A Phonology of the Məkaá Language by Daniel Heath and Teresa Heath February 1982 This paper was modernized in 2012 using Unicode fonts and IPA characters. Table of Contents 0. Introduction ................................................................................................................... 2 1. Inventory of Phones ....................................................................................................... 5 2. Interpretation of Ambiguous Clusters ............................................................................. 6 2.1. Interpretation of Prenasalized Stops........................................................................ 6 2.2. Interpretation of Labial and Palatal Phones ............................................................ 8 2.3. Interpretation of Sequences of Similar Vocoids ..................................................... 11 3. Consonant Phonemes ................................................................................................... 12 3.1. Stops ..................................................................................................................... 13 3.2. Nasals ................................................................................................................... 13 3.3. Fricatives .............................................................................................................. 14 4. Vowel Phonemes .......................................................................................................... 16 5. Tone ............................................................................................................................. 39 6. Units above the Phoneme ............................................................................................ 41 6.1. Syllable ................................................................................................................. 41 6.2. Radical .................................................................................................................. 42 6.3. Phonological Word ............................................................................................... 45 7. Bibliography ................................................................................................................ 48 1 0. Introduction The Məkaá language is a member of the ‘Maka-Njem Group’ identified by Guthrie as ‘Makaa (A 83)’. It is a Bantu language spoken in the Haut Nyong department in the Eastern Province of the United Republic of Cameroon (see map 1). The people number about 70,000. Map 1: Area in Cameroon where Məkaá is spoken The Məkaá language is comprised of four main dialects, the Bəbən d, the Mbwaanz, the Mpaŋ Sekunda, and the Bəs əp (see Map 2)1. The present phonology is of the Mbwaanz dialect and more specifically of the area along Ndjonkol (Abong Mbang-Mayos) road. The Mbwaanz dialect was chosen because it is mutually intelligible by all the other dialects, it is the one spoken by the largest number of people (close to half the total population), and it is geographically the most central. For a further discussion of the dialect situation the reader may refer to a paper by the same authors entitled ‘The dialect situation of the Makaa language’. 1 This map was adapted from a map of the United Republic of Cameroon (scale 1/100.000), drawn and published by the Institut Géographique National, Centre Yaoundé (1ère edition, 1972). 2 We have carried out our research2 primarily in the village of Andjou, 25 kms west of Abong Mbang along the Ndjonkol. As far as we know the only previous linguistic information written about the Məkaá language was done by Marie-Anne Toreton, a Catholic sister, who wrote a brief grammatical sketch and a 52-page lexicon. Some anthropological work has been carried out by Peter Geschiere, an anthropologist at the Free University of Amsterdam. He has published several articles and a doctorate dissertation on the Məkaá people. Keith Beavon, a member of the Société Internationale de Linguistique, has done research in the Koonzime language, a bordering language to the south and also a member of the Maka- Njem group. He has written a phonology (1977) and several papers on various aspects of the grammar, some published and some in manuscript form. Patrick Renaud (1976) has written a phonology and a morphology of the nouns of the Bajɛle language, a member also of the same family. Bajɛle is spoken by the pygmies in southwest Cameroon around Kribi and Lolodorf and is classified by Guthrie as A 80. 2 Research was done under research authorizations Nos. 237 (1978) and 18 (1981), issued by the General Delegation for Scientific and Technical Research. To our many Məkaá friends we owe a big thanks, especially to Zé Mpouéné Saturnin, Nguélé Gregoire and Mikouas Henri who gave their time to help us learn the language and gather the necessary data. The help and motivation of the members of the ‘Association des Promoteurs des Recherches Linguistiques et de la Conservation du Patrimonie Culturel du Peuple Məkaá’ also proved an encouragement and stimulation for us. We are indebted to Dr. Ursula Wieseman for her comments and helpful criticisms on the organization of the paper. Soeur Marie-Anne Toreton also greatly helped and encouraged us with her observations on the language, both written and oral. To all these people we express our sincere thanks. 3 Map 2: Dialects of Məkaá 4 1. Inventory of Phones The Məkaá language has the consonantal and vocalic phones sho n on the follo ing charts Chart 1: Inventory of Consonantal Phones Labio- Labial Alveolar Palatal Velar velar t c k kp voiceless cw kw cj kj b, p d, t ɟ g, k voiced bw dw ɟw gw bj dj ɟj gj STOPS mp nt ɲc ŋk prenasalized mpw ɲcw ŋk voiceless mpj ɲcj ŋkj mb, mp nd, nt ɲɟ ŋg, ŋk prenasalized mbw ɲɟw ŋg voiced mbj ndj m n ɲ ŋ NASALS mw ɲw ŋ mj f s ʃ voiceless fw ʃw FRICATIVES fj ʃj v z ʒ voiced ʒw l LATERALS lw lj SEMI- w j h CONSONANTS 5 Chart 2: Inventory of Vocalic Phones FRONT CENTRAL BACK oral nasal oral nasal oral nasal close i iː ː ɘ ɘː u uː ː HIGH open ɪ HIGHER -THAN- close e eː MID MID close e eː o oː open ɛ ɛː ɛ ɛː a aː ː ɔ ɔː ɔ ɔː Sequences: ia, uo, ao, au, a , aːw, ɛw, əw, ej, ow 2. Interpretation of Ambiguous Clusters Some of the consonantal and vocalic phones are ambiguous as to their phonemic interpretation. In the following sections, we will discuss whether consonantal clusters are single phonemes or sequences of phonemes when they involve a stop preceded by a homorganic nasal. Also, we will discuss the interpretation of palatal and labial phones following consonants, following vowels, and intervocalically. Finally, we will discuss whether sequences of similar vocoids are one long or two short vowels. 2.1. Interpretation of Prenasalized Stops All stops can be preceded by a homorganic nasal, resulting in the following clusters: [nt, ɲc, ŋk, mb, nd, ɲɟ, ŋg]. [mp] also occurs, even though [p] does not occur except in [mp] and [kp]. These may be interpreted as either consonant sequences or as unit phonemes. One might interpret these as sequences for the following reasons: 6 1) The nasal-oral cluster may have both a voiced segment and a voiceless segment, such as in [mp]. 2) Both nasals and oral stops occur independently, except [p]. 3) Handling these as sequences would simplify the phoneme inventory. 4) A morpheme division can occur between the nasal and the oral segments of a sequence, as illustrated in the way some nouns form their plurals. Examples: [ɲ-ɟôŋ] ‘stranger’ [o-ɟôŋ] ‘strangers’ [m-bwô] ‘arm’ [mə-bwô] ‘arms’ (In all examples low tone is unmarked.) One of the prefixes on class 1 and class 3 nouns is a homorganic nasal. Thus, in the above words, the radicals are [ɟôŋ] and [bwô], with a homorganic nasal prefix in the singular, and the plural prefixes o- and mə- respectively. This puts a morpheme boundary between the nasal and oral segments. However, the following arguments lead us to interpret the prenasalized stops as unit phonemes. 1) Prenasalized stops have the same distribution as their oral counterparts (occurring word initially, medially, and finally). Examples: [kwan] 3 ‘bee’ [ŋk an] ‘ edding gift’ [b ] ‘break’ (vb) [mb ] ‘year’ [ɟoŋ] ‘sky’ [ɲɟôŋ] ‘stranger’ [kwɛ: k ] ‘lengthen’ (vb) [kwɛ: ŋk ] ‘purify’ (vb) [ŋkət ] ‘ ithout pity’ [ŋkən t ] ‘kind of mouse’ [kanda] ‘tale’ [nada] ‘stick’ (vb) 3 Reminder: low tones are unmarked in the phonetic transcriptions. 7 2) There are no unambiguous sequences of consonantal phones, such as [st] or [kl]. In fact, when two consonantal phones do occur together across word boundaries or in reduplicated words, a transitional [ə] is inserted, showing that consonants cannot be pronounced in sequence. 3) Interpretation of the prenasalized stops as sequences would complicate the syllable structure. 4) Due to a gap in the phoneme inventory, as has been mentioned earlier, there is no phoneme /p/ that could be combined with the phoneme /m/ to form the sequence /mp/, which does occur. Therefore must be posited as a phoneme. The phoneme /p/ does occur in one other dialect. Borrowed words containing /p/ are usually pronounced with an /f/ instead. 2.2. Interpretation of Labial and Palatal Phones Many consonantal phones can be labially or palatally released, as illustrated in charts 4 and 5. This labial or palatal release can be interpreted either as a semi-consonant or as a vowel. These releases are phonetically contoid rather than vocoid because they never carry tone contrasting with the following vocoid. Therefore they are interpreted as consonants, or rather, as semi-consonants, /w/ and /j/. This also avoids long vowel sequences. There are ambiguous sequences that could result in a VV or VVV sequence if the releases were interpreted as vowels, such as [bja] /bja/ ‘give birth’ (vb) and [kwej] /kwej/ ‘find’ (vb). Once the releases have been interpreted as semi-consonants, a second question arises. Are these semi-consonants, /w/ and /j/, and the preceding consonants sequences of phonemes, or unit phonemes? They are interpreted here as sequences of two phonemes for the following reasons: 8 1) The sequences do not have the same distribution as unit phonemes. They occur syllable–initially, but not syllable-finally, whereas the unit phonemes occur both initially and finally. Also there are gaps in the distribution of the palatal and labial sequences on the phone chart. Not all the unit phonemes have a palatal and a labial counterpart. (As has been mentioned previously, all the stops do have prenasalized counterparts.) /w/ does not occur after alveolars (except /l/, and /d/ in one example [dwo] d- o ‘cl. 5 prefix – your’); nor does / / occur after /v/, hich has a limited distribution anyway. Like /w/, /j/ does not occur after /v/ or after alveolars, except /l/ and /d/, such as [ŋk indje] ‘elbo ’. In some rare cases, /j/ does occur after palatals, such as /cj/, /ɟj/, and /ʃj/. Examples: [cj ŋ] ‘voice, neck’ [ɟjuga] ‘key, cap (of bottle)’ contrasting with [ɟuga] ‘stopper’ [ɟj m] ‘tongue’ [ʃjɛ] ‘kind of tree’ 2) There are many occurrences of Cw and Cj (more than of prenasalized stops) so that interpreting these as units would greatly increase the phoneme inventory. /w/ and /j/ already exist as separate phonemes. Another question that might be raised regarding palatal phones is whether palatal consonants are really palatalized alveolar consonants, that is, whether [c, ɲc, ɟ, ɲɟ, ɲ, ʃ, ʒ] are to be interpreted as /tj, ntj, dj, ndj, nj, sj, zj/. This is not possible for the following reasons: 1) Some dialects have /dj/ in addition to /ɟ/, and the dialect being analyzed has this contrast. Examples: [ŋk indje] ‘elbo ’ [ɲɟiɲɟamp ] ‘sorcerer’ 2) [n], [ʃ], and [ʒ] occur word finally, whereas palatally-released consonants do not. 9