘safe’ settings.

On the basis of these interviews, the two villages with Ganza speakers in the survey area are marked with G on map 6; the green-yellow colour indicates Omotic Mao.

Very little linguistic research has been done on Ganza, and only based on my limited material the relationship between Ganza and other Omotic languages such as Seze/Hozo or Northern Mao is impossible to state with certainty.^[95] An analysis of the recorded word list (cf. appendix 3) shows – with all reservation – the following cognates: 39 % with Seze/Hozo, 11 % with Gwama and 5.5 % with Komo.^[96] For 44.5 % no clear cognates with any of those languages could be found.

My tentative findings support the classification of Ganza as an the Omotic Mao language. In the column of comments, a couple of observations are listed which might indicate regular differences –like [w]/[b] in Seze/Hozo vs. [p]/[ɸ] in Ganza, or [-a]/[-ɔ] endings in Ganza where Seze/Hozo has [-i]/[-]. It would be particularly interesting to compare the 1700 items list recorded by David Ford (unpublished) with a Northern Mao word list for assessing the family relationship between those two languages. Generally we need more linguistic research on Ganza to able to make any final statements.

In summary my investigation has revealed that there are around 200 - 300 Ganza speakers in Mao Komo Special Woreda in Ethiopia. These Ganza live in villages with a majority of Gwama speakers and are aware of their descent from the Yabus area in Sudan, where the majority of Ganza speakers live. The wordlist shows a connection to Seze/Hozo that indicates that Ganza together with Seze, Hozo and Northern Mao belongs to the Omotic Mao languages.^[97] The relationship to Seze and Hozo and probably also Northern Mao is, however, beyond mutual intelligibility.

5. The status of the “Mao – Komo languages”

5.1 Number of speakers

It is difficult to give dependable statistics of the number of speakers of each of the “Mao - Komo languages” in the survey area. One of the reasons for this is the general lack of reliable data with regard to inhabitants and language use. The arbitrary and confusing ethnic and linguistic designations are another challenge. Finally, there is probably a considerable number of bilingual people in this area, many of whom are about to lose the language of their ancestors and acquire a new mother tongue, and who would classify themselves differently depending on the context in which and by whom the data was collected.^[98]

The Ethiopian census only uses the categories Mao and Komo. According to the census of 2007, there are 46,026 ethnic Mao in Ethiopia; 33,683 of them state that Mao is their mother tongue. There is no indication of the distribution of these Mao speakers to the different language groups.^[99]

With regard to the number of speakers of the single languages, relevant literature mainly refers to The Ethnologue (Lewis 2009), which provides the following figures: Seze 3,000; Hozo 3,000; Bambasi Mao 5,000; Gwama 15,000; Komo 1,500 (Ethiopia) and Ganza 5,400 (Sudan). Some of these figures are controversial. Ahland (2012:13) assesses that the number of Northern Mao speakers (Bambasi and Didessa area) may not exceed 2,000 – 3,000; according to Krell (2011:11) the number of Ganza speakers in Sudan may only be around 2,600. With regard to Gwama, Kievit et al. (2011:11) suppose that the number of 15,000 speakers is too high. They argue that the total population of Mao Komo Special Woreda is put to 42,050 (2007 census) and estimate that only 5 % of them are Gwama speakers, i.e. little more than 2000 people. However, in this calculation they have not taken the Koman Mao speakers in eastern Mao Komo Special Woreda and in Begi Woreda in Oromia Regional State into account. Many of them may either be counted as Gwama in The Ethnologue or as Mao in the Ethiopian census.

My own material obtained through the interviews is not very consistent either. The difference between “households” and “people” (cf. appendix 1, questions 8 and 9) was frequently misunderstood. Furthermore, the questionnaire did not differentiate between people with full mother tongue competence and others who only have a basic knowledge of the language of their ancestors. But even if the numbers resulting from the addition of the tentative figures given in the interviews should be notably too high, we can gain interesting insights by e.g. looking at the mutual size ratio of the languages regarding the number of speakers.

For the Omotic Mao we got the following tentative figures: Seze: around 13,000 speakers; Hozo (including Shuluyo): around 6,000 speakers. These figures reflect the impression we got during the interviews - that Seze is the dominating language with a solid and compact language area^[100], while the Hozo area is more mixed with Seze speakers in the south and south-west, and scattered in the remote mountains and swamps in the south-eastern and eastern part. On the basis of our information, the number of Ganza speakers in Ethiopia is estimated to be less than 500 people. Together with the estimation for Sudan mentioned in Krell (2011), there may not be more than 3,000 Ganza speakers in both countries together.

For Koman Mao and Gwama the situation is more complicated and there is much room for interpretation. If we add all entries for Koman Mao (Afaan Mao, T’ wa Sit Shwala; T’wa Kirin, T’wa Kuro), we come to the impressive figure of 16,000 speakers. The number of Gwama speakers obtained by this method amounts to 8,700.^[101] Together there would be nearly 25,000 speakers of this dialect cluster. Even if this estimation of Koman Mao speakers is significantly too high and many people who have already abandoned this language in favour of the Oromo language have been counted,^[102] it is obvious that the Koman Mao speakers are a non-negligible group. It is not really surprising that there are at least historically more Koman Mao speakers in the densely populated highlands around Begi and Tongo than “proper Gwama” in the vast and sparsely populated lowland. Still, the average Gwama speaker surely has considerably better language skills and a higher linguistic self-confidence than most of the Koman Mao speakers who are linguistically and culturally highly influenced by the dominating Oromo surroundings.

On the basis of my material, it is not possible to give any reliable estimation of the number of Komo speakers in the survey area. Komo is in all places mentioned as a minority language with Gwama as the dominating language. The number of Komo speakers in Ethiopia may not exceed a few thousand. The situation in Sudan seems to be similar where Krell (2011:11) does not report a particular number of speakers, but we get the impression that there are not more than a few thousand either. For Gambela there are rather different estimations regarding the number of Komo speakers, but even here they do not exceed 1,000 – 2,000 people. So we would not be far off if we assume that, all together, there are no more than 5,000 to 10,000 Komo speakers – probably with a majority in Sudan.^[103]

5.2 Degrees of Endangerment and Development

Even if we add the highest estimates with regard to the number of speakers for all 5 languages subject to this survey, their number will not exceed 50,000 – 0,06 % of the Ethiopian population of at least 90 million; the number of speakers with full language competence will still be significantly lower. What makes the situation worse is that all Mao and Komo people are highly marginalised and underprivileged with low self-esteem and with less education, influence and access to participation in the civil society than the members of the dominating people groups in the area such as the Oromo.^[104]

In the field interviews, most of the interviewees drew a picture of languages degraded to “house languages” that are highly endangered.

There are only few statements that indicate the opposite, some from Gwama speakers in the lowlands of Mao Komo Special Woreda, and a few from Seze speakers from Guma Gara Arba in K’ondala Woreda.

It goes beyond the capacity of this survey to make profound statements on the degree of endangerment of the five languages subject to this survey – e.g. as Ahland (2012)^[106] is able to do for the Northern Mao around Bambasi and in the Didessa Valley. He analyses the situation for these language according to the indicators for measuring the degree of endangerment of languages in the report of UNESCO’s Ad Hoc Group on Endangered Languages (2003) and concludes clearly that “the Northern Mao language is endangered.” (Ahland 2012:31):

Through my field research and discussions with native speakers I got a similar impression of the sociolinguistic situation for the indigenous languages within the survey area. Schools seem to play an important role in the negative development of these small languages – at least in Oromia, where Oromo as the only school language reinforces the dominating role of this language in the society. The same can be said for at least the protestant churches where only Oromo is used as a medium of reading, preaching and instruction.

The Omotic Mao languages Seze and Hozo are exclusively spoken in Oromia Regional State. In Oromia, Oromo is the only medium of instruction and the only language used in public, and it is the only language recognised in the constitution of the regional state. As long as this situation continues, the negative development for Seze and Hozo probably will continue. As far as I know, there are no plans to develop neither Seze nor Hozo to be used as languages of instruction or even to be recognised as existing languages.

What makes it even more difficult is that Seze and Hozo/Shuluyo are not easily mutually understandable. Speakers of these languages live in the same area, often together in the same villages, but Oromo most often functions as a means of communication and takes a more and more important role as the home language of the younger generation. Younger speakers with full mother tongue competence are therefore often found only in very remote areas.

Up to now, very little linguistic research and data collection has been done on these two languages^[107], no orthography has been developed and no script decision has been made. SIL Ethiopia supports some linguistic work on Seze and Hozo, and starting from 2014 a couple of workshops on these languages will be arranged within a Mao Komo language development project. Through this more clarity on the current sociolinguistic situation and the degree of language erosion amongst young speakers is expected. It can be hoped that it is not too late, and that in the near future, small texts in Seze and Hozo can be developed for informal use.

The situation for Koman Mao/Gwama is even more complicated. The Gwama speaking area is mainly limited to the lowlands in Benishangul Gumuz Regional State. It is a compact area with rather little influence from outside and where Gwama even plays the role as lingua franca in communication with speakers of even smaller languages such as Komo, Ganza and Uduk. Most Gwama speakers in this area seem to have a well-developed awareness of their language, and you will easily find young people with fully developed language competence.

Gwama is included in a multilingual education programme run by Benishangul Gumuz Regional State and facilitated by SIL, and it is on the way to become a medium of instruction in at least some pilot schools in the lowlands in Mao Komo Special Woreda. A script decision has been made, and the last decision on the orthography is imminent. First texts in the preliminary Gwama orthography are developed will be printed in 2014. More linguistic research will be done, and a comprehensive grammar and a dictionary will hopefully soon be developed.^[108]

There seems to be an interest in developing this language both from the regional government’s side and from the side of the speakers around Zebsher, Keser, Yangu and Lake in Mao Komo Special Woreda. What obviously makes a difference is that Amharic functions as the language of instruction in Benishangul Gumuz Regional sate, a fact that lessens the pressure from the side of Oromo. It will remain interesting to see in how far those who classify themselves as Sit Shwala (Koman Mao) and would not accept their language to be called Gwama can benefit from this development programme in Benishangul Gumuz Regional State.

The situation for the Koman Mao speakers is much more uncertain; this people group seems to be neglected or overseen since they do not belong to the Omotic Mao, nor are they included in the development of Gwama. Whether Koman Mao speakers around Tongo will profit from the development of Gwama in Benishangul Gumuz Regional State will probably depend on the question if the “Maos” accept Gwama as their language – and if they are accepted by “proper Gwama speakers”. In Oromia, where the majority of the Koman Mao speakers live, there will barely be efforts to introduce Koman Mao as a school language or in public life, and there is fear that this language will be crowded out by Oromo within a relatively short period of time. The workshops in 2013 have clearly shown that it already now is almost impossible to find young and literate Koman Mao speakers with fully developed language competence. The majority of the young Sit Shwala from Begi Woreda only have basic skills in Koman Mao and prefer to use Oromo in daily conversations, even with each other.^[109] This does not mean that these Koman Mao speakers are not interested in the use and development of their language. In contrary, the loss of their mother tongue on one hand and the lack of acceptance in the Oromo society on the other hand, seem to make it even more difficult to find a positive cultural identity. This increases their interest in the language of the ancestors – particularly for the active young people who suffer from the loss of their mother tongue. To actively involve Koman Mao speakers in the development of the Gwama orthography and literature and to stimulate the use of Mao Koman orally and in writing in the non-state sector seem to be the only option to counteract the negative development.

Here it is very difficult to say something about the degree of endangerment of Komo Since . Komo is an extreme minority language throughout the area, a high degree of vulnerability is likely. On the other hand, Komo is exclusively spoken in the Gwama language area in Benishangul-Gumuz, and there it is far less exposed to the pressure from the side of Oromo than e.g. Koman Mao or Mao Omotic in Oromia Regional State. Interestingly Komo, not Gwamis one of the languages mentioned in the constitution of Benishangul-Gumuz regional state, what gives this language group huge prestige and a far greater level of visibility than the small number of speakers would suggest.

Komo is part of the same governmental development programme for multilingual education in Benishangul Gumuz Regional State as Gwama, and has even come farther. The orthography is decided and first writer’s workshops are accomplished; there is on-going linguistic research on The problem for Komo may be that the language area is extremely scattered with few if any areas where Komo is spoken as the only language, that most of the Komo speakers are bilingual in Gwama or in another local language and that the number of native speakers is very small in all areas – no good conditions for the use of Komo in schools and in other public fora. Komo.^[110]

My field research has provided evidence that there is a vibrant Ganza language community in Ethiopia. Here Ganza is spoken as an extreme minority language in some villages in Benishangul Gumuz. The majority of the Ganza speakers may live in South Sudan, but the total number of speakers is small. Research or development activities on Ganza are not known, and it is not likely that there will be any development programme for Ganza as a medium of instruction in schools in Ethiopia.

Surprisingly the information both from Ethiopia and from Sudan indicate that the Ganza speaking community is stable and that the children still show fluency in this language.^[111] The reason for this may be the relative remoteness of all Ganza speakers, their limited exchanges with other groups in the society and that their children do not join any schools. With only a few hundred speakers within a changing multilingual society, the survival of the Ganza language will anyway be doubtful.

5.3 Summary and Outlook

In conclusion, we can state that all the information gathered in this survey indicates that all Mao-Komo languages are endangered. Koman Mao already seems to be highly reduced and displaced by Oromo as the native language for the younger generation. Gwama in the lowlands of Mao Komo Special Woreda appears to have a much better chance for surviving as a mother tongue. Since Koman Mao and Gwama are closely related dialects of the same language and Gwama is subject to a governmental language development project and it can be hoped that the development of Gwama may have a positive effect on the survival of Koman Mao.

The risk status of Seze and Hozo is difficult to assess, and more sociolinguistic studies are needed. Since these languages are exclusively spoken in Oromia and the local population, e.g. through schools and churches, is exposed to a severe linguistic assimilation in favour of Oromo, it can be taken as granted that these languages are highly endangered and already declining both quantitatively and qualitatively. No official development programmes are planned, but there are a few attempts through private organisations and churches to implement development tasks such as recordings of music and poems and printing of traditional stories on an informal level.

Ganza and Komo are only spoken by very small groups that are exclusively minorities in their respective settlements. The relatively good survival of these languages may amongst other reasons be related to the lack of formal education of its speakers. However, small changes may have large impacts and may quickly bring these languages to the brink of extinction if appropriate measures are not taken. Ganza is by far the least explored indigenous language of the area. It will, however, be included in a research project by SIL and should be given priority for further research. The development and research work of Komo is on-going and currently, much attention is given to this language. Whether the use of Komo as a language of instruction will be successful or even possible since the number of Komo students will always be a minority within the Gwama majority, remains to be seen. In the worst case, the lack of skilled teachers and the unclear use of ethnic and linguistic designations could lead to increased confusion.

References

Ahmad, Abdussamad 1999. Trading in slaves in Bela-Shangul and Gumuz, Ethiopia: border enclaves in history, 1897-1938. The Journal of African History No. 40, 10/01

Ahland, Colleen. 2010. The classification of Gumuz and Koman languages. Paper presented at the Workshop on African Isolates, Lyon, France, Universite Lumiere, Dec. 3–4, 2010. Manuscript.

Ahland, Michael Bryan. 2012. A grammar of Northern Mao (Màwés Aas’è). A Dissertation Presented to the Department of Linguistics and the Graduate School of the University of Oregon.

Bender, M. Lionel. 1975. The beginnings of ethnohistory in Western Wellegga: The Mao problem. Patterns in language, culture, and society in Sub-Saharan Africa: Proceedings of the sixth African linguistics conference, ed. by Robert K. Herbert, Ohio State University Working Papers in Linguistics 19:125–141. Columbus, OH: Department of Linguistics, Ohio State University.

Bender, M. Lionel. 1983a. Proto-Koman phonology and lexicon. Afrika und Uebersee 66.259-297.

Bender, M. Lionel. 1983b. (ed.) Nilo-Saharan Language Studies (Monograph 13, Committee on Northeast African Studies), East Lansing, MI 1983

Bender, M. Lionel. 1997. The Nilo-Saharan languages: a comparative essay. 2nd edition. Munich: Lincom Europa.

Bender, M. Lionel. 2000. Comparative morphology of the Omotic languages. Müncher: Lincom Europa.

Bender, M. Lionel. 2003. Omotic Lexicon and Phonology, Carbondale, IL 2003 (Published by the author).

Bender, M. Lionel. 2007a. Hozo-Sezo. Siegbert Uhlig (ed.). Encyclopaedia Aethiopica. 3.85-86. Wiesbaden: Harrassowitz Verlag.

Bender, M. Lionel. 2007b. Koman. Siegbert Uhlig (ed.). Encyclopaedia Aethiopica. 3.416-418. Wiesbaden: Harrassowitz Verlag.

Bryant, Michael and Ralph Siebert. 2007. Kwama. Uhlig, Siegbert (ed.). Encyclopaedia Aethiopica 3.460-462. Wiesbaden: Harrassowitz Verlag.

Davis, Angela, Hussein Mohammed, Linda Jordan. 2011. A Sociolinguistic Survey Report of the Northern Mao, Kwama, and Komo Speech Varieties of Western Ethiopia. SIL Electronic Survey Reports, 2011-043. Dallas: SIL International. Online: http://www.sil.org/silesr/abstract.asp?ref=2011-043.

Federal Democratic Republic of Ethiopia Population Census Commission. 2008. Summary and statistical report of the 2007 population and housing census: Population size by age and sex. Addis Ababa. Online: http://www.csa.gov.et/pdf/Cen2007_firstdraft.pdf

Fleming, Harold C. 1976. Omotic overview. Marvin Lionel Bender. (ed.). The non-Semitic languages of Ethiopia , 299-323. African Studies Center, Michigan State University.

Fleming, Harold C. 1986. Mao's ancestor: consonant phonemes of proto-Mao stage one. Proceedings of the Ninth International Congress of Ethiopian Studies, 5. Moscow, 26-29 August 1986.

Fleming, Harold C. 2007. Mao. Uhlig, Siegbert (ed.). Encyclopaedia Aethiopica. Vol 3:755-756 Wiesbaden: Harrasowitz Verlag.

Ford,David. 2013. Ganza Word List (CWL 1700 words). SIL Ethiopia, Addis Ababa, October 2013 (unpublished).

González-Ruibal, Alfredo. 2014. Archeology of Resistance: Materiality and Time in an African Borderland. Rowman & Littlefield, Lanham, Maryland USA

Greenberg, Joseph .1966. The Languages of Africa. Indiana University Press.

Grottanelli, Vinigi L. 1940. I Mao . Missione etnografica nel Uollega occidentale Vol. 1, Roma: Reale Accademia d'Italia.

Hayward, Richard J. 2000. Omotic: the "empty quarter" of Afroasiatic linguistics. Jacqueline Lecarme. (ed.).Research in Afroasiatic Grammar II: selected papers from the fifth conference on Afroasiatic languages, Paris 2000, 241-261. Amsterdam: John Benjamins

Hellenthal, Anne-Christie. 2005. Notes on Mao. Addis Ababa: SIL Ethiopia. Unpublished ms.

Hofmeister, Lorianne. 2010. Ganza language learning manual. Sudan. SIM-Sudan. MS

James, Wendy.1979. ‘Kwanim Pa’: The making of the Uduk people. Clarendon Press, Oxford.

James, Wendy. 1981. Ethnic Terms and Ambiguities on the Sudan-Ethiopian Border. Journal of Research on North East Africa 1/1, 16–31

Kievit, Dirk and Erika Robertson. 2011. Notes on Gwama grammar with texts and word lists. SIL Ethiopia.

Krell, Amy. 2011. A Sociolinguistic Survey of the Ganza, Komo, and “Baruun be Magtole” Language Groups (Blue Nile Province, Sudan). SIL Electronic Survey Reports, 2011-039. Dallas: SILInternational. Online: http://www.sil.org/silesr/2011/silesr2011-039.pdf.

Lewis, M. Paul. (ed.) 2009. The Ethnologue. Languages of the world. Sixteenth edition. Dallas:SIL International. Online: http://www.ethnologue.com/

Moges Yigezu. 1995. Dying twice: the case of Anfillo languages. Afrikanistische Arbeitspapiere 43. Cologne: University of Cologne, Institute of African Studies.

Neudorf, Andreas. 2011. Portfolio on Mao-Komo in Benishangul Gumuz Regional State Ethiopia; MA Literacy Programme Development; SIL Ethiopia, unpublished.

Regional Government of Benishangul Gumuz. 2002. The Revised Constitution of the Benishangul Gumuz Regional State. 8th edition. Assosa: Government of Benishangul Gumuz Regional State.

Siebert, Kati, Ralph Siebert and Klaus Wedekind. 1993. Survey on languages of the Asosa-Begi-Komosha area. S.L.L.E. Linguistic Report no. 11. Addis Ababa: Institute of Ethiopian Studies, Addis Ababa University.

Siebert, Kati, Ralph Siebert and Klaus Wedekind. 2002. Sociolinguistic survey report on languages of the Asosa, Begi, Komosha area: Part I. SIL Electronic Survey Reports 2002-054. Online at: http://www.sil.org/silesr/2002/SILESR2002-054.pdf

Siebert, Ralph, Charlotte Wedekind and Klaus Wedekind. 1994. Third S.L.L.E. survey on languages of the Begi/Asosa area. S.L.L.E. Linguistic Report no. 15. Addis Ababa: Institute of Ethiopian Studies, Addis Ababa University.

Siebert, Ralph, Charlotte Wedekind and Klaus Wedekind. 2002. Third S.L.L.E. survey on languages of the Begi/Asosa area. SIL Electronic Survey Reports 2002-056. Online: http://www.sil.org/silesr/2002/SILESR2002-056.pdf

Smidt, Wolbert. 2007. Mao, ethnography. Uhlig, Siegbert. (ed.). Encyclopaedia Aethiopica. 3.756-760 Wiesbaden: Harrasowitz Verlag.

Theis, Joachim. 1995. Nach der Razzia. Ethnographie und Geschichte der Koma. München: Trickster Verlag.

UNESCO’s Ad Hoc Expert Group on Endangered Languages. 2003. Language Vitality and Endangerment. Paris: UNESCO, MS. Online: http://www.unesco.org/culture/ich/doc/src/00120-EN.pdf

Wedekind, Charlotte and Klaus Wedekind. 1993. Survey on languages of the Asosa-Begi-Komosha area part II. S.L.L.E. Linguistic Report no. 12. Addis Ababa: Institute of Ethiopian Studies, Addis Ababa University.

Wedekind, Charlotte and Klaus Wedekind. 2002. Sociolinguistic survey report on languages of the Asosa, Begi, Komosha area: Part II. SIL Electronic Survey Reports 2002-055. Online: http://www.sil.org/silesr/2002/SILESR2002-055.pdf

Zelealem Leyew. 2005. Gwama: A little known endangered language of Ethiopia: A sketch of its grammar and lexicon. Addis Ababa: Addis Ababa University. Online: http://www.uni-koeln.de/gbs/images/guama.pdf (accessed 3/12/11).

Appendix 1

1. General Information

2. Questions about the place, the peoples and the languages

Appendix 2

Villages covered by this survey – organised on the basis of language identification on the word lists^[112]

The following tables give an overview over the villages investigated in this survey. One of the interesting findings of the interviews and the word lists^[113] is the fact that there are very few villages where speakers of Omotic Mao and Koman languages mix, but there is a high number of villages where speakers of different Koman languages and speakers of different Omotic Mao respectively live together. Exceptions are the speakers of Ganza, an Omotic language with only few speakers who live as minorities in villages with speakers of Koman languages. In a few Koman Mao villages, Hozo and Seze speakers are reported as a minority. Mimi Akobo seems to be the only Koman Mao village with a considerable number of Seze speakers, which is due to recent immigration.

On the basis of these findings, the investigated villages are organised in two groups – villages with (almost) exclusively Koman Mao/Gwama and Komo speakers (Table 1) and villages with (almost) exclusively speakers of Omotic Mao (Seze and Hozo/Shuluyo) (Table 2). Ganza is, thus Omotic, entered into Table 1 since there is no village with a majority of Ganza speaers.

Appendix 3