Foundation for Endangered Languages

Home | Manifesto | Membership details | Proceedings | Grant Applications | Newsletter | Links | Bibliography


5. Fieldwork Reports

Field Trip to Record the Status of some little-known Nigerian Languages. Roger Blench

Between December 18th 1998 and 23rd January 1999 I made a number of field trips in Central and South-East Nigeria to obtain more data on the status of some little-known languages.

In 1992, I had published the Index of Nigerian Languages (Crozier and Blench 1992); many entries inevitably relied on very old field data, some of which has turned out to be highly inaccurate. Blench (1998) is a summary of the situation as it was known in 1997; much of that material has had to be revised. Nigeria is developing an ‘Endangered Languages’ infrastructure, as far as headed notepaper and workshops go; unfortunately this seems to have had very limited results in terms of new field data.

The following notes describe the conclusions of what were inevitably rapid visits. The lexical data was taped and is being transcribed. Papers including preliminary data on each language are available from the author as email attachments. Comments on the status of the languages are necessarily impressionistic; more detailed socio-linguistic surveys are clearly essential.


The Ningye language is spoken in Ningeshen Kurmi village, Kaduna State, Nigeria. A wordlist was collected by Roger Blench with the assistance of Selbut Longtau from the chief, Abubakar Salihu Samu and Bulus Magaji and a group of villagers in Ningeshen Kurmi on the 19th of December 1998. The Ningye people do not seem to be recorded in any previous literature (Crozier & Blench 1992). Ningye is the name of the language and the people. The language appears to be most closely related to Numana and Gwantu.

Ningeshen Kurmi is some 19 km. south of Fadan Karshi on the Akwanga road. There are three other settlements, Akwankwan, Kobin and Ningeshen Dutse, due north of Ningeshen Kurmi but these are all very small. The Ningye language is spoken by perhaps 3-4000 speakers at a maximum. Ningye is still regularly spoken at present. The Ningye are multilingual; Numana and Gwantu are the main additional languages they speak, but Hausa is widely known and some younger people also speak English.


The Cara language is spoken in Teriya village, Bassa LGA, Plateau State, Nigeria, some 5 km. west of Gurum, which is 3 km. south-west of the main Jos-Kaduna road, 11 km from Jos town. Teriya is a Hausa name, describing a series of scattered sections, ipup, of which the principal one on the road is Anjòk. The Cara people occur in the literature under the names Teriya, Tariya, Pakara, Fakara and Fachara. A wordlist was collected by Roger Blench with the assistance of Selbut Longtau from the Village Head, Peter Maguni Kusaru, the Wakili, Hamidu Taita and a group of elders including Kudaru Tanko, Culu Gado and Jinga Kunangaru on the 21st of December 1998. Cara is the name of the language and the people.

The Cara language is spoken by less than 3000 speakers at a maximum. The Cara tend to know Hausa and some younger people also speak English, but generally do not speak the languages of their neighbours. The older people have the impression that younger people are giving up the language in favour of Hausa. Certainly they do not have an easy command of the complex morphology required to be a competent speaker, although this may develop slowly.


The Bu language is spoken in four villages in Plateau State, Nigeria. A wordlist was collected by Roger Blench with the assistance of Selbut Longtau from Joshua Chaga (27 years old) and a group of elders in Nakere on the 8th of January 1999. Our thanks to the Wakili of Nakere, Gambo Nagwe, for making possible the meeting and ensuring the terms elicited were as accurate as possible.

The Bu are not referred to in primary sources but appear first in Hansford et al. (1976) as the ‘Jidda-Abu’ a group classified with Eggon, Nungu and Ake. Ibut and Nakare are given as alternative names; Ibut is clearly a version of Bu and Nakare the name of the first village reached from the road.

The villages of the Bu people are reached from Gbodu village, 18 km east of Akwanga on the road to Wamba. A road leading northwest from Gbodu reaches Nakere after 6 km. and then Rago, Maiganga and Abu. The villages of the Ningkada [=Jidda] are reached by a turnoff some 7 km. north of the Wamba junction. The road turns east at Kango and Jidda (the main settlement) is some 6 km. away. Two hamlets, Ningkada and Lago, are southwest of Jidda respectively 1 and 3 km. away. A further settlement on the main road, Wanga, is some 5 km. north of Kango.

The Bu regard their language and culture as distinct from the Ningkada [Jidda] and they are now geographically separated, but the differences between the two would appear to be slight. A complete wordlist of Ningkada was not taken but some twenty words elicited in a rapid visit suggested that the two languages are the same, with some minor lexical and phonological differences. There are perhaps 4000 speakers of Bu and about 2000 or less speakers of Ningkada (the Ningkada hamlets are extremely small). In both locations, immediate evidence suggests that both language and culture are very strong and are not immediately threatened. The classification of Bu would appear to be in error; the language closest to it is undoubtedly Ninzam, with which is shares some highly distinctive features such as suppletive plurals.


The Hasha language is spoken in three villages in Nassarawa State, Nigeria. A wordlist was collected by Roger Blench with the assistance of Selbut Longtau from a group of villagers in Bwora (Yashi Sarki) on the 7th of January 1999. The Hasha people have been generally known in the literature as ‘Yashi’ a Hausaised form of their name. Correctly, however, a single person is / haSa / and the plural is /h´haSa/; the language also appears to be called / haSa /. There seems to be no reason use any other term than Hasha, which is now what the community prefers.

There are three villages where Hasha is the main language, Hashasu (=Yashi Pa), Kusu (Yashi Madaki) and Bwora (Yashi Sarki). Bwora is the largest settlement and generally regarded as the administrative centre. Hasha villages fall within Nassarawa State. They are all east of the main road from Fadan Karshi to Wamba, reached from a turnoff some 25 km. south of Fadan Karshi.

None of the Hasha settlements are very large; the population cannot be more than ca. 3000. However, the language is still spoken by young people at present. Hausa is widely known and English is spoken by some secondary school students.


The Rukul language is spoken in Barkul village, Plateau State, Nigeria. A wordlist was collected by Roger Blench with the assistance of Selbut Longtau from a group of villagers in Barkul on the 13th of January 1999. Samuel Musa (ca. 40 years old) kindly spoke the examples on to the tape, but the forms cited are a collective product.

Existing references to the Rukul people and language appear as the Barkul element in the name ‘Mabo-Barkul’ given in Crozier & Blench (1992). The ba- element is a nominal prefix and ought properly to be dropped in a reference name. The name of the closely related Mabo people similarly incorporates an person affix which is better eliminated. The correct terminology is as follows;

One person	People	Language
ama rukul	ba rukul	kap ma rukul
ama bO	ba bO	kap ama bO

A standard reference name mirroring the existing formulation would thus be ‘Bo-Rukul’ and this has been suggested as the head-entry for the forthcoming Ethnologue 2000.

Barkul village is some 8 km. (20 mins drive) east of Richa over an extremely bad road. Mabo village is about a half-hour walk further on but cannot be reached by road. Richa is at the edge of the Jos Plateau some 2 hours drive SE of Jos and is the principal market-town for the area. All the other villages in this area speak varieties of Kulere, a Chadic language, with the exception of Horom and Mwa. The settlement pattern is fairly dispersed, but there appear to be no other hamlets speaking this language.

To judge by the visible houses, there must be between 500 and 1000 speakers of Rukul and the same number of speakers of Bo. Rukul remains the main language of communication and appears to be healthy, inasmuch as a language with so few speakers can be. Hausa is widespread and eliciting the wordlist suggested that younger speakers had a tendency to replace some items with their Hausa equivalent.

Mambila field trip

Between 19th and 23rd January 1999, Bruce Connell and I made a field trip to the Mambila Plateau to try and complete the listing of lects in Bruce’s Mambila database. A more comprehensive report will be published on the Mambila Web site managed by David Zeitlyn.

The first trip was to Zongo Ajiya in the extreme northwest of the Mambila Plateau, to visit the MvanIp people, first recorded by Meek as Magu in the 1920s. Far from being ‘less than 10,000’ as it says in Index of Nigerian Languages (1992), there are only 100 speakers (chief’s estimate) living in one quarter of Zongo Ajiya. Despite this, the language seems to be alive –the Jauro assured us that all the children still speak it, which appears to be true. A long wordlist was taped and there is no doubt that this is the same as the language given in Meek. Oral traditions of migration were recorded from a monoglot speaker of MvanIp, Mr. Yi Neman, who is now over 100 years old. Despite the small number of speakers and the evidence of fluency in vehicular Fulfulde, MvanIp is still being transmitted to children.



When we asked for the language closest to MvanIp, we were given the name of the Ndunda people (which Meek noted as speaking the same language as Magu). Ndunda is a village some 5km. from Yerimaru, past Kakara on the tea estate road, northwest of Gembu. The Ndunda are a distinct people and language whose existence seems previously to have been unrecorded. Ndunda resembles MvanIp but the two are sufficiently distinct as to be regarded as separate languages. Apart from a wordlist, we also took a limited amount of historical data. There are probably 3-400 speakers of Ndunda. It appears the language is still being transmitted to children.

We wanted to reach Antere, beyond Ndunda on the Cameroun border, as the exact language(s) spoken there are unknown. However, the road has now collapsed and we were forced to give up that project. We were able to contact Antere people in Yerimaru, and were told, much to our surprise, there are numerous languages spoken in Antere, in different quarters. These are;

Fum	 	= Mfumte
Nshi (home village Nkiri) ? = Wushi
Bùkwák people from Kwak, 
speaking Kwak ? = Yamba
Bunta people from  Nca village, 
speaking áncá ? = Manta
BItI people speaking (?) NdE!gbítE~	
VItI people		

We were able to record a wordlist of VItI and it is definitely a Grassfields language of Camerounian Bantu type –but of what type and whether it is new cannot yet be known. As for the others we guess they may also be Grassfields although there is absolutely no evidence for this except proximity. The identifications in the table above are based on the entries in the Linguistic Atlas of Cameroun. But they all seem to be spoken in Nigeria and should thus be added to the list of Nigerian languages. The numbers of speakers must be very small.

* * *

The most striking result of this survey was that in general, despite the small numbers of speakers of many languages, they continue to be spoken. This is not to say they should not be regarded as ‘endangered’ – any language with so few speakers can disappear rapidly through cultural or economic change. The main threat is that the rise of Hausa and English will tend to pidginise the languages, that younger people will no longer be able to master the complex system of plurals and tend to replace common lexical items with their Hausa equivalents.

Since the map accompanying the Index of Nigerian Languages was published in 1994 (but representing field data up to the end of 1990) a substantial new body of data has been collected on names, location and existence of languages. These have been drawn on a base map prepared by SIL, and are presently being digitised in Nairobi. This map should appear to accompany the millennium edition of the Ethnologue.


Blench, R.M. 1998. The status of the languages of Central Nigeria. In: Endangered Languages in Africa. ed. M. Brenzinger. 187-205. Köln: Rüdiger Köppe.

Crozier, D. and Blench, R.M. 1992. Index of Nigerian Languages (edition 2). Dallas: SIL.

Hansford, K. Bendor-Samuel, J. & Stanford, R. 1976. An Index of Nigerian Languages.Ghana: Summer Institute of Linguistics.

Recent Fieldwork in Ghana: Report on Dompo and a note on Mpre
Roger Blench

In April 1998 I was able to visit a community in Ghana, whose language has been uncertain until now. Indeed I was informed that Dompo was either dead or was simply a subgroup of the Gonja not speaking a distinct language. This turned out to be entirely false, and reports of the death of Dompo exaggerated. I have added a note on Mpre, a language more likely to be extinct, but which no-one has comprehensively searched for since its first appearance in the literature.


The existence of the Dompo language has been known previously only through the 100 words given in Painter (1967) and the animal names transcribed by Cansdale (1971) . It has several times been reported not to exist, or to be just an ethnonym or a submerged clan. As the present document shows, this is false.

The Dompo live in a settlement adjacent to the main town of the Nafaanra people, Banda, Brong-Ahafo Region, Ghana. Painter (1967) gives a map reference as 8° 09´N 2° 22´W. Banda is reached from Wenchi by going northwards from the main road to Bondoukou in Côte d’Ivoire and is still south of the Black Volta. A new road is being built that will eventually join Banda to Kintampo via Mo, but this is presently only passable with a four-wheel drive. The quarter of Banda where the Dompo live is called Dompofie. One or two women who married out are said to reside in nearby villages, but this is the only settlement of the Dompo.

As far could be ascertained, Dompo is the name of both the people and the language. The name ‘Ndmpo’ given in some documents seems to be a garbled version of this. The map accompanying Painter (1967) shows two very small areas near Banda marked ‘Ndmpo’ and a much larger area south of Larabanga marked NDMPO (Kwa Akan) inside the Gonja area. To judge by a slight mismatch between the list of surveyed villages (p. 31) and the data tables (p. 46 ff.) Painter probably found that only one of his sites actually spoke Dompo but subsequently failed to correct the map. This may also be the source of the idea that Dompo is only a Gonja subgroup, as this probably is the status of the larger group.

A wordlist of the Dompo language was collected by Roger Blench from Mr. Kosi Mila (ca. 55 years old) and a group of villagers on the 2nd of April 1998. The oral traditions of Dompo were narrated by the Chief, Mr. Nanas SianO, assisted by the elders. I am grateful to all who took part for their enthusiastic co-operation.

Dompo is spoken by ten households, all of whom are also fluent in Nafaanra. All members of these households seem to have some command of the language, but whether it will be transmitted to the next generation in more than an attenuated form is open to doubt. Nonetheless, there is no evidence that it has ever been spoken by many people and bilingualism in Nafaanra must long have been a feature of the community. There are presently some 60-70 people with some command of the language and perhaps 10 who can recall obscure lexical items. As such, intensive further study is certainly a priority.


The most recent influence on the Dompo language is clearly Nafaanra, but Dompo is clearly not related to Nafaanra. Painter (1967) assumes that Dompo belongs to North Guang, which it does on the basis of the lexicostatistic counts used in his paper. The numerous correspondences with Gonja suggest either that;

a) it is a dialect of Gonja that has come under heavy influence from other languages

b) it is a Guan language related to Gonja that has been relexified from Gonja and other languages

c) it is a language of unknown provenance that has been heavily relexified from Gonja and other languages

One argument in favour of the latter is that no names for wild animals or trees in Dompo resemble those of Gonja in any way. Some of them resemble Mo and the Senufoid languages in the area (not only Nafaanra but GyOgO and Banda Ashanti) but others seem to have no external parallels. Given the lack of detailed lexical material on many neighbouring languages this cannot be the basis for any firm conclusions. Nonetheless, it does suggest that Dompo might be a relic of the pre-Niger-Congo languages of West Africa.

A note on Mpre

Our only information about this language is in an article by Cardinall (1931) published in the Gold Coast Review. He calls it ‘A survival’ and it consists of some ethnographic notes and a wordlist of a language called Mpre, spoken in a village named Butie. Cardinall says;

‘Butie… stands more or less on the watershed of the Black and white Voltas not far from their confluence on the Southern path from Mpehe to Kabilipe.’

This is not remarkable for its precision, but it seems that area must have gone under the floodwaters when the Akosombo Dam was created. Where the populations were moved remains to be discovered.

Cardinall gives some 70 words of Mpre. Of these, the first five numerals resemble the Gonja numerals but most of the other seem to have no parallel anywhere. The list has been typed up and I am presently doing comparative work with it, but it might be a survival of a pre Niger-Congo language, or else the population was severely displaced and the language remains to be recognised from elsewhere in West Africa. No-one has yet gone in search of any remaining speakers of Mpre, but the experience of Dompo suggests that it would be worth trying.

Draft annotated wordlists of Dompo and Mpre are available from the author at r.blench(at) by email attachment in Word 97 format.


Cansdale, G.S. 1971. A list of the scientific and vernacular names of the fauna of Ghana. Accra: Legon University Press.

Cardinall, A.W. 1931. A survival. Gold Coast Review, V,1:193-197.

Painter, C. 1967. The distribution of Guang in Ghana and a statistical pre-testing on twenty-five idiolects. Journal of West African Languages, 4,1:25-78.